Evaluation of a Landscape Analysis Approach for Migratory Birds Conservation Planning in

Evaluation of a Landscape Analysis Approach for Migratory Birds Conservation Planning in
Evaluation of a Landscape Analysis
Approach for Migratory Birds
and Species at Risk Habitat
Conservation Planning in
the Mixedwood Plains Ecozone:
Case Study in Lake Saint-Pierre
Benoît Jobin, Renée Langevin, Matthieu Allard,
Sandra Labrecque, Diane Dauphin,
Martine Benoit, Pierre Aquin
Quebec Region
Canadian Wildlife Service
March 2013
CANADIAN WILDLIFE SERVICE TECHNICAL REPORT SERIES
This series of reports, introduced in 1986, contains technical and scientific information on Canadian Wildlife Service
projects. The reports are intended to make available material that is either of interest to a limited audience or is too
extensive to be accommodated in scientific journals or in existing CWS series.
Demand for the Technical Reports is usually limited to specialists in the fields concerned. Consequently, they are
produced regionally and in small quantities. They are numbered according to a national system but can be obtained
only from the address given on the back of the title page. The recommended citation appears on the title page.
Technical Reports are available in CWS libraries and are listed in the catalogue of Library and Archives Canada,
which is available in science libraries across the country. They are printed in the official language chosen by the
author to meet the language preference of the likely audience, with an abstract in the second official language. To
determine whether there is sufficient demand to make the Reports available in the second official language,
CWS invites users to specify their official language preference. Requests for Technical Reports in the second
official language should be sent to the address on the back of the title page.
SÉRIE DE RAPPORTS TECHNIQUES DU SERVICE CANADIEN DE LA FAUNE
Cette série de rapports, créée en 1986, donne des informations scientifiques et techniques sur les projets du Service
canadien de la faune (SCF). Elle vise à diffuser des études qui s'adressent à un public restreint ou sont trop
volumineuses pour paraître dans une revue scientifique ou une autre série du SCF.
Ces rapports techniques ne sont habituellement demandés que par les spécialistes des sujets traités. C’est pourquoi
ils sont produits à l'échelle régionale et en quantités limitées. Ils sont toutefois numérotés à l'échelle nationale. On ne
peut les obtenir qu'à l'adresse indiquée au dos de la page titre. La référence recommandée figure à la page titre.
Les rapports techniques sont conservés dans les bibliothèques du SCF et figurent dans le catalogue de Bibliothèque
et Archives Canada, que l’on retrouve dans les principales bibliothèques scientifiques du Canada. Ils sont publiés
dans la langue officielle choisie par l'auteur, en fonction du public visé, accompagnés d’un résumé dans la deuxième
langue officielle. En vue de déterminer si la demande est suffisante pour publier ces rapports dans la deuxième
langue officielle, le SCF invite les usagers à lui indiquer leur langue officielle préférée. Les demandes de
rapports techniques dans la deuxième langue officielle doivent être envoyées à l'adresse indiquée au dos
de la page titre.
EVALUATION OF A LANDSCAPE ANALYSIS APPROACH
FOR MIGRATORY BIRDS AND SPECIES
AT RISK HABITAT CONSERVATION PLANNING
IN THE MIXEDWOOD PLAINS ECOZONE:
CASE STUDY IN LAKE SAINT-PIERRE
Benoît Jobin
Renée Langevin
Matthieu Allard
Sandra Labrecque
Diane Dauphin
Martine Benoit
Pierre Aquin
March 2013
Canadian Wildlife Service
Quebec Region
This report may be cited as follows:
Jobin B., R. Langevin, M. Allard, S. Labrecque, D. Dauphin, M. Benoit and P. Aquin. 2013.
Evaluation of a Landscape Analysis Approach for Migratory Birds and Species at Risk Habitat
Conservation Planning in the Mixedwood Plains Ecozone: Case Study in Lake Saint-Pierre.
Technical report series No. 527. Environment Canada, Canadian Wildlife Service, Quebec Region,
Quebec. 69 p. and appendices.
A copy of this report can be obtained by contacting:
Environment Canada
Canadian Wildlife Service
Quebec Region
801-1550 D’Estimauville Avenue
Québec (Québec) G1J 0C3
ISBN 978-1-100-22361-2
Cat. No.: CW69-5/527E-PDF
Information contained in this publication or product may be reproduced, in part or in whole, and by
any means, for personal or public non-commercial purposes, without charge or further permission,
unless otherwise specified.
You are asked to:
• Exercise due diligence in ensuring the accuracy of the materials reproduced;
• Indicate both the complete title of the materials reproduced, as well as the author organization;
and
• Indicate that the reproduction is a copy of an official work that is published by the Government of
Canada and that the reproduction has not been produced in affiliation with or with the
endorsement of the Government of Canada.
Commercial reproduction and distribution is prohibited except with written permission from the
Government of Canada's copyright administrator, Public Works and Government Services of Canada
(PWGSC). For more information, please contact PWGSC at 613-996-6886 or at
[email protected]
Photos: © Photos.com, Thinkstockphotos.ca
© Her Majesty the Queen in Right of Canada, represented by the Minister of Environment, 2013
Aussi disponible en français
ACKNOWLEDGEMENTS
We would like to thank Daniel Robitaille and Luc Bélanger of the Canadian Wildlife Service,
Quebec Region (CWS-QC) for their support throughout the project. We thank Daniel Bordage,
Vincent Carignan, Bruno Drolet, Gilles Falardeau, Christine Lepage, Jean-François Rail,
François Shaffer and Josée Tardif, also from the CWS-QC, for data sharing and for their
comments on the habitat criteria for priority species. André Desrochers and Monique Poulin
from Université Laval were also consulted for their expertise on peat bogs, and we thank them
for their input. We would also like to thank the following people for the information they shared:
Martin Jean of Environment Canada (Water Quality Monitoring and Surveillance Division);
Pascale Dombrowski, Marc Mingelbier, Réjean Dumas and Charles Racine of the Ministère des
Ressources naturelles du Québec (MRN); Jacques Perron of the Ministère du Développement
durable, de l’Environnement, de la Faune et des Parcs (MDDEFP), Jason Beaulieu of Ducks
Unlimited Canada (DUC); Pascal Hébert of Nature Conservancy of Canada (NCC); Christine
Bélanger of the Fondation de la faune du Québec (FFQ); Ludyvine Millien of the Réseau de
milieux naturels protégés (RMN); and Roxanne Dugas of the Lake Saint-Pierre Biosphere
Reserve (LSPBR).
Project Team
This project was conducted by the Landscape Assessment and Planning Unit of the Canadian
Wildlife Service, Quebec Region, as part of a pilot project designed to develop a methodology
for habitat conservation planning.
Benoît Jobin
Renée Langevin
Matthieu Allard
Sandra Labrecque
Diane Dauphin
Martine Benoit
Pierre Aquin
i
ABSTRACT
Environment Canada’s Canadian Wildlife Service (CWS), Quebec Region, has conducted a pilot
project to develop and test a method to identify priority sites for migratory bird conservation
within Bird Conservation Region (BCR) 13, located in the Mixedwood Plains ecozone. The
approach is based on the landscape ecology theory, making it possible to associate habitat needs
of priority bird species with a finer description of habitat composition and spatial distribution.
This landscape-based approach is more integrative and allows for work on a broader scale
instead of the more conservative approach based on known priority sites (hot spots) traditionally
used in conservation planning. A logic model illustrating the steps for preparing a conservation
plan was developed and tested. The Lake Saint-Pierre region (included within BCR 13) was
selected as the study area.
The goal of the project was to determine the current and potential structure of the landscape
in order to maintain and restore functional and viable habitats for the priority species for
this project. A total of 48 species identified in the BCR 13 conservation plan as species of
conservation concern were selected, including species at risk for which critical habitat was
proposed or designated. The issue regarding those priority species is for the most part associated
with breeding habitat availability. A land cover map of the study region has been produced and
validated. A total of 7 general (anthropogenic, shrubland, annual crops, perennial crops, water,
forest, wetland) and 21 detailed land cover classes were delineated. Data on the protected areas
and species at risk present in the study area were also compiled.
The analysis performed with ArcGIS and FRAGSTATS software was divided into two major
components: 1, descriptive analysis; and 2, landscape functionality. The analysis was performed
at the study area level, at the regional county municipality (RCM) scale and at the watershed
scale. The last two spatial units were selected because they foster effective implementation of the
conservation recommendations whereby priority sites can be considered in regional land-use
planning activities. The study area is largely dominated by agriculture: annual and perennial
crops cover 31% and 20% of the study area respectively, followed by forest (24%), open water
(10%), wetlands (10%), anthropogenic areas (4%) and shrubland (which accounts for only 1% of
the area). A detailed analysis also identified portions of the study area most suitable for forest
birds, where forest fragmentation is reduced and where forest interior habitats still prevail.
Landscape functionality was analyzed by comparing the composition of the landscape with
known habitat thresholds, by identifying movement corridors for forest birds, and by assessing
the availability of certain classes of priority habitat. The thresholds used to compare the
landscape of the study area, RCM and watersheds were taken from the document How Much
Habitat is Enough? and focused on forest habitat, wetlands and riparian buffer strips. Forest
habitats are under-represented in the study area, though forest interior habitats could probably
support forest bird populations. Wetlands are abundant around Lake Saint-Pierre, but their
presence is limited in the rest of the area. Furthermore, the integrity of these habitats is at risk
because the adjacent habitats are strongly influenced by human activity. The same is true for
riparian buffer strips.
iii
The identification of movement corridors for forest birds focused on connecting forest patches
> 1000 ha. Using Corridor Designer software, 14 movement corridors were selected based on
pre-established criteria (width > 300 m; distance between woodlots < 200 m). Priority breeding
habitats were then identified using the coarse- and fine-filter approaches. Hundreds of habitat
patches occupying the minimum surface area necessary to meet the needs of multiple priority
species (coarse-filter approach) were identified throughout the study area and in various types of
environments (forest > 100 ha, perennial crops > 40 ha, shrubland > 5 ha, marsh > 5 ha, shrub
swamp > 5 ha, peatland > 20 ha). All patches of forest swamp and wet meadow were considered
priority sites because no minimum area threshold is known for those habitat classes. The best
patches in each habitat class were then prioritized according to a series of criteria related to their
significance for the establishment and maintenance of nesting bird populations (e.g. patch shape,
% of interior habitat) or their ecological role in the landscape (e.g. creation of a buffer zone
around protected areas, presence of species at risk). Finally, other habitat components specific to
certain species (fine-filter approach) such as sand pits and rocky outcrops in forest environments
were also identified.
A diagnosis of the ability of the landscape to provide functional habitat for priority bird species
was performed. The deficiencies noted in the study area included the limited surface area
occupied by shrubland (1%), the lack of forest cover (< 30%), the inadequate distribution
of wetlands (few are located outside the immediate vicinity of Lake Saint-Pierre), severe
disturbance of riparian buffer strips, and forest corridors that do not meet the established criteria.
A conservation plan for migratory bird and species at risk habitat was developed taking into
account the description and analysis of landscape functionality, as well as regional development
issues. The conservation plan proposes detailed conservation actions at the RCM and watershed
scales: prioritization of spatially explicit habitat patches (habitat of avian species at risk, coarseand fine-filter patches, forest corridors), protection of non-spatially explicit habitat components
(e.g. large-diameter snags, Purple Martin nest boxes), and landscape attributes to be considered
for the maintenance of ecological processes (e.g. vegetated riparian buffer strips). The
conservation plan must be validated at the site level because the data on certain habitats may
be outdated. Possible actions and proposals for the implementation of the conservation plan
are also presented. Lastly, a general summary of the project points out the benefits and some
shortcomings of the landscape-based approach that was used and highlights some problems
encountered. A suite of recommendations are proposed to help with the application of the
approach and to support a joint involvement of partners and stakeholders in land use planning.
iv
RÉSUMÉ
Le Service canadien de la faune (SCF) d’Environnement Canada, région du Québec, a réalisé un
projet pilote afin de développer et de tester une méthodologie permettant de déterminer les sites
prioritaires pour la conservation des oiseaux migrateurs à l’échelle de la région de conservation
des oiseaux (RCO) 13 située dans l’écozone des Plaines à forêts mixtes. L’approche utilisée est
basée sur l’écologie du paysage et permet de coupler les besoins réels en matière d’habitat des
espèces prioritaires à une analyse fine de la composition et de la répartition spatiale des habitats.
Ceci permet donc d’avoir une vision plus intégratrice et à plus grande échelle du territoire au lieu
de cibler la protection de sites déjà connus comme importants pour les oiseaux (approche par
« hot spot » traditionnellement utilisée en conservation). Un modèle logique, qui illustre les
étapes nécessaires à la réalisation d’un plan de conservation selon cette approche paysage, a été
développé et testé. La région du lac Saint-Pierre (incluse dans la RCO 13) a été choisie comme
aire d’étude.
Le but du projet était de déterminer la structure actuelle et potentielle du paysage dans le but de
maintenir et de rétablir des habitats fonctionnels et viables pour les espèces considérées comme
prioritaires pour ce projet. Au total, 48 espèces présentant des enjeux de conservation identifiées
dans le plan de conservation de la RCO 13 ont été retenues, incluant les espèces en péril pour
lesquelles des habitats essentiels sont proposés ou désignés. Ces enjeux sont presque tous
associés à la disponibilité des habitats de nidification. Une carte d’occupation du sol a été
réalisée et validée. Au total, sept classes générales (anthropique, arbustif, culture annuelle,
culture pérenne, eau, forestier, milieu humide) et 21 classes détaillées d’occupation du sol ont
été retenues. Les données sur les aires protégées et sur les espèces en péril présentes dans l’aire
d’étude ont aussi été compilées et utilisées.
L’analyse réalisée à l’aide des logiciels ArcGIS et FRAGSTATS se divise en deux grands
volets : 1- l’analyse descriptive et 2- la fonctionnalité du paysage. Cette analyse a été réalisée à
l’échelle de l’aire d’étude, ainsi qu’à celle des MRC et des bassins versants. Ces deux derniers
découpages spatiaux ont été retenus car ils facilitent la mise en œuvre des recommandations de
conservation en les intégrant au processus usuel de planification du territoire au Québec. L’aire
d’étude est largement dominée par l’agriculture : les cultures annuelles et pérennes couvrent
respectivement 31 % et 20 % du territoire. Suivent les milieux forestiers (24 %), les zones d’eau
libre (10 %), les milieux humides (10 %), les milieux anthropiques (4 %) et les milieux arbustifs,
qui ne couvrent que 1 % du territoire. Une analyse détaillée a aussi permis de localiser les
secteurs de l’aire d’étude où la fragmentation forestière est réduite et où se situent les habitats
forestiers d’intérieur.
La fonctionnalité du paysage a été analysée en comparant la composition du paysage à des seuils
de référence connus, en déterminant des corridors de déplacement potentiels pour les oiseaux
forestiers et en évaluant la disponibilité de certaines classes d’habitats prioritaires. Les seuils de
référence utilisés ont été extraits du document « Quand l’habitat est-il suffisant » et ciblaient les
habitats forestiers, les milieux humides et les bandes riveraines. Les habitats forestiers sont
sous-représentés dans l’aire d’étude, bien que les habitats forestiers d’intérieur qui s’y trouvent
permettent de soutenir des populations d’oiseaux forestiers. Les milieux humides sont abondants
v
autour du lac Saint-Pierre, mais leur présence est limitée ailleurs sur le territoire. De plus,
l’intégrité de ces habitats est menacée puisque les milieux adjacents sont fortement anthropisés.
On observe la même situation pour les bandes riveraines.
Des corridors de déplacement pour les oiseaux forestiers visant à relier les massifs forestiers
> 1000 ha ont été déterminés à l’aide du logiciel Corridor Designer. Quatorze corridors ont été
retenus en fonction de critères préétablis (largeur > 300 m, distance entre les boisés < 200 m).
Enfin, les habitats prioritaires de nidification ont été déterminés en appliquant les principes de
filtre grossier et de filtre fin. Des centaines de parcelles d’habitat occupant des superficies
minimales requises pour combler les besoins des espèces prioritaires (filtre grossier) ont ainsi été
sélectionnées dans toute l’aire d’étude et pour différentes classes de milieux (forêt > 100 ha,
culture pérenne > 40 ha, milieu arbustif > 5 ha, marais > 5 ha, marécage arbustif > 5 ha,
tourbière > 20 ha). Toutes les parcelles de marécage arboré et de prairie humide ont été
considérées comme prioritaires puisqu’aucun seuil de superficie minimale n’est connu pour ces
classes d’habitat. Une priorisation des meilleures parcelles de chacune des classes d’habitat a
ensuite été faite à l’aide d’une série de critères portant sur leur importance pour l’établissement
et le maintien de populations d’oiseaux nicheurs (ex., forme des parcelles; % d’habitat
d’intérieur) ou sur leur rôle écologique dans le paysage (ex., mise en place de zone tampon
autour des aires protégées; présence d’espèces en péril). Finalement, d’autres composantes
d’habitats recherchées par certaines espèces (filtre fin) ont été localisées sur le territoire
d’étude comme des sablières et des sols dénudés en milieu forestier.
Un diagnostic sur la capacité du paysage à procurer des habitats fonctionnels aux espèces
d’oiseaux prioritaires a ensuite été réalisé. Parmi les lacunes relevées, on note la faible superficie
occupée par les friches arbustives (1 %), le manque de couverture forestière (inférieur au seuil de
30 % établi), la répartition inadéquate des milieux humides (peu présents en dehors de la région
immédiate du lac Saint-Pierre), une forte perturbation des bandes riveraines adjacentes aux cours
d’eau, ainsi que des corridors forestiers qui répondent peu aux critères de sélection. Un plan de
conservation des habitats de nidification des oiseaux migrateurs et des espèces en péril a été
développé qui tient compte de la description et de l’analyse de la fonctionnalité du paysage, de
même que des enjeux de développement régional. Ce plan de conservation propose des actions
de conservation qui sont détaillées à l’échelle des MRC et des bassins versants : la priorisation
de parcelles d’habitats avec référence spatiale (habitats d’espèces d’oiseaux en péril, parcelles
du filtre grossier et du filtre fin, corridors forestiers), la protection de composantes d’habitat sans
référence spatiale (ex., chicots de grand diamètre, nichoirs à Hirondelle noire) et les éléments
du paysage à considérer pour le maintien de processus écologiques (ex., bandes riveraines
végétées). Une validation est toutefois nécessaire car les données relatives à certains habitats
peuvent dater de plusieurs années. Des pistes et des propositions pour la mise en œuvre du plan
de conservation sont aussi présentées. Enfin, un bilan général du projet soulève les différents
avantages et certains inconvénients de l’approche paysage retenue et met en lumière certaines
problématiques rencontrées. Diverses recommandations sont proposées permettant d’appliquer
la méthode d’analyse utilisée afin de favoriser l’arrimage des outils existants et la concertation
des intervenants impliqués dans l’aménagement du territoire.
vi
TABLE OF CONTENTS
Acknowledgements .......................................................................................................................... i
Abstract .......................................................................................................................................... iii
Résumé .............................................................................................................................................v
List of Acronyms and Abbreviations ............................................................................................. ix
1.0
Introduction ..........................................................................................................................1
2.0
Concepts of Landscape Ecology ..........................................................................................1
2.1
2.2
2.3
2.4
2.5
Definition ............................................................................................................................ 1
Structure of the Landscape ................................................................................................. 2
Landscape Fragmentation ................................................................................................... 2
Buffer Zone ......................................................................................................................... 2
Landscape Study ................................................................................................................. 2
3.0
Development of a Logic Model ...........................................................................................3
4.0
Delineation of the Study Area for the Pilot Project .............................................................4
5.0
Step 1 − Goal and Objectives of the Pilot Project ...............................................................6
6.0
Step 1a − Identifying Targets: Priority Species and Habitats ..............................................6
6.1 Selection of Priority Species for the Pilot Project .............................................................. 6
6.2 Selection of Priority Habitats for the Pilot Project ............................................................. 8
7.0
7.1
7.2
7.3
7.4
8.0
Step 2 – Data collection and management .........................................................................10
Data Sources ..................................................................................................................... 10
Production of the Final Land Use Map ............................................................................. 11
Data on Protected Areas ................................................................................................... 11
Data on Species at Risk and Critical Habitats .................................................................. 12
Step 2a – Planning and Mapping Tools .............................................................................14
8.1 Reviewing Existing Softwares .......................................................................................... 14
8.2 Decisions Made Prior to Analysis .................................................................................... 15
8.2.1
8.2.2
8.2.3
8.2.4
Spatial analysis scale .............................................................................................. 15
Overlapping of habitat patches ............................................................................... 15
Selection of landscape metrics ................................................................................ 15
Forest fragmentation ............................................................................................... 16
vii
9.0
Step 3 – Landscape Analysis .............................................................................................17
9.1
Descriptive Analysis ...................................................................................................... 17
9.1.1
9.1.2
Land use: Study area ............................................................................................... 17
Land use: RCM ....................................................................................................... 20
9.2 Functionality of the Landscape ......................................................................................... 24
9.2.1
9.2.2
9.2.3
9.2.4
9.2.5
Comparison of the landscape with known reference thresholds............................. 24
Identification of forest corridors ............................................................................. 30
Application of coarse filter criteria ......................................................................... 36
Prioritization of coarse filter patches ...................................................................... 39
Application of fine filter criteria ............................................................................. 43
Step 4 − Final Analysis and Specific Issues ......................................................................43
10.0
10.1 Analysis of Current Situation ......................................................................................... 43
10.2 Regional Issues and Threats ........................................................................................... 45
11.0
Step 8 – Conservation Plan for the Study Area .................................................................46
11.1
11.2
11.3
11.4
11.5
Priority Habitat Patches with Spatial Reference ............................................................ 46
Priority Habitats Without Spatial Reference .................................................................. 47
Landscape Elements to Consider for Maintaining Ecological Processes ...................... 49
Special Considerations of the Conservation Plan .......................................................... 50
Conservation Plan Specific to Each RCM and Watershed ............................................ 51
11.5.1 Example of a detailed conservation plan: Bécancour RCM ................................... 52
11.6 Limits of the Conservation Plan ..................................................................................... 54
12.0
Step 9 − Implementation of the Conservation Plan: Approaches and Proposals...............55
13.0
Additional Information ......................................................................................................55
13.1 Overall Assessment of the Project ................................................................................. 55
13.1.1
13.1.2
13.1.3
13.1.4
Advantages of the landscape approach ................................................................... 55
Disadvantages of the landscape approach .............................................................. 56
Involvement of other CWS units, partners, etc. ...................................................... 57
Steps to complete .................................................................................................... 57
13.2 Problems Encountered and Recommendations .............................................................. 57
13.2.1 Data ......................................................................................................................... 57
13.2.2 Softwares................................................................................................................. 58
13.2.3 Teamwork ............................................................................................................... 59
14.0
Conclusion .........................................................................................................................59
15.0
Literature Cited ..................................................................................................................60
viii
LIST OF ACRONYMS AND ABBREVIATIONS
AAFC: Agriculture and Agri-Food Canada
AARQ: Atlas of Amphibians and Reptiles of Quebec
BCR: Bird Conservation Region
CDPNQ: Centre de données sur le patrimoine naturel du Québec
COSEWIC: Committee on the Status of Endangered Wildlife in Canada
CWI: Canadian Wetland Inventory
CWS: Canadian Wildlife Service
CWS-QC: Canadian Wildlife Service, Quebec Region
DUC: Ducks Unlimited Canada
EC: Environment Canada
EHJV: Eastern Habitat Joint Venture
FFQ: Fondation de la faune du Québec
GIS: Geographic information system
HSP: Habitat Stewardship Program
HPP : Habitat Priority Planner
LSPBR: Lake Saint-Pierre Biosphere Reserve
MBS: Migratory bird sanctuary
MDDEFP: Ministère du Développement durable, de l’Environnement, de la Faune et des Parcs
MRN: Ministère des Ressources naturelles
NABCI: North American Bird Conservation Initiative
NAESI: National Agri-Environmental Standards Initiative
NCC: Nature Conservancy of Canada
NGO: Non-governmental organization
NTDB: National Topographic Data Base
RCM: Regional county municipality
RMN: Réseau de milieux naturels protégés
ROW: Right-of-Way
SARA: Species at Risk Act
SIEF: Ecoforestry information system of Quebec
SLAP: St. Lawrence Action Plan
SOS-POP: Suivi de l’occupation des stations de nidification des populations d’oiseaux en péril
du Québec
ix
1.0
INTRODUCTION
In 2005, Canada, the United States and Mexico signed the NABCI (North American Bird
Conservation Initiative) Declaration of Intent in an effort to strengthen international cooperation
in the field of bird conservation in North America (NABCI International 2012). Bird
conservation strategies are developed for Bird Conservation Regions (BCRs) and contain
population goals, habitat needs and conservation issues for a number of priority species within a
BCR. In this context, the Canadian Wildlife Service (CWS) – Quebec Region has developed and
evaluated a methodology to identify priority sites for the conservation of migratory birds. The
approach used under this pilot project is based on the landscape ecology theory, which allows for
associating actual habitat needs of priority species with a detailed analysis of the composition
and spatial distribution of habitats. This conservation planning process is based on an integrated
assessment of the landscape instead of the more traditional approach used in conservation
planning where the identification of priority sites for bird conservation is based on past surveys
(“hot spot” approach). The landscape approach not only offers the advantage of working on a
larger scale, it also allows to integrate in the analysis various components of the landscape
(biological, geographical, physical, socio-economic and heritage-related) that characterize the
study area (habitats, species at risk, protected areas, hydrology, climate, human activities, etc.).
This report presents the method used, its advantages and disadvantages, and the resulting
recommendations that constitute the conservation plan. It aims to provide information to anyone
interested in learning more about the application of a landscape-based approach to determine
priority sites for conservation and will be useful to land managers acting at various levels
(government, municipalities, non-governmental organizations). The project was carried out in
the Lake Saint-Pierre region, a portion of the BCR 13 located in the Mixedwood Plains ecozone,
and recognized for its high level of biodiversity and its significance for migratory birds, but also
for the intensive anthropogenic pressure affecting natural habitats. This report summarizes the
main points and results of the pilot project. The reader is invited to consult a related report (Jobin
et al. 2013) for a detailed description of the methodological development, analyses, and results.
A few concepts underlying the landscape-based approach are described in the following section.
2.0
CONCEPTS OF LANDSCAPE ECOLOGY
2.1
DEFINITION
Landscape ecology focuses on the spatial and temporal dynamics of biological, physical and
social landscapes (humanized and/or natural) where humans are also a component of the
landscape (Turner et al. 2001). Through their activities, humans can affect the structure and
integrity of the landscape and interfere with ecological processes. In the context of this project,
landscape ecology is defined as the study of the interaction between landscape structure (its
composition and configuration) and the processes that determine the abundance and distribution
of species.
1
2.2
STRUCTURE OF THE LANDSCAPE
A landscape is a heterogeneous and dynamic mosaic composed of three main components: the
matrix, the patch (habitats) and the corridors (Forman 1995). A landscape is thus a mosaic of
important habitats (patches) for a species or species group that are spread across a dominant
component of the landscape (the matrix). Corridors are the elements that connect patches.
Habitat composition of the various elements and their configuration in the landscape, in other
words, their juxtaposition relative to each other, is what characterizes the structure of the
landscape. A landscape approach therefore requires consideration of the ecological requirements
of the species (habitats), but also their method and limits for travel between habitat patches.
2.3
LANDSCAPE FRAGMENTATION
The natural environments of southern Quebec have been fragmented by human activities (roads,
agriculture, etc.) (Bélanger and Grenier 1998, 2002; Latendresse et al. 2008). This fragmentation
of the landscape, i.e. the replacement of landscape elements by others and the decrease in patch
size, has significant consequences for the biodiversity of an area (Saunders et al. 1991; Andrén
1994; Fahrig 2003; Stephens et al. 2003), due to patch isolation and edge effects. With habitat
patches becoming smaller and more isolated from each other, the resulting "insularization" has
implications for the dispersal of individuals and genetics exchange. For several species, corridors
help compensate for landscape fragmentation.
2.4
BUFFER ZONE
A buffer zone is a strip of more or less natural vegetation that reduces the contrast between a
given habitat and adjacent habitats, thereby maintaining or improving ecological integrity
(Bentrup 2008). Depending on the context and objectives set, the buffer zone can be contiguous
to specific sites such as protected areas. It may also extend along watercourses and reduce
runoff, siltation and nonpoint source pollution. These are called riparian buffer strips. The
establishment of buffer zones adjacent to protected areas helps reduce anthropogenic pressures
and edge effects that can affect many species.
2.5
LANDSCAPE STUDY
The development of new technologies such as remote sensing, as well as specialized GIS
(geographic information systems) softwares facilitates the study of landscapes and their
evolution. In addition, a variety of landscape metrics have been developed to describe the
composition, structure and spatial configuration of habitat patches and landscapes (McGarigal et
al. 2002). The use of landscape ecology concepts, along with the use of these analytical tools in
studies aimed at understanding the distribution and abundance of living organisms and their
habitats, is now common practice (Huber et al. 2011; Thompson 2011; Watling et al. 2011),
and their use in the study of bird communities is widespread (Naugle et al. 2000; Renfrew
and Ribic 2008; Holzmueller et al. 2011; Schwenk and Donovan 2011; Shanahan et al. 2011;
Uezu and Metzger 2011).
Moreover, the spatial scale under analysis varies depending on the issue or study species and
should consider the needs of individuals (e.g. the landscape extent of a salamander is very
different from that of a black bear), such as nesting or feeding needs, and needs for genetic
2
exchange and dispersal within a population (metapopulation concept). In the case of bird
populations, the spatial scale under analysis can cover an area that extends over hundreds
of square kilometres, such as an administrative region or an ecoregion.
3.0
DEVELOPMENT OF A LOGIC MODEL
We developed a logic model designed to organize and visually present the steps necessary to
achieve a conservation plan, regardless of the spatial scale (ecozone, particular ecosystem, BCR,
etc.) (Figure 1). Used as part of this project, the model is divided into several steps and presents
the logical links to the available resources to ensure the successful completion of the project
(such as partners, inventories), as well as factors that may influence its outcome (such as laws
and regulations).
The goal and general objectives of the project initially depend on the issues and priorities that
are specific to the landscape under study, ideally in collaboration with the project partners. This
is followed by a series of steps ranging from data collection and the selection of appropriate tools
(e.g. software), to a detailed analysis of the landscape leading to a diagnosis of the state of
ecosystems and current issues. It may be useful to model changes or disruptions to the landscape
in order to define the best strategies for landscape planning, taking into account future
anthropogenic developments. The conservation plan integrates the results of the previous steps
and identifies priority areas for conservation. Naturally, implementation of the actions identified
in the conservation plan entails a variety of conservation and habitat restoration options and rests
on the involvement of local partners and stakeholders (municipalities, RCMs, etc.) in the context
of policies, laws and regulations that may have an impact on the land. Finally, actions taken
should be monitored in order to determine if the results meet the objectives. Finally, this model
is adaptative as new information that becomes available can improve its outcome.
Since this was a pilot project with a set deadline, some of the steps in the logic model were
intentionally omitted (steps 4a, 5, 6, 7, 9 and 10, Figure 1). Similarly, the collection of data was
limited to biological, physical and geographical data. Some crucial links in the logic model were
also not established during this project. For example, stakeholders and partners working in the
study area were not invited to participate in the project since the exercise was to develop and test
a methodology. However, the implementation of this pilot project helped test the logic and
usefulness of the model.
3
Figure 1 − Diagram of the logic model 1 (the steps completed are in yellow)
4.0
DELINEATION OF THE STUDY AREA FOR THE PILOT
PROJECT
The Mixedwood Plains ecozone is one of the priority ecosystems for habitat conservation in Canada.
This ecozone ranges from the Great Lakes in Ontario to Québec (Quebec City) and includes the
St. Lawrence River fluvial system where habitat conservation and restoration are major objectives of
the St. Lawrence Action Plan for over 20 years (www.planstlaurent.qc.ca/en/home.html). The Lake
Saint-Pierre, which is a widening of the St. Lawrence River upstream of Trois-Rivières, and its
floodplain, form an area that sustain a very rich biodiversity, which led to its designation as a
RAMSAR site (www.ramsar.org) and as a Biosphere Reserve (www.biospherelac-st-pierre.qc.ca).
Human activities (agriculture, navigation, urban and industrial developments) create significant
pressures on the natural environment, and conservation actions are needed to protect and restore
1
Adapted from Leitão and Ahern 2002, and Ahern 2006.
4
key habitats. Other reasons support our choice of this region as the study area for the pilot
project:
•
•
•
•
•
•
This sector is included in BCR 13;
This sector is home to a wide diversity of species and habitats and is important for the
migration, feeding and breeding of numerous migratory birds and species at risk;
Environment Canada holds several territories in the region; there is a migratory bird
sanctuary (Nicolet MBS), numerous public and private protected areas, and two
Aboriginal reserves;
Several projects of the Habitat Stewardship Program (HSPs) are underway;
The study area is located within the range of the Eastern Habitat Joint Venture (EHJV)
and the St. Lawrence Action Plan (SLAP);
A large amount of biological, physical, geographical and hydrological data are available
for this region.
The study area for the project was established using the boundaries of the Lake Saint-Pierre
Biosphere Reserve, while retaining municipalities with more than half of their area contained in
BCR 13 (Figure 2). Some municipalities were excluded in the western part in order to limit the
study area at the head of Lake Saint-Pierre, but the industrial centre of Sorel-Tracy and all the
islands of the Berthier-Sorel archipelago have been preserved. The study area covers an area of
4194 km2.
Figure 2 − Delineation of the study area for the pilot project
5
5.0
STEP 1 − GOAL AND OBJECTIVES OF THE PILOT PROJECT
The goal of the pilot project is to "determine the current and potential structure 2 of the landscape
in order to maintain and restore functional and viable habitats of species considered priority for
this project." More specifically, the objectives are to:
•
•
•
•
•
6.0
determine priority species in the study area;
determine the types of habitat that are essential to the life cycle of these priority species;
evaluate the functionality of the landscape (compare with reference values, determine
travel corridors for forest birds, locate potential habitats for priority species);
identify regional issues and threats to habitats;
produce a landscape conservation plan.
STEP 1A − IDENTIFYING TARGETS: PRIORITY SPECIES
AND HABITATS
The first two objectives of the project are to determine priority species in the study area and to
determine the habitats necessary for their life cycle. The following sections describe the steps
needed to meet these objectives.
6.1
SELECTION OF PRIORITY SPECIES FOR THE PILOT PROJECT
The priority species in the context of the pilot project are:
•
•
Priority species for conservation purpose identified in the BCR 13 bird conservation plan;
Species at risk listed under Schedule 1 of the Species at Risk Act (SARA) for which
critical habitats are identified.
In Quebec's integrated BCR 13 plan (November 2010 version, Fournier et al. 2010), 67 species
of birds are considered a priority. However, many of these species are not a priority in the
current project because some are very rare or do not inhabit the study area, or other species do
not require immediate conservation action (priority species for reasons of stewardship or
overabundant species). In collaboration with experts from the CWS, 48 priority species were
retained (33 species of landbirds, 4 species of waterfowl, 6 species of marshbirds/waterbirds,
5 species of shorebirds), 9 of which were selected according to precautionary principle (Table 1).
Conservation issues for priority species largely deal with breeding habitats. Appendix 1 lists the
19 species that were not selected. The study area includes identified critical habitats for only
1 species at risk designated under SARA, the Least Bittern, which is also a priority species in the
BCR 13.
2
The structure is defined as follows: configuration and composition of spatio-temporal components of the landscape.
6
Table 1 − List of 48 priority bird species found in the BCR 13 and selected for the pilot project
English name1
Latin name
Group2
American Kestrel
Baltimore Oriole
Bank Swallow
Barn Swallow
Barred Owl
Belted Kingfisher
Black-billed Cuckoo
Bobolink*
Brown Creeper
Brown Thrasher
Canada Warbler*
Chimney Swift*
Common Nighthawk*
Eastern Kingbird
Eastern Meadowlark
Eastern Screech-Owl
Eastern Wood-Pewee
Horned Lark
Long-eared Owl
Nelson's Sparrow
Northern Flicker
Northern Harrier
Northern Rough-winged Swallow
Northern Saw-whet Owl
Palm Warbler
Peregrine Falcon (anatum)*
Purple Martin
Savannah Sparrow
Sedge Wren
Short-eared Owl*
Vesper Sparrow
Whip-poor-will*
Wood Thrush
American Woodcock
Killdeer
Upland Sandpiper
Wilson's Phalarope
Wilson's Snipe
American Bittern
Falco sparverius
Icterus galbula
Riparia riparia
Hirundo rustica
Strix varia
Megaceryle alcyon
Coccyzus erythropthalmus
Dolichonyx oryzivorus
Certhia americana
Toxostoma rufum
Wilsonia canadensis
Chaetura pelagica
Chordeiles minor
Tyrannus tyrannus
Sturnella magna
Megascops asio
Contopus virens
Eremophila alpestris
Asio otus
Ammodramus nelsoni
Colaptes auratus
Circus cyaneus
Stelgidopteryx serripennis
Aegolius acadicus
Dendroica palmarum
Falco peregrinus anatum
Progne subis
Passerculus sandwichensis
Cistothorus platensis
Asio flammeus
Pooecetes gramineus
Caprimulgus vociferus
Hylocichla mustelina
Scolopax minor
Charadrius vociferus
Bartramia longicauda
Steganopus tricolor
Gallinago delicata
Botaurus lentiginosus
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Land.
Shor.
Shor.
Shor.
Shor.
Shor.
Mar.
7
English name1
Latin name
Group2
Black Tern
Common Tern
Least Bittern*
Sora
Virginia Rail
Blue-winged Teal
Greater Scaup
Lesser Scaup
Wood Duck
Chlidonias niger
Sterna hirundo
Ixobrychus exilis
Porzana carolina
Rallus limicola
Anas discors
Aythya marila
Aythya affinis
Aix sponsa
Mar.
Mar.
Mar.
Mar.
Mar.
Wat.
Wat.
Wat.
Wat.
1
The species followed by an asterisk are listed as species at risk in SARA or by COSEWIC. Species in italics are selected according to
precautionary principle (no data on population trends, but actual or perceived issues and threats are identified).
2
Land. = Landbirds; Shor. = Shorebirds; Mar. = Marshbirds/Waterbirds; Wat. = Waterfowl
6.2
SELECTION OF PRIORITY HABITATS FOR THE PILOT PROJECT
Priority habitats are selected based on the coarse filter/fine filter approach since no specific,
quantitative criteria (e.g. the minimum area for species "X") are presented for priority species in
the BCR 13 conservation plan and the preferred habitat classes are not determined for each
species. The coarse filter approach consists of identifying the most common habitats that meet
the needs of many species and determining the minimum area thresholds required for breeding,
while the fine filter approach identifies habitat components specific to certain species that are not
identified by the coarse filter. The priority habitat classes used in the coarse filter approach fall
under three (3) major types of habitat:
1. Farmland: perennial crops, old fields
2. Forests: deciduous, mixed, coniferous
3. Wetlands: marshes, swamps, bogs, wet meadows
These habitats were determined by analyzing the integrated BCR 13 conservation plan (Fournier
et al. 2010), descriptive data on nesting habitats used to produce the plan (according to CWS-QC
experts), conservation plans produced for each species group (Chapdelaine and Rail 2004; Aubry
and Cotter 2007; Environment Canada 2010a, 2010b; Lepage et al. 2010) and the Quebec
Breeding Bird Atlas (Gauthier and Aubry 1996). All of these are breeding habitats, except for
the migration habitats of two species of scaups. These habitat classes must be discernible on the
digital land-use layers. The minimum area thresholds were determined for each habitat class
based on information taken from conservation plans specific to four groups of birds, as well as
from scientific literature and expert opinions. Appendix 2 presents the species targeted for each
habitat class. The criteria and thresholds selected, as well as a justification for their selection, are
described below:
Farmland
•
•
Maintain perennial crops (forage, pastures) > 40 ha (Environment Canada 2010a, 2010b)
Maintain old fields > 5 ha (Dettmers 2003; Schlossberg and King 2008; G. Falardeau,
CWS-QC, pers. comm.)
8
•
Favour square or rectangular fields that are non-elongated to minimize edge effects
(Renfrew et al. 2005)
Forests
•
•
•
Maintain forest patches > 1000 ha and increase connectivity (B. Drolet, CWS-QC, pers.
comm.)
Maintain woodlots > 100 ha in farming and urban areas (Environment Canada 2010a,
2010b)
Favour square or rectangular patches to minimize edge effects (Langevin and Bélanger
1994; Langevin 1997; Environment Canada 2004)
Wetlands
•
•
•
•
•
•
•
Preserve the critical habitats of the Least Bittern
Maintain large marshes (> 5 ha) (Brown and Dinsmore 1986; Gratton 2010)
Maintain large shrub swamps (> 5 ha) (Brown and Dinsmore 1986; Gratton 2010)
Maintain large bogs (> 20 ha) (Poulin et al. 2006)
Maintain wooded swamps (no area specified)
Maintain wet meadows (no area specified)
Maintain areas that are abundant in wetlands (wetland complex) (Calmé 1998; Naugle et
al. 2000; Fairbairn and Dinsmore 2001; Riffell et al. 2003; Tozer et al. 2010)
Finally, breeding habitats for 7 of the 48 priority species are not targeted by the coarse filter
criteria, while two species were selected because they use significant areas for feeding during
migration periods. The habitats identified by the fine filter criteria are described in Table 2.
Table 2 − Habitats identified by the fine filter criteria for priority species, whose needs are not targeted by the
coarse filter criteria
Species
Specific need
Common Nighthawk*
Nesting: Gravel roofs; bare soils in forested areas; rocky outcrops;
forest disturbance (fire and logging)
Nesting: Human structures; cliffs
Nesting: Sandy shores; sand pits
Nesting: Sandy shores; sand pits
Nesting: Cavities (natural or human-made)
Nesting: Sandy shores; sand pits
Nesting: Low-lying islands along the St. Lawrence waterway
Nesting: Aquatic beds along the shore (within 150 m) of Lake Saint-Pierre
Nesting: Aquatic beds along the shore (within 150 m) of Lake Saint-Pierre
Peregrine Falcon (anatum)*
Northern Rough-winged Swallow
Bank Swallow
Purple Martin
Belted Kingfisher
Common Tern
Greater Scaup
Lesser Scaup
*
Species listed at risk according to SARA or COSEWIC
9
7.0
STEP 2 – DATA COLLECTION AND MANAGEMENT
The data collected and used in the analyses were mainly geographical, physical (land use,
administrative boundaries) and biological (flora, fauna, species at risk, protected areas, critical
habitats for species at risk) in nature.
7.1
DATA SOURCES
The following spatial and mapping data sources were used to produce the land use map for
the study area:
Farmland: Classification of land use generated by Agriculture and Agri-Food Canada (AAFC)
using Landsat-7 images (2001–2002; 25 m resolution). Annual crops, perennial crops, old fields
and shrublands (land expanses occupied by relatively low woody vegetation, typically ± –2 m)
can be seen.
Forests: Ecoforestry information system of Quebec (SIEF) maps 3rd decadal on a scale of
1:20 000 generated by Quebec’s Ministère des Ressources naturelles. Three main classes of
forests (deciduous, mixed, coniferous), as well as signs of burning, logging and other
disturbances can be seen. The sheets were produced between 1991 and 2006, but the majority of
them were produced before 1996. The resolution used to convert the data into matrix format was
set at 5 m.
Urban areas: The land use map produced by the CWS (1999–2003) using Landsat-7 imagery
(25 m resolution). This was considered alongside the AAFC classification, which provides
a better delineation of urban areas. Green spaces in urban areas were classified as "other
anthropogenic". The resulting layer was filtered (3x3) to eliminate isolated pixels.
Wetlands: Four sources of information were combined by importance (quality of data, accuracy,
precision, date) in the following order:
1. Detailed mapping of wetlands by Ducks Unlimited Canada in the Montérégie area
(orthophotos from 2006) (GéoMont 2008);
2. Centre St-Laurent mapping (Ikonos imagery from 2000) on the banks of the St. Lawrence
river;
3. Modelling by Ducks Unlimited Canada using a formula applied to ecoforestry maps
(SIEF) (Ménard et al. 2006);
4. The CWS Conservation Atlas of Wetlands in the St. Lawrence Valley (combination of
Landsat-5 imagery from 1993–1994 and Radarsat imagery from 1999) (Bélanger and
Grenier 2003).
Eight classes of wetlands were identified: 1) shallow water (including submerged vegetation),
2) marsh, 3) swamp, 4) shrub swamp, 5) forest swamp, 6) wet meadow, 7) bog and
8) unidentified wetland.
In addition to these habitat classes, a few other landscape elements are required for certain
priority species:
•
The SIEF class "bare soil" corresponds to rocky outcrops in forested areas.
10
•
•
•
Sand and gravel pits were extracted using a combination of SIEF data and topographic
maps 1:50,000 (NTDB). Polygons were visually validated using high resolution images
and additional data (photos taken by helicopter, list of mining establishments [Institut de
la statistique du Québec 2010]).
Sandy shores were extracted from information on the CWS database (1994) about the
banks of the St. Lawrence River (Cornwall – Montmagny). A combination of attributes
(vegetation, slope) was performed to determine the location of sandy shores with steep
slopes.
Linear elements (electrical ROWs, watercourses/hydrography, railroads, etc.) were
extracted from topographic maps 1:20 000. The SIEF road cover was chosen because
roads in wooded and agriculture areas are identified.
7.2
PRODUCTION OF THE FINAL LAND USE MAP
The final land use map integrated data from the different sources into a single information layer.
This integration in raster mode involved using the AAFC classification as the "background"
layer and overlaying the other layers of information in order of priority based on the reliability
and quality of data, with the highest priority layer being added last:
•
•
•
•
Priority 1 – Polygons of wetlands
Priority 2 – AAFC classification for farmland and old fields
Priority 3 – SIEF for forested areas
Priority 4 – Map of urban areas
This land use map is the foundation for all geospatial landscape analyses. Given that several
sources of information date back more than 10 years, and significant changes could have
occurred in the landscape since then, a validation exercise was carried out based on the method
developed as part of the Canadian Wetland Inventory (CWI) (Grenier et al. 2007). This method
consists of randomly selecting polygons from each land use class that 2 photo-interpretation
teams then identify using recent high-resolution images (SPOT, Quickbird). The validation of
general habitat classes resulted in an overall accuracy of 76.0% and 71.8% for both teams.
Steps and methods used in the production of the land use map, detailed results and problems
encountered during the validation exercise are presented in Jobin et al. (2013).
7.3
DATA ON PROTECTED AREAS
In order to have a better understanding of the study area and to guide the prioritization of habitat
patches, a list of existing protected areas 3 was compiled. Several departments (MRN, MDDEFP,
EC-CWS) and conservation organizations that manage properties for the conservation of natural
habitats (DUC, NCC, FFQ, RMN, LSPBR) were consulted along with the Municonsult report
(2002). Some information on land tenure was also validated, in part with the help of the Registre
foncier du Québec (Quebec Land Register) (MRNF 2012a), because several designation and
names could be attributed to the same sites.
3
The MDDEFP (2002) defines a protected area as follows: a geographically defined expanse of land or water established under a legal and
administrative framework designed specifically to ensure the protection and maintenance of biological diversity and of related natural and
cultural resources.
11
Only protected areas with a protection status high enough to justify the actions covered by the
pilot project, such as the creation of buffer zones around them, were included in our analyses.
These include ecological reserves, nature reserves, wildlife refuges, migratory bird sanctuaries,
rare forests, habitats of threatened or vulnerable plant species, federal and provincial lands
without reserve status and land protected by an NGO charter or private organization. The
following areas were not included: wildlife habitats (muskrat habitat, waterfowl gathering area),
regional parks, city parks, managed marshes but not legally protected, exceptional forest
ecosystem proposed but not protected, conservation landmarks identified in a development
plan. Overall, there are 47 protected areas in the study area, covering a total of approximately
10 500 ha, more than half of which are sites protected by an NGO charter (Figure 3).
Figure 3 − Protected areas located in the study area and selected for analysis
7.4
DATA ON SPECIES AT RISK AND CRITICAL HABITATS
Four sources of information were used to obtain data on species at risk in the study area :
Critical habitats for species listed under SARA
As of July 2011, critical habitats were identified for only one species, the Least Bittern. Eight
sites were identified as critical habitats 4 in the study area, seven of which are human-made
impoundments (only Baie Saint-François is a natural wetland) (Figure 4).
Habitat polygons of avian species from CDPNQ and SOS-POP data
As of February 2011, breeding habitat polygons of certain avian species at risk, all associated
with wetlands, have been delineated from known observations. These polygons were extracted
from the Centre de données sur le patrimoine naturel du Québec (CDPNQ) and were traced for
4
Critical habitat refers to suitable breeding habitat (marshes and shrubby swamps containing tall and robust emergent herbaceous and/or
woody vegetation interspersed with areas of open water) located within these polygons (Environment Canada 2011).
12
the Least Bittern (n=14 polygons), the Short-eared Owl (n=3), the Nelson's Sharp-tailed Sparrow
(n=2) and the Sedge Wren (n=4) (Figure 5). In addition to these polygons, well-located and
recently used nesting sites for the Chimney Swift (n=29) and the Peregrine Falcon (n=4) and one
probable nesting site for the Bald Eagle were extracted from the SOS-POP database.
Figure 4 − Location of identified critical habitats of the Least Bittern
Figure 5 − Location of habitat polygons and breeding sites for avian species at risk
CDPNQ data for species other than birds
Records of species at risk collected in Quebec for decades, both for animals and flora, are
collated in the CDPNQ database. As of December 14, 2010, there were 131 well-located and
recent records of 57 species at risk on the pilot project territory, the majority of which were
vascular plants (Figure 6). Records of avian species were not considered because they are
already included in the SOS-POP database.
13
Figure 6 − Location of records of species at risk (other than avifauna)
Data from the Atlas of Amphibians and Reptiles of Québec
Records of the Snapping Turtle (Chelydra serpentina), a species of special concern in Canada,
were extracted from the database of the Atlas of Amphibians and Reptiles of Quebec (AARQ).
Among the 13 sightings in the study area (as of July 2011), 2 recent and accurate records were
retained (Figure 6).
8.0
STEP 2A – PLANNING AND MAPPING TOOLS
8.1
REVIEWING EXISTING SOFTWARES
In order to locate priority conservation areas, statistics must be calculated for each habitat patch
(e.g. number, geometry) and the connectivity between them, or any other spatial element that
may have an impact on the distribution and movement of wildlife in the area, must be assessed.
Existing softwares were assessed to determine which is most appropriate for achieving the
project objectives. Compatibility with ArcGIS and the frequency of updates to the software were
key criteria in the selection process.
Nine software packages frequently used in habitat conservation and landscape planning were
assessed, including CLUZ, ConsNet, C-Plan, Habitat Priority Planner, LINK, Marxan,
P.A.N.D.A., Vista and ZONATION. The Habitat Priority Planner (HPP) software was selected
for its ease of use, features for calculating landscape metrics based on FRAGSTATS (McGarigal
et al. 2002), compatibility with ArcGIS (ArcGIS 9.3 or 10 + Spatial Analyst) as well as its
capability to produce different scenarios based on changes in land use. However, a technical
problem made HPP incompatible for several months, and the landscape metrics calculated with
HPP could not meet all needs of this project. Therefore, FRAGSTATS had to be used to obtain
more comprehensive statistics on the landscape components.
14
The Corridor Designer software was selected amongst eight packages frequently used to design
wildlife corridors (Circuitscape, Connectivity Analysis Toolkit, Connefor Sensinode, Corridor
Designer, FunnConn, Guidos, Marine Geospatial Ecology tools [MGET] and Pathmatrix)
because of its ease of use, compatibility with ArcGIS and ability to create habitat suitability
models. The quality of the proposed corridors can be assessed with an additional optional
module by calculating several statistics, such as corridor width, distance between viable habitat
patches and the location of bottlenecks.
8.2
DECISIONS MADE PRIOR TO ANALYSIS
8.2.1
Spatial analysis scale
In addition to a descriptive analysis across the entire study area, the landscape was described
in smaller spatial units in order to facilitate the implementation of the conservation plan. The
landscape has been described through an administrative division of Quebec, i.e. the regional
county municipalities (RCMs), and an ecological division (watershed).
8.2.2
Overlapping of habitat patches
Landscape metrics calculated to describe the spatial configuration of habitat patches may be
biased if patches overlapping more than one RCM or a watershed are artificially cut at their
boundaries (e.g. a large wooded area located on the edge of two RCMs would be considered two
separate patches). Habitat patches overlapping more than one RCM or watershed were assigned
to each RCM or watershed, and landscape metrics were calculated on the actual boundaries of
the patch. Thus, the selection of priority habitat patches in the conservation plan is based on the
intrinsic character of the patches (total area) and not on an artificial division. However, habitat
patches located along the boundaries of the study area whose scope extends beyond those
boundaries were cut at the boundaries of the pilot project territory. This may have biased the
calculation of landscape metrics for these patches; however, large woodlots located at the fringe
of the study area were considered to select forest corridors.
8.2.3
Selection of landscape metrics
Landscape metrics were calculated using FRAGSTATS for the three landscape divisions (entire
study area, RCM, watershed) and at three spatial scales (patch, habitat class, entire landscape) to
describe the composition and configuration of habitats in the study area. The metrics calculated
for the habitat patches provide insight into the intrinsic nature of the patches (e.g. their shape).
The metrics calculated for the habitat classes provide insight into the relative importance of each
habitat class (e.g. the total area covered by each class) or the spatial configuration of habitat
patches (e.g. the proximity of patches of the same class). Finally, the metrics calculated for the
entire landscape inform us on the distribution patterns of habitat classes and thus the
heterogeneity and diversity of the landscape as a whole (e.g. Simpson's diversity index).
Several metrics calculated by FRAGSTATS are redundant and/or difficult to interpret. We
selected a few metrics based on knowledge gained in past studies (Jobin et al. 2001; MaheuGiroux et al. 2006; Latendresse et al. 2008), the metrics’ ease of interpretation and the literature
(Gustafson 1998; Hargis et al. 1998; Trani and Giles 1999; Jaeger 2000; Shao et al. 2001;
15
McGarigal et al. 2002; Corry 2004). Correlations were also used to select otherwise redundant
metrics. Landscape metrics were calculated on general and detailed land use habitat classes
(Table 3). Interior forest habitats were calculated using three edge widths (100 m, 200 m, 300
m). To calculate the proximity index (PROX) of patches in the same habitat class, the analysis
range was determined at 200 m for forests, 1 km for perennial crops and 5 km for wetlands.
Appendix 3 presents the matrices formed to calculate the edge contrast index (ECON).
Table 3 − Landscape metrics selected for the pilot project and calculated with FRAGSTATS
Utility
FRAGSTATS
acronym
AREA
CORE
CAI
ECON
FRAC
PROX
Unit
ha
ha
%
%
none
none
Spatial scale
Patch
Landscape metric
Area
Core area
Core area index
Edge contrast
Fractal dimension (shape)
Proximity index
Habitat
class
Number of patches
Class area
% cover of the landscape
Mean patch area
Coef. var. in patch area
Area-weighted mean patch area
Total core area
Mean core area
Core area percentage of landscape
Patch density
Clumpiness index
Splitting index
NP
CA
PLAND
AREA_MN
AREA_CV
AREA_AM
TCA
CORE_MN
CPLAND
PD
CLUMPY
SPLIT
none
ha
%
ha
%
none
ha
ha
%
n/100 ha
%
none
Shannon’s diversity index
Shannon’s evenness index
Simpson’s diversity index
Simpson’s evenness index
Contagion index
Aggregation index
Interspersion & juxtaposition index
Edge density
Patch density
Landscape shape index
Patch richness density
SHDI
SHEI
SIDI
SIEI
CONTAG
AI
IJI
ED
PD
LSI
PRD
none
none
none
none
%
%
%
m/ha
n/100 ha
none
n/100 ha
Landscape
(study area)
*
Forest
Patch
Land use Fragmentation corridors prioritization
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
See McGarigal et al. 2002 for a description of the selected metrics.
8.2.4
Forest fragmentation
Certain species of forest birds avoid crossing open areas that lie between two forest patches for
various reasons (increased risk of predation, unsuitable habitat, etc.). Several studies conducted
in Quebec (Desrochers and Hannon 1997; Rail et al. 1997; Duchesne et al. 1998; Bélisle and
Desrochers 2002) show that, in general, forest birds easily cross gaps < 30–50 m wide.
Therefore, anthropogenic structures in the landscape whose width is less than 50 m do not
16
contribute to the fragmentation of forest cover. Only highway right-of-ways (ROWs) contribute
to the fragmenting of forest cover, as their width is > 65 m (Bélanger et al. 2006). A habitat class
called "highway ROW" has been created so that highways and their ramps are shown on the land
use map by creating a buffer zone of 40 m extending from the centre of the highway to both
sides, for a total ROW of 80 m. Power line ROWs were visually inspected individually and only
those with a width greater than 50 m were considered to influence forest fragmentation. No
change in forest patches has been made to reflect the fragmentation caused by watercourses.
9.0
STEP 3 – LANDSCAPE ANALYSIS
Analysis of the landscape of the study area allows us to describe and understand the distribution
and the interaction of the various landscape elements. The approach used quantifies the
availability (composition) and spatial distribution (configuration) of habitats in order to assess
whether the landscape of the study area is able to maintain viable populations of priority bird
species and ensure the integrity of habitats. The landscape analysis is divided into two main
parts, each broken down into distinct sections:
1. Descriptive analysis
•
Description of the land use (study area, RCM, watersheds)
2. Functionality of the landscape
•
•
•
•
•
Comparison of the landscape with known reference thresholds
Identification of forest corridors
Application of coarse filter criteria
Prioritization of coarse filter patches
Application of fine filter criteria
All analyses presented in this report for RCMs were also done for watersheds.
Results for each watershed are presented in the detailed methodological report
(Jobin et al. 2013).
9.1
DESCRIPTIVE ANALYSIS
9.1.1
Land use: Study area
Figures 7 and 8 illustrate land use in the study area according to 7 general classes and 21 detailed
classes. Farmlands cover more than half of the territory, 31% of which is annual crops (corn,
soybean) and 20% is perennial crops (forage, hayfield, pasture) (Table 4). Patches of perennial
crops are, on average, much smaller (18 ha) than those of annual crops (40 ha). There are
143 patches of perennial crops covering > 100 ha. Forests cover 24% of the study area,
dominated by deciduous forests (12%) or mixed forests (9%). All types combined, the average
forest covers 45 ha, but 118 of these forests cover > 100 ha. More than 1500 forest patches
17
provide interior habitats when a 100 m edge is eliminated, covering more than 12% of the
territory (Table 5). These numbers are cut in half and by a quarter when wide edges (200 m
and 300 m respectively) are eliminated.
Figure 7 − Land use in the study area, general classes
Figure 8 − Land use in the study area, detailed classes
18
Table 4 − Area and description of the general and detailed habitats classes in the study area
Area
General class
Patch
2
%
Number
Mean (ha)
Coef. Var.
172
4.1
1649
10.4
1741.5
128
3.1
1835
7.0
791.9
Anthropogenic – Other
20
0.5
337
5.9
247.9
Highway ROW
24
0.6
4
595.5
130.2
1316
31.4
3340
39.4
1311.1
837
20.0
4644
18.0
525.6
837
20.0
4644
18.0
525.7
0
0.0
4
2.8
43.0
Detailed class
km
Anthropogenic
Anthropogenic
Annual crop
Perennial crop
Perennial crop
Orchard
Old field/Shrubland
42
1.0
1184
3.5
257.1
998
23.8
2221
45.0
774.7
Forest – Coniferous
65
1.5
720
9.0
157.8
Forest – Deciduous
506
12.1
2902
17.4
416.3
Forest – Mixed
354
8.5
2292
15.5
307.6
72
1.7
976
7.3
154.5
1
0.0
38
3.4
83.6
398
9.5
1082
36.8
840.7
Forest
Forest – Disturbance
Bare soil
Wetland
120
2.9
733
16.4
699.6
Unidentified swamp
Marsh
48
1.1
689
7.0
281.1
Shrub swamp
15
0.4
357
4.2
281.7
Forest swamp
125
3.0
545
23.0
267.7
Wet meadow
36
0.9
391
9.2
290.8
Bog
40
1.0
301
13.3
371.3
Shallow water
12
0.3
279
4.3
233.9
2
0.0
217
0.9
75.1
431
10.3
356
121.2
1543.5
Unidentified wetland
Open water
Table 5 − Description of interior forest habitats in the study area based on three forest edge widths
Edge width
Interior forest habitat
100 m
200 m
300 m
Number of patches
Total area (ha)
Percent cover of the
study area
Mean patch area (ha)
1534
51 014
839
24 585
419
12 012
12.2
5.9
2.9
23.0
11.1
5.4
Wetlands cover nearly 10% of the landscape, mostly located in the Lake Saint-Pierre flood plain.
The wooded swamps and marshes bordering the lake cover large areas, particularly on the south
shore of Lake Saint-Pierre, in the Berthier-Sorel archipelago and in the Lavallière, SaintFrançois and Maskinongé bay areas. We also note the presence of large bogs in the TroisRivières and Daveluyville areas. Apart from a few large rivers flowing into the study area (SaintMaurice, Yamaska, Saint-François, Richelieu, Bécancour), open water areas (10% of the
territory) are represented by Lake Saint-Pierre and the St. Lawrence River. Anthropogenic areas
(4%) are concentrated around the cities of Trois-Rivières (130 000 inhabitants) and Sorel-Tracy
(35 000 inhabitants) and in municipalities with a lower population, including Bécancour
19
(11 000 inhabitants), Nicolet (8 000 inhabitants), Louiseville (8 000 inhabitants) and
Berthierville (4 000 inhabitants). Finally, old fields and shrublands cover only 1% of the
territory, and are generally small in size (avg. = 3.5 ha). The largest ones fall under the electric
ROWs.
9.1.2
Land use: RCM
The study area was divided up based on the boundaries of the RCMs in order to describe the
landscape in terms of territorial divisions favourable to the implementation of the habitat
conservation plan. Some of the RCMs are partially located in the study area, and the areas
analyzed vary greatly between the RCMs (Table 6). Thus, only 36% of the Maskinongé RCM
is included in the study area while the Nicolet-Yamaska and Trois-Rivières RCMs are fully
included. Similarly, the Nicolet-Yamaska and Maskinongé RCMs together cover more than half
of the study area, while the other four RCMs each cover less than 15% of the territory.
Table 6 − Area covered by the RCMs in the study area
Total
2
RCM
Area within the
% of the RCM located
% cover of the
2
area (km )
study area (km )
within the study area
study area
D'Autray
Maskinongé
Trois-Rivières
Bécancour
Nicolet-Yamaska
Pierre-De Saurel
1353
2643
335
1234
1189
639
587
957
335
584
1190
542
43.4
36.2
100.0
47.3
100.0
84.9
14.0
22.8
8.0
13.9
28.4
12.9
Total
7393
4195
56.7
100.0
Table 7 presents the absolute (ha) and relative (%) area of the general and detailed habitat
classes in each RCM, and figures 9 and 10 are used to compare the RCMs to determine the
habitats present and the distribution of habitat classes in each RCM. We observe that:
•
•
•
•
•
•
•
•
•
•
The relative importance of forest cover is higher in the northern (Maskinongé and
D'Autray RCMs) and eastern RCMs (Bécancour and Trois-Rivières RCMs).
Forests cover less than 20% of the Pierre-De Saurel and Nicolet-Yamaska RCMs.
About two thirds of anthropogenic areas are found in the Pierre-De Saurel and TroisRivières RCMs.
There are few old fields in the study area and virtually none in the Pierre-De Saurel
RCM.
Farmland covers more than half of the RCMs, except for the Trois-Rivières RCM, where
it covers less than 20% of the territory.
Annual crops dominate in the Pierre-De Saurel and D'Autray RCMs.
Wetlands (all types combined) cover between 6% and 12% of each RCM.
There are very few marshes, shrub swamps, wet meadows and shallow water areas in the
Trois-Rivières RCM.
There are no wet meadows in the Bécancour RCM.
There are very few bogs in the D'Autray and Pierre-De Saurel RCMs.
20
Table 7 − Area (km2 and %) of general and detailed classes of habitats in the RCMs
D'Autray
Maskinongé
Bécancour
Trois-Rivières
Nicolet-Yamaska
Pierre-De Saurel
km 2
%
km 2
%
km 2
%
km 2
%
km 2
%
km 2
%
14
2.4
24
2.5
68
20.3
12
2.1
19
1.6
35
6.4
10
1.7
15
1.6
54
16.0
8
1.4
13
1.1
28
5.1
Anthropogenic – Other
2
0.3
2
0.2
8
2.3
1
0.1
2
0.2
6
1.0
Highway ROW
3
0.4
7
0.7
7
2.1
3
0.6
3
0.2
1
0.2
Annual crop
225
38.3
299
31.2
36
10.6
128
22.0
396
33.2
233
43.0
Perennial crop
103
17.5
177
18.5
27
7.9
158
27.1
272
22.9
101
18.6
103
17.5
177
18.5
27
7.9
158
27.1
272
22.8
101
18.6
0
0.0
0
0.0
0
0.0
0
0.0
0
0.0
0
0.0
7
1.2
9
0.9
10
3.1
5
0.9
9
0.8
2
0.3
147
25.1
259
27.1
118
35.4
170
29.2
227
19.0
77
14.1
Forest – Coniferous
6
1.0
18
1.9
17
5.1
10
1.7
8
0.7
6
1.1
Forest – Deciduous
63
10.8
125
13.1
47
14.1
101
17.3
125
10.5
44
8.2
Forest – Mixed
68
11.6
97
10.1
47
13.9
44
7.6
76
6.4
23
4.2
Forest – Disturbance
9
1.5
19
1.9
7
2.2
15
2.6
18
1.5
4
0.7
Bare soil
1
0.2
0
0.0
0
0.0
0
0.0
0
0.0
0
0.0
49
8.3
63
6.5
33
9.9
60
10.4
132
11.1
62
11.4
18
3.1
21
2.2
3
0.9
11
1.9
45
3.8
21
4.0
Unidentified swamp
4
0.7
3
0.3
4
1.3
15
2.6
11
0.9
11
2.0
Shrub swamp
5
0.8
2
0.2
0
0.0
1
0.2
7
0.6
0
0.1
Forest swamp
13
2.3
24
2.5
14
4.2
19
3.2
41
3.4
14
2.6
Wet meadow
6
1.1
5
0.5
0
0.1
0
0.0
14
1.2
10
1.9
Bog
0
0.1
7
0.8
11
3.2
10
1.8
9
0.8
2
0.4
Shallow water
1
0.2
0
0.0
0
0.1
3
0.6
5
0.4
2
0.4
Unidentified wetland
0
0.1
0
0.0
0
0.1
1
0.1
0
0.0
0
0.0
43
7.3
126
13.2
43
12.8
49
8.5
136
11.4
34
6.3
General class
Detailed class
Anthropogenic
Anthropogenic
Perennial crop
Orchard
Old field/Shrubland
Forest
Wetland
Marsh
Open water
21
Figure 9 − Cover (%) of the seven general habitat classes by RCM
Figure 10 − Breakdown of general habitat classes in the RCMs
A more detailed analysis of the quality of forests and woodlots in the RCMs indicates that
woodlots offering the largest and most numerous areas of interior habitats are found in the
D'Autray and Maskinongé RCMs, regardless of the width of the forest edge (100 m, 200 m,
300 m) (Figure 11). On the other hand, the Pierre-De Saurel and Nicolet-Yamaska RCMs offer
the least amount of interior forest habitats. In addition, the four metrics used to describe the
degree of forest fragmentation (Figure 12) indicate that woodlands are the rarest and most
fragmented in the Pierre-De Saurel and Nicolet-Yamaska RCMs. The combination of this
information on the quality of forest habitats (interior forest habitats, fragmentation) makes it
possible to determine the RCMs that offer the most suitable woodlots to forest birds (Figure 13).
22
Figure 11 − Relative importance (%) of the cover and mean area (ha) of interior forest habitats
in the RCMs for forest edges that are 100 m, 200 m and 300 m wide
Figure 12 − Landscape metrics (4) illustrating woodlot fragmentation in the RCMs
23
Figure 13 − RCMs ordered by the importance of interior forest habitats and woodlot fragmentation
9.2
FUNCTIONALITY OF THE LANDSCAPE
Different approaches have been used to assess whether the landscape in our study area provides
functional habitats for priority species. The availability and integrity of nesting habitats, as well
as landscape permeability allowing for the movement of forest birds, were evaluated in several
ways: comparison with known reference thresholds, identification of forest corridors, application
of coarse filter and fine filter criteria.
9.2.1
Comparison of the landscape with known reference thresholds
Habitat availability in the study area was compared to reference thresholds values known to
sustain minimum viable wildlife populations and help maintain selected ecosystem functions.
The thresholds used are those created by Environment Canada in Ontario in the development
of guidelines for the conservation and restoration of wetland, riparian and forest habitats
(Environment Canada 2004). Comparing current wildlife habitat cover in the various spatial
units with these thresholds makes it possible to determine the parts of the study area where
habitat cover is adequate and regions where conservation and restoration needs would be
justified. This approach was also adopted as part of the National Agri-Environmental Standards
Initiative (NAESI) in Canada (Maheu-Giroux and Belvisi 2007; McPherson et al. 2009; Neave et
al. 2009).
Forests
Studies have shown that when forest cover of a landscape is above a certain threshold, ranging
from 20 to 35%, the persistence of bird species was virtually ensured or that habitat configuration
had little or no effect on species richness or abundance (Andrén 1994; Fahrig 1997; Tate 1998;
Villard et al.1999; Rompré et al. 2010). Also, certain bird species avoid forest edges, and their
abundance increases based on the area of interior forest habitat (Austen et al. 2001). It is even more
imperative to determine the availability of interior forest habitat knowing that forest cover is highly
reduced and fragmented in several RCM in southern Quebec (Bélanger and Grenier 2002).
•
•
•
At least 30% of the territory (study area, RCM, watershed) should be forested.
At least 10% of the territory (study area, RCM, watershed) should be forested,
representing interior forest conditions at 100 m or more from the forest edge.
At least 5% of the territory (study area, RCM, watershed) should be forested,
representing interior forest conditions at 200 m or more from the forest edge.
24
Forests cover less than 24% of the study area (Table 8), which means that the current landscape
is not adequate to sustain viable bird communities. On a smaller scale, the threshold of 30%
forest cover is reached only in the Trois-Rivières RCM (Figure 14). In the Nicolet-Yamaska and
Pierre-De Saurel RCMs, forests are very scarce, covering less than 20% of the territory.
Table 8 − Comparison of the study area and the RCMs landscape with reference thresholds values known to
support forest bird communities and maintain wetlands and watercourses functions
Forest
Region
Study area
RCM D'Autray
Maskinongé
Trois-Rivières
Bécancour
Nicolet-Yamaska
Pierre-De Saurel
% total
cover
% interior habitat
100 m
200 m
> 30%
23.8
> 10%
12.2
> 5%
5.9
25.1
27.1
35.4
29.2
19.0
14.1
16.1
21.1
14.3
14.7
8.8
5.8
9.6
11.8
6.7
7.2
3.5
1.9
Threshold value
Wetland
% of sites with 100%
natur. veget.
% total
cover
Buffer zone=100m
(max. number of
> 6% ou 10%
sites)
9.5
43.5 (n=3512)
8.3
6.5
9.9
10.4
11.1
11.4
48.2 (n=745)
44.5 (n=661)
37.6 (n=340)
29.7 (n=445)
45.8 (n=918)
46.6 (n=474)
Riparian habitat
% natur.
veget.
% natur. veget. (buffer zone)
Width=30m Width=100m
> 75%
36.2
> 75%
34.3
> 75%
32.2
33.2
41.2
59.0
38.4
30.8
30.5
31.6
39.2
57.8
36.0
29.2
27.6
28.6
35.9
54.7
35.1
27.7
25.9
Note: The numbers highlighted in green indicate that the threshold is reached or exceeded.
Figure 14 - Cover (%) of forest cover by RCM
Despite the fact that forests cover less than 30% of the territory, they still offer numerous interior
forest habitats, regardless of whether the forest edge is 100 m or 200 m wide. In fact, the
respective thresholds of 10% and 5% of interior forest coverage for both forest edge widths
are reached across the entire study area, as well as in the majority of RCMs and watersheds
(Table 8). However, these thresholds are not met in the Nicolet-Yamaska and Pierre-De Saurel
RCMs (Figures 15 and 16).
25
Figure 15 − Cover (%) of interior forest habitat by RCM (edge = 100 m)
Figure 16 − Cover (%) of interior forest habitat by RCM (edge = 200 m)
Wetlands
Wetlands should cover at least 10% of a watershed, or at least 6% of a subwatershed, in order
to ensure adequate spatial distribution of wetlands across the landscape (Detenbeck et al.
1999; Environment Canada 2004). Also, to maintain key wetland functions and attributes,
it is recommended to preserve an area 100 m wide of natural vegetation around wetlands
(Environment Canada 2004).
•
•
5
At least 10% of a watershed and at least 6% of a subwatershed should be
covered by wetlands.
A buffer zone of 100 m or more in width, composed of natural vegetation 5
should be preserved around wetlands.
Natural vegetation includes the following habitat classes: coniferous forests, mixed forests, deciduous forests, other wetlands, old fields and
shrublands, water, rocky outcrops, sand dunes, plantations.
26
With 9.5% wetland coverage (Table 8), the study area almost reaches the minimum threshold
required to ensure suitable habitats for wetland species. The 6% threshold targeted for
subwatersheds here refers to the parts of the RCM that are located in the study area. This
threshold is reached for almost all spatial units, with wetlands even covering > 10% in several
RCMs (Figure 17). Note, however, that the wetlands are largely concentrated in the immediate
vicinity of Lake Saint-Pierre and are scarce elsewhere in the study area.
Figure 17 − Percentage of wetland coverage by RCM
There are 3512 patches of wetlands in the study area and less than half of them (43.5%) have a
100-m-wide buffer zone that is completely covered with natural vegetation (including aquatic
areas) (Figure 18, Table 9). Less than 60% of wetlands in the RCMs have such a vegetated
buffer zone. Interestingly, of the 1986 wetlands whose buffer zone do not have full natural
vegetation coverage, 407 have a buffer zone with 90% coverage, and 818 are 75% covered with
natural vegetation. On the other hand, 100 wetlands have less than 10% of their buffer zone
covered with natural vegetation. The shrub swamps and wet meadows are those wetland types
mostly bordered by a buffer zone of 100% natural vegetation, with 70% and 62% of patches,
respectively, while less than 35% of undefined wetlands, undefined swamps, wooded swamps
and peat bogs have a buffer zone completely covered with natural vegetation.
There are significant differences between the RCMs (Table 9). A high proportion (74%) of wet
meadows in the Nicolet-Yamaska RCM has a buffer zone completely covered with natural
vegetation; this figure is only 14% in the Trois-Rivières RCM.
27
Figure 18 − Percentage of natural vegetation cover (including aquatic areas)
in a 100-m-wide buffer zone around each wetland patch
Table 9 − Number (n) and proportion (%) of wetland patches with 100-m-wide buffer zones that are completely
covered with natural vegetation (including aquatic areas)
Study area
Wetland type
Shallow water
Marsh
Forest swamp
Shrub swamp
Unidentified swamp
Unidentified wetland
Wet meadow
Bog
Total
Total
n
279 135
733 363
545 166
357 249
689 211
217
57
391 242
301 103
3512 1526
RCM
Wetland type
Shallow water
Marsh
Forest swamp
Shrub swamp
Unidentified swamp
Unidentified wetland
Wet meadow
Bog
Total
D'Autray
Total
n
34
16
251 126
90
43
122
78
80
17
50
7
102
67
16
5
745 359
%
48.4
49.5
30.5
69.7
30.6
26.3
61.9
34.2
43.5
%
47.1
50.2
47.8
63.9
21.3
14.0
65.7
31.3
48.2
Maskinongé
Total
n
%
8
4 50.0
148
80 54.1
136
32 23.5
80
54 67.5
98
40 40.8
27
13 48.1
99
53 53.5
65
18 27.7
661 294 44.5
Trois-Rivières
Total
n
%
3
0 0.0
72
38 52.8
60
13 21.7
1
0 0.0
80
34 42.5
26
6 23.1
14
2 14.3
84
35 41.7
340 128 37.6
Bécancour
Total
n
%
55
20 36.4
58
20 34.5
70
16 22.9
8
6 75.0
133
30 22.6
62
12 19.4
0
–
–
59
28 47.5
445 132 29.7
NicoletYamaska
Total
n
113
47
119
60
148
48
137 107
183
44
27
9
119
88
72
17
918 420
%
41.6
50.4
32.4
78.1
24.0
33.3
73.9
23.6
45.8
Pierre-De Saurel
Total
n
%
76
52 68.4
103
47 45.6
61
19 31.1
11
4 36.4
124
51 41.1
27
12 44.4
64
36 56.3
8
0 0.0
474 221 46.6
Riparian habitats
Maintaining natural habitats along watercourses helps improve water quality. An Ontario study
revealed a degradation of watercourses where vegetation cover was less than 75% along the
banks (Steedman 1987; McPherson et al. 2009).
•
•
At least 75% of of stream length should be naturally vegetated.
At least 75% of a 100-m-wide riparian habitat along a watercourse should be
naturally vegetated.
28
There are more than 31 616 km of riparian habitats in the study area, and only 36% of those are
naturally vegetated, which is far from the goal of 75%. The riverbanks are mostly covered with
natural vegetation (67%), while the banks of small streams have less coverage of natural habitats
(25%). If the vegetation is not natural, the banks are often covered by annual or perennial crops
(52% and 41% respectively). Few banks in the study area are in urban settings (5%). By
including perennial crops in the natural vegetation class, the percentage of natural vegetation
along the watercourses greatly increases but does not reach the 75% threshold for small streams
and creeks.
Table 10 shows the proportion of riparian habitats composed of natural vegetation by RCM.
Apart from riparian habitats in the Bécancour area, the percentage of riparian habitats covered
with natural vegetation is still below the threshold of 75% for the three main types of
watercourses (gullys, streams and rivers) in each RCM. It is particularly worrying that these
percentages are often less than 30% in several RCMs. Areas with the most highly modified
riparian habitats are the Pierre-De Saurel and Nicolet-Yamaska RCMs, while the area where
riparian habitats are mostly natural is the Trois-Rivières RCM.
Table 10 − Total length of the riparian habitats (watercourses, small bodies of water) and total length of riparian
habitats of natural vegetation based on the type of watercourses in each RCM
Total length
RCM
Type
Metres
Length natur. veget.
%
Metres
Total length
%
RCM
D'Autray
Type
Length natur. veget.
Metres
%
Metres
%
Bécancour
Gully
2300992
50.2
442427
19.2
Gully
2597903
53.6
755511
29.1
Stream
1338149
29.2
493643
36.9
Stream
1546169
31.9
562237
36.4
River
774058
16.9
482359
62.3
River
537009
11.1
442097
82.3
Pond
152921
3.3
95432
62.4
Pond
98631
2.0
44852
45.5
Lake
14032
0.3
7890
56.2
Lake
55961
1.2
50019
89.4
4761
0.1
837
17.6
536
0.0
536
100.0
4585448 100.0
1523125
33.2
Culvert
Pool
Total
Culvert
4618
0.1
1620
35.1
Pool
4503
0.1
3730
82.8
4844794 100.0
1860065
38.4
Total
Maskinongé
Nicolet-Yamaska
Gully
3680931
54.1
1065736
29.0
Gully
6049761
65.9
1305956
21.6
Stream
2294198
33.7
1188677
51.8
Stream
1743505
19.0
536839
30.8
River
459837
6.8
255177
55.5
River
1184313
12.9
869310
73.4
Pond
258663
3.8
192559
74.4
Pond
161670
1.8
85490
52.9
Lake
44689
0.7
44039
98.5
Pool
16939
0.2
16939
100.0
Reservoir
44664
0.7
40778
91.3
Basin
9233
0.1
7505
81.3
Pool
16597
0.2
14080
84.8
Lake
6588
0.1
6588
100.0
9303
0.1
2559
27.5
Culvert
1741
0.0
116
6.7
6808881 100.0
2803606
41.2
Total
9173750 100.0
2828744
30.8
Culvert
Total
Trois-Rivières
Pierre-De Saurel
Stream
875601
45.3
478362
54.6
Gully
2384250
55.6
463581
19.4
Gully
755917
39.1
483719
64.0
Stream
926458
21.6
255206
27.5
River
183095
9.5
108364
59.2
River
768028
17.9
478969
62.4
Pond
101998
5.3
67172
65.9
Pond
138551
3.2
57792
41.7
14044
0.7
2030
14.5
Lake
32312
0.8
32312
100.0
2984
0.2
553
18.5
Pool
17267
0.4
17267
100.0
1933640 100.0
1140199
59.0
Culvert
11688
0.3
734
6.3
9783
0.2
0
0.0
4288338 100.0
1305860
30.5
Culvert
Lake
Total
Basin
Total
29
At least 75% of a 30-m-wide buffer adjacent to streams should be naturally vegetated on both
sides of the streams to maintain satisfactory water quality (Environment Canada 2004). A buffer
zone of > 100 m wide along watercourses is often required to provide suitable habitats for
several species of birds (Fischer 2000). The percentage of natural vegetation in riparian corridors
30 m and 100 m in width was calculated. 6 Similar to the analysis of riparian habitats along
watercourses (previous reference threshold), we have observed that natural vegetation covers
less than 35% of the riparian corridors for both widths analyzed (Table 11). The 75% threshold
was also not reached for any RCM. In fact, the trends observed with the previous reference
threshold (% of stream length naturally vegetated) are the same as those observed when riparian
habitats are 30 m or 100 m wide. The threshold of 75% cover of natural vegetation in these
riparian habitats is not met for the study area as a whole or for the RCMs even when perennial
crops are considered as natural vegetation.
Table 11 − Area (ha and %) of natural vegetation in 30 m and 100 m wide riparian habitats in the study area and
the RCMs. The figures in parentheses include perennial crops in the calculation of natural vegetation
cover.
Buffer zone = 30 m
Buffer zone = 100 m
Natural vegetation
Region
ha
%
14956.7
34.3 (60.0)
Study area
D'Autray
2023.8
31.6 (53.7)
RCM
Maskinongé
3757.5
39.2 (62.4)
Maskinongé
Trois-Rivières
1589.2
57.8 (69.3)
Bécancour
2308.7
Nicolet-Yamaska
Pierre-De Saurel
Study area
RCM
9.2.2
Natural vegetation
Region
ha
%
43240.6
32.2 (58.0)
5627.5
28.6 (51.2)
10627.2
35.9 (59.6)
Trois-Rivières
4684.1
54.7 (67.5)
36.0 (69.9)
Bécancour
6912.4
35.1 (69.0)
3666.6
29.2 (58.7)
Nicolet-Yamaska
10713.0
27.7 (57.2)
1610.9
27.6 (50.3)
Pierre-De Saurel
4676.3
25.9 (48.1)
D'Autray
Identification of forest corridors
While identifying the nesting habitats of priority bird species (coarse filter and fine filter), one
must also consider the travel corridors between forest patches because several species of birds
require a continuous forest cover to move through the landscape on a daily basis and for the
dispersal of populations (Beier and Noss 1998). The conservation plan for BCR 13 does not
specify the criteria required for forest bird species to move across the landscape. However, one
of the criteria used is to increase connectivity between forest patches > 1000 ha. The Corridor
Designer software used pre-set parameters to identify potential corridors to connect those forest
6
The criterion suggested by Environment Canada (2004) reads as follow: "Streams should have a minimum 30 m wide naturally vegetated
adjacent-lands areas on both sides, greater depending on site-specific conditions." This criterion is measured by analyzing the type of habitat
adjacent to a buffer zone 30 m or 100 m in width along the watercourse (G. Bryan, CWS-Ontario, pers. comm.). We adapted this criterion to
measure only the percentage of natural vegetation found in riparian corridors 30 m and 100 m in width.
30
patches > 1000 ha. The proposed corridors were then evaluated based on criteria related to their
spatial configuration and discontinuities.
Recent studies have focused on methods for identifying corridors and on associated decisions
required to develop adequate predictive models (Beier et al. 2008; García-Feced et al. 2011).
However, there is no consensus in the literature on a "minimum corridor width" or a "minimum
distance between woodlots" because of the large variability of bird communities studied, the
landscape context, the geographical location of the studies, etc. Some criteria have been
proposed for the minimum corridor width (Stauffer and Best 1980; Keller et al. 1993; Spackman
and Hugues 1995; Hodges and Krementz 1996; Duchesne and Bélanger 1997; Environment
Canada 2004; Mason et al. 2007) and for the distance between woodlots (Duchesne et al. 1998,
1999). Kampf and Stavast (2005) defined distance thresholds between habitat patches within a
given corridor based on the size of the birds: 1000 m for large birds, 500 m for medium-sized
birds and 200 m for small birds. These habitat patches located inside corridors are called
stepping stones and their presence is vital for discontinuous corridors or if they include less
suitable habitats such as areas with intensive farming activities (Bennett 1999; Van der Sluis et
al. 2004). The following criteria were selected in this study to evaluate corridor functionality for
forest birds:
•
•
The corridors should have a minimum width of 100 m, or ideally 200–300 m.
The distance between woodlots should be < 200 m.
Creating a map of potential habitats
The first step in creating corridors consisted of selecting criteria that can be represented spatially
and used to create the map of potential habitat use by forest birds (Habitat Suitability Model).
Each criterion is weighted according to its relative importance to the movement of birds, and a
score is then assigned to each habitat patch in order to rank the quality of each patch for each of
the criteria. In the context of this study, six criteria were used (Table 12). A quality index was
first assigned to each land use habitat class according to their probability of use by forest birds;
some habitat classes are more conducive than others to the movement of forest species (e.g. old
fields compared to corn fields). The permeability of the matrix of non-forested habitats that
would prove hostile to the movement of forest birds has been considered, whereby the habitats
were not categorized into a simple dichotomy of suitable and unsuitable habitats (Baum et al.
2004; Debinski 2006; Watling et al. 2011). To do this, the relative importance of each habitat
class for the movement of forest birds was assessed by five CWS experts and the average
assessment was used. This approach of calling on experts is a common practice for Corridor
Designer users (Majka et al. 2007). We considered that birds prefer woodlots located more than
a kilometre from urban centres (> 50 ha) and avoid woodlots located close (within 250 m) to
them. Landscape metrics calculated in FRAGSTATS for each of the habitat patches were used to
assign weights to the four other criteria: TOTAL AREA for size, PROXIMITY for isolation,
FRACTAL DIMENSION INDEX for shape and EDGE CONTRAST INDEX for contrast. Three or
four classes were created for each of the criteria, and the Jenks optimization method was used to
determine the best distribution of values or each criterion (de Smith et al. 2011). For each habitat
patch in the study area, the weight of each criterion was multiplied by the score associated with
each patch, and a map of potential habitats was created as a basis for creating corridor scenarios
(Figure 19).
31
Table 12 − Weighting of the criteria to determine corridors in Corridor Designer
Land use habitat class
Annual crop
10
Anthropogenic
20
Bog
30
Forest
100
Forest swamp
75
Marsh
10
Old field/Shrubland
60
Open water
5
Perennial crop
20
Shallow water
5
Shrub swamp
50
Unidentified swamp
50
Unidentified wetland
10
Wet meadow
20
Weight
50
Total area
0 – 30 m
30 – 100 m
100 – 200 m
200 m and over
Weight
Distance from a urban centre
0 – 250 m
0
250 – 1000 m
50
1000 m and over
100
Weight
5
Proximity index
0–5m
0
5 – 250 m
50
250 m and over
100
Weight
12
Fractal Dimension Index (shape)
1.00 - 1.06
100
1.06 – 1.12
50
1.12 and over
0
Weight
5
25
50
75
100
20
Edge Contrast Index
0 – 40
100
40 – 65
50
65 and over
0
Weight
8
Figure 19 − Habitat suitability model for priority birds, generated by Corridor Designer
Creating corridor scenarios
Corridor scenarios are designed to connect the 13 forest patches > 1000 ha found in the study
area, all of which are located in the northern part of study region. Four woodlots > 500 ha
(patches #14, 15, 16 and 17, Figure 20) located south of the St. Lawrence River were added in
order to study the connectivity of forest patches throughout the study area. In total, 21 corridors
were proposed (Figures 20 and 21). Only those habitat patches with a suitability value > 60 were
selected to establish the corridors; they were selected based on the cost of travelling, in other
32
words, the habitat characteristics that influence the species' ability to move between two areas.
The selected corridor is the best biological choice for the species and may contain one or more
branches. Several scenarios are proposed according to the desired size of the corridor; this size
is calculated based on its relative cover to the rest of the territory. Thus, the smallest possible
corridor is the one that covers 0.1% of the landscape, representing the best route based on
geographical and biological habitat characteristics between two patches. However, it is often
unnecessary to select a scenario that covers a large proportion of the landscape, because
the software will be forced to identify habitats that are unsuitable for forest birds.
Figure 20 − Selected forest patches and location of established corridors
Figure 21 − Examples of proposed corridor scenarios
33
Selecting corridors
We used a minimum corridor width of 300 m, as a corridor width between 100 and 300 m
facilitates the movement of forest birds. However, several of the proposed corridors cross
through habitats that are not conducive to forest birds (potential < 60 as for annual crops).
Therefore, it was determined that the selected corridors should have a width of at least 300 m
across at least 75% of their area. For this, the statistics for the smallest scenario proposed (0.1%)
were calculated, and the size was increased until the thresholds were met. This was done using
the Evaluation Tools extension tool. This extension tool also allows one to identify bottlenecks
based on the minimum width identified, and to represent them spatially. Some corridors were
eliminated because they provided too close of a connection between forests (e.g. scenario 1_2) or
because the best travel options for species between the forests were outside of the study area
(e.g. scenario 2_3). On the other hand, other scenarios were selected, such as scenario 2_17,
because they provide a connection between two forest patches even if only a section of the route
is found within the study area. In total, 14 proposed corridors were selected (Figure 22).
Figure 22 − Final selection of potential corridor scenarios (P = main area branch, S = secondary area branch)
and location of bottlenecks (width < 300 m)
Assessment of the quality of potential corridors
A detailed analysis of the proposed corridors helped identify the functional corridors (those
where the criteria are met) as well as corridors where problem areas are noted (bottlenecks,
distance between woodlots). Duchesne et al. (1999) presented a detailed methodology for
assessing the quality of forest corridors, including various criteria such as the length of the forest
corridor, the average minimum width, the number and size of forest cover interruptions, the
number of bottlenecks, and habitat heterogeneity. Many of these criteria were calculated
automatically by Corridor Designer, including the frequency and location of bottlenecks
(Figure 22), as well as the length and proportion of the corridor with a width greater than the
threshold of 300 m. In terms of the habitat heterogeneity criterion, the cover (%) of each general
habitat class was evaluated for each corridor (Figure 23). Overall, forests cover half of the area
34
where the proposed corridors are found, while annual and perennial crops cover 24% and 18%
respectively. This helps identify corridors where habitats are less suitable for forest birds, such
as corridors 2_17 and 15_16, which cross large areas covered with annual crops. On the other
hand, some corridors, such as 6_8_S and 9_10_S, are almost entirely composed of wooded areas,
and priority conservation actions could concentrate on the bottlenecks (width < 300 m) and on
areas where the distance between woodlots is > 200 m so that corridor functionality and quality
could be increased. For example, plantation programs could increase forest cover, habitats that
are less conducive to forest bird dispersal in the matrix could be converted into more suitable
habitats (e.g. crop abandonment, conversion of annual crops to perennial crops), and windbreaks
and riparian corridors could be planted to increase connectivity.
100%
90%
80%
Forest
70%
Perennial crop
60%
Annual crop
50%
Old field
40%
Open water
30%
Wetland
20%
Anthropogenic
10%
0%
Figure 23 − Cover (%) of the general habitat classes in the potential corridors
Figure 24 shows an enlarged section of corridor 10_13. We can see the proposed route, located
in a suitable environment, where the length and width of the corridor are adequate (in blue). We
also see the sections of the corridor that cross unsuitable environments (red) and the bottlenecks
(in purple), most of which are between 500 m and 1 km long. Portion of the proposed corridor go
through unsuitable habitats, such as areas covered with annual crops, but these are short
distances (< 200 m), which will have little impact on the movement of species. However, in
cases where this distance is > 200 m, small forest patches could be created to serve as stepping
stones, or more realistically, a conversion of annual crops to perennial crops could significantly
improve the quality of the corridor.
35
Figure 24 − Detailed analysis of corridor 10_13
9.2.3
Application of coarse filter criteria
Coarse filter criteria were used to determine minimum area thresholds for several habitat types
needed to sustain viable populations of several priority bird species (see section 6.2). These
thresholds were determined for agricultural habitats (perennial crops, old fields), forest habitats
and wetlands (marshes, shrub swamps, bogs), and the habitat polygons that fulfill these
thresholds were extracted from our land use maps. Tables 13 and 14 present the results of
applying coarse filter criteria in the study area and in each of the RCMs, while Figure 25 shows
the spatial location of patches that meet the coarse filter criteria for each habitat class.
Table 13 − Description of the habitat patches that meet the coarse filter criteria in the study area
Area (ha)
Habitat
Number
Mean
Stand. error
Min
Max
Total
Perennial crop > 40 ha
Old field > 5 ha
Forest > 100 ha
Marsh > 5 ha
Shrub swamp > 5 ha
359
171
118
169
62
171.6
15.8
697.8
65.7
17.4
15.8
1.5
125
17.9
3.1
40.1
5.1
101.1
5.1
5.2
2728.3
141.5
9570.9
2279.9
143.6
61 590
2704
82 344
11 104
1080
Bog > 20 ha
34
93.2
20.9
21.4
666.9
3168
36
Table 14 − The number of habitat patches that meet the coarse filter criteria in the study area and by RCM
Area
Number of patches/habitat class
Farmland
Forest
Wetland
Perennial Old
Shrub
crop
field
Patch Woodlot
Marsh swamp
Bog
> 40 ha > 5 ha
> 1000 ha > 100 ha
> 5 ha > 5 ha > 20 ha
Region
(km2)
Study area
4194.8
359
171
13
118
169
62
34
RCM D'Autray
Maskinongé
Trois-Rivières
Bécancour
Nicolet-Yamaska
Pierre-De Saurel
586.8
957.1
334.9
583.7
1189.9
542.4
54
76
15
58
116
54
20
39
42
23
42
8
2
6
3
2
4
0
11
22
15
23
42
15
54
34
15
22
28
27
24
13
0
6
19
1
0
3
12
9
9
3
Note: The sum of all patches for all RCMs may be higher than the total in the study area because some patches overlap > 1 RCM.
(1)
(2)
(3)
(4)
37
(5)
(6)
Figure 25 − Location of all habitat patches that meet the coarse filter criteria for (1) perennial crops,
(2) marshes, (3) old fields, (4) shrub swamps, (5) woodlots, and (6) bogs
•
A total of 359 perennial crop patches, or 7.7% of 4 644 patches found in the study area,
are larger than 40 ha. These patches cover an average of 172 ha and their total area is
61 590 ha. The greatest number of these large perennial crop patches, many of them
covering more than 1 000 ha, are found in the Nicolet-Yamaska and Maskinongé RCMs.
•
A total of 171 old fields, or 14.4% of the 1 184 polygons of old fields found in the study
area, are larger than 5 ha. These old fields cover an average of 16 ha and their total area
is 2 700 ha. The greatest numbers of these old fields, many of them covering more than
100 ha, are found in the Trois-Rivières, Nicolet-Yamaska and Maskinongé RCMs.
•
There are 13 forests > 1 000 ha in the study area, 6 of which are entirely or partly located
in the Maskinongé RCM. None of these forests are found in the Pierre-De Saurel RCM.
A total of 118 forest patches, representing just 5.3% of 2 221 patches found in the study
area, are larger than 100 ha. These patches cover an average of nearly 700 ha and their
total area is 82 350 ha. The greatest numbers of these large forest patches are found in the
Nicolet-Yamaska, Bécancour and Maskinongé RCMs.
•
A total of 169 marshes, or 23.1% of the 733 marshes found in the study area, are larger
than 5 ha. These marshes cover an average of 66 ha and their total area is 11 100 ha.
Many marshes > 5 ha can be found in each RCM, but most of them are found in the
D’Autray and Maskinongé RCMs. Several marshes cover more than 100 ha.
•
Only 62 shrub swamps (17.4% of 357 shrub swamps identified in the study area) > 5 ha
are found in the study area. These marshes cover an average of 17 ha and their total area
is 1 080 ha. The majority of shrub swamps > 5 ha are found in the D'Autray, NicoletYamaska and Maskinongé RCMs, while 1 can be found in the Pierre-De Saurel RCM
and none are found in the Trois-Rivières RCM. Only 9 shrub swamps are > 25 ha. Note
that other shrub swamps are certainly present in the study area, but the images used to
generate the land use mapping cannot distinguish between shrub swamps and wooded
swamps in many areas.
•
Only 34 bogs (11.3% of the 301 bogs found in the study area) > 20 ha are found in
the study area. These bogs cover an average of 93 ha and their total area is 3 170 ha.
38
The majority of bogs > 20 ha are found in the Trois-Rivières, Nicolet-Yamaska
and Bécancour RCMs, while none are found in the D'Autray RCM. Only 8 bogs
cover > 100 ha.
Finally, the 545 wooded swamps and 391 wet meadows found in the study area have been
considered priority sites because no minimum area threshold is known for these types of habitat
(Table 15). Figure 26 shows their spatial distribution. The wooded swamps are mainly found on
the shores of Lake Saint-Pierre, in the Berthier-Sorel archipelago and in the eastern part of the
study area, while wet meadows are found mainly on the shores of Saint-Pierre, in the BerthierSorel archipelago and on the southern end of the Saint-François and Lavallière bays.
Table 15 − Number of wooded swamps and wet meadows in the study area and by RCM
Region
Forest swamp
Wet meadow
545
391
D'Autray
Maskinongé
Trois-Rivières
Bécancour
Nicolet-Yamaska
90
136
60
70
148
102
99
14
0
119
Pierre-De Saurel
61
64
Study area
RCM
Note: The sum of all patches for all RCMs may be higher than the total in the study area because some patches overlap > 1 RCM.
(1)
(2)
Figure 26 − Location of (1) wooded swamps, and (2) wet meadows in the study area
9.2.4
Prioritization of coarse filter patches
Figure 27 shows the spatial location of habitat patches that meet the coarse filter criteria. All
these plots are, a priori, important for nesting birds and deserve to be protected, or at the very
least, anthropogenic pressures that may affect them should be reduced. However, due to their
large number, these plots must be prioritized in order to identify those that can contribute more
to the needs of priority species in the study landscape.
39
Figure 27 − Location of all habitat patches that meet the coarse filter criteria
There are numerous ways to prioritize habitats (see Langevin and Bélanger 1995; McGarigal et
al. 2005; Qiu 2010; Holzmueller et al. 2011), and these methods are often dictated by common
factors such as the presence of species at risk or rare ecosystems, the proximity to a protected
area, the size and shape of habitat patches, and the identification of criteria specific to target
species (e.g. threat reduction around habitats of species at risk). A weighting factor is then added
to the criteria in order to calculate an index for each habitat patch. This type of multi-criteria
analysis was used to prioritize farm woodlots and wetlands in southern Quebec (Langevin 1997;
Nature-Action 2009; Gratton 2010; CRECQ 2012).
This type of multi-criteria analysis method was used to prioritize coarse filter habitats for each
habitat class. Habitat criteria were identified based on the specific needs of bird guilds and the
landscape context where the habitat patches are located, and a weighting factor was given to
each criterion according to its relative importance. For each criterion, a score was then assigned
to each patch according to its value in relation to other patches within the established classes
(percentiles, ranges with determined limits, etc.). Finally, the weighting used for each criterion
was multiplied with the score given to each patch for each criterion, and the sum of these
multiplications produced a final prioritization index for each patch, where C is the weight of the
criterion i, P is the score given to each patch for criterion i, and n is the number of criteria:
n
Index = ∑ CiPi
i =1
The criteria were divided into two groups. The first group refers to the attributes of the patches
given their importance for the establishment and maintenance of breeding bird populations. The
shape index and % of interior habitat reduce edge effects, the edge contrast index focuses on
patches found in a landscape matrix that is less hostile for birds, the proximity index favours the
selection of patches located in areas dominated by the same habitat class, and the % of natural
40
vegetation in a buffer zone of 100 m around wetlands is used to select patches that are less prone
to anthropogenic pressures. The second group of criteria is used to prioritize patches based on
their ecological role in the landscape of the study area, such as the creation of buffer zones
around critical habitats of species at risk, wetlands or existing protected areas in order to reduce
pressures and threats that can affect those sites. Habitat patches already located in a protected
area (see section 7.3) received a score of "0" for the criterion that is designed to prioritize
patches according to their proximity to a protected area, because they do not require
conservation action. On the other hand, those patches that are partially included in a protected
area received the maximum score, because the portions bordering protected areas should be
prioritized so that buffer zones can be created around them. The distance to a significant urban
centre (> 50 ha) is also considered. Similarly, forests, shrublands and swamps located in a
proposed forest corridor are prioritized. Finally, the presence and number of species at risk
designated under SARA (species that are endangered, threatened or of special concern) or An Act
Respecting Threatened or Vulnerable Species in Quebec (species that are threatened, vulnerable,
or likely to be designated threatened or vulnerable) are also considered when prioritizing
patches. Prioritization of patches was done across the study area and all criteria were considered
in one analysis. The criteria chosen and the relative weight given to each criterion are presented
in Table 16. A justification of the criteria chosen for each habitat class and of the score given to
the habitat patches for each criterion are presented in Jobin et al. (2013).
10
10
20
10
10
10
20
10
Presence/abundance of species at risk
15
15
10
Wet meadow
Marsh
Bog
10
10
10
10
10
10
10
10
10
10
20
20
20
20
20
15
15
15
15
15
15
15
15
15
15
15
15
10
20
15
10
10
15
10
15
10
15
10
5
15
10
20
10
10
10
10
10
10
10
% natural vegetation (buffer = 100 m)
Ecological role in the landscape
Proximity of a critical habitat
Proximity of a wetland
Proximity of a protected area
Distance to a urban centre > 50 ha
Located in a potential corridor
Shrub swamp
Forest
Old field
10
10
Forest swamp
Prioritization criteria
Patch attribute
Shape index
Edge contrast index
% interior habitat (edge = 200 m)
Proximity index
Perennial crop
Table 16 − Criteria chosen and weighting used to prioritize patches for each coarse filter habitat class
15
15
15
Note: The proximity index is calculated for a distance of 200 m for forests, 1 km for perennial crops and 5 km for wetlands.
The habitat patches that meet the coarse filter criteria are then classified according to the final
prioritization index, which allows one to select the higher quality patches for each habitat class,
with a high index being representative of a high-priority patch. This selection can be made
randomly or based on statistics. Various scenarios were tested (see Jobin et al. 2013), and two
41
selected scenarios are shown in Figure 28 (top 25 patches for each habitat class) and in Figure 29
(patches included in the top 10th percentile of each habitat class). Notably, Pearson's correlations
indicate that the final patch prioritization index is independent of their size, which shows that
the criteria chosen are appropriate for prioritizing conservation sites compared to traditional
methods, which are often based on habitat size. Other prioritization criteria have also been
considered but were not retained in the final process due to data availability or duplication with
selected criteria (patch uniqueness, largest patch index in a RCM, etc.).
Figure 28 − Location of the 25 patches with the highest priority index for each coarse filter habitat class
Figure 29 − Location of the patches whose prioritization index is in the top 10th percentile
for each coarse filter habitat class
42
9.2.5
Application of fine filter criteria
The spatial distribution of the habitats identified by the fine filter criteria (forest disturbances,
bare soil in forests, sand/gravel pits, sandy shores with steep slopes) is shown in Figure 30. This
information is incomplete for certain habitat types such as sand pits, because there is not enough
information about their spatial distribution in the study area. There are forest disturbances in
almost all of the forest patches in the study area (except the wooded swamps bordering Lake
Saint-Pierre). Sandy shores with steep slopes are located along the banks of some islands and
watercourses in the Sorel region, and the bare soil (rocky outcrops) are mostly found at the top of
forests in the northwestern portion of the study area (D'Autray RCM).
Figure 30 − Location of habitat patches that meet the fine filter criteria
Habitats identified by the fine filter approach are also used by species already included in the
coarse filter criteria, such as the Eastern Whip-poor-will, which may nest in sand pits, or the
American Kestrel and the Northern Flicker, which also nest in burnt-out areas and logging areas.
Finally, following the identification of priority habitats using coarse filter and fine filter criteria,
the nesting needs of the two species, the Peregrine Falcon (anatum) and the Purple Martin, still
have not been considered. Similarly, it is well known that the Common Nighthawk nests on
gravel roofs in urban areas. These species prefer to nest on anthropogenic structures, and these
needs are considered in section 11.
10.0 STEP 4 − FINAL ANALYSIS AND SPECIFIC ISSUES
10.1 ANALYSIS OF THE CURRENT SITUATION
Analyzing the functionality of the landscape can help produce a diagnosis of the landscape’s
ability to provide functional habitats for priority bird species. Although many patches in each
priority habitat class are adequate to meet breeding needs of priority species, comparison of the
43
landscape with habitat thresholds known to support sustainable bird populations identifies gaps
in habitat availability in the landscape. Also, some of the proposed corridors present problems
that make them less suitable for the movement of forest birds.
It should be pointed out that the existing protected areas are almost exclusively located around
Lake Saint-Pierre, reflecting past efforts in wetland protection in this region. Hence, the majority
of terrestrial habitats in the study area do not hold any protection status.
Farmland
The study area includes farmland habitats suitable for bird species that require large areas of
perennial crops (e.g. Bobolink), particularly in the Maskinongé, Bécancour and NicoletYamaska RCMs. However, the marked dominance of annual crops indicates that farmlands are
generally not suitable to grassland birds. Similarly, old fields and shrublands are scarce in the
study area (they are practically absent from the Richelieu region) and are mostly located under
power line ROWs, therefore subject to periodic perturbations for maintenance purposes.
Forests
Although the forests found in the study area provide quality interior habitats for area-sensitive
forest bird species in almost all of the RCMs, the forest cover in the study area (24%) is below
the minimum 30% required to sustain forest bird communities. Therefore, it is imperative to
preserve existing forested areas and to try to increase forest cover, particularly in the southern
part of the study area (Pierre-De Saurel and Nicolet-Yamaska RCMs) where highly fragmented
forests cover less than 20% of the territory. This is also particularly urgent as very few forests
are located in existing protected areas. Note also that the northern (Maskinongé and D'Autray
RCMs) and eastern (Bécancour and Trois-Rivières RCMs) parts of the study area are well
forested. Various reasons, both historical and current, may explain how these habitat patches
were preserved in the landscape (e.g. low-quality soils for agriculture, difficult access, woodlots
used for logging or sugar maple production), which reduces their conservation needs. Note that
the bare soils found in forests are almost entirely located at the top of the mountains in the
northwest region of the study area (D'Autray RCM) and deserve special attention.
Wetlands and riparian corridors
Although wetlands cover 9.5% of the study region, nearing the 10% threshold required to
maintain quality habitats, and the 6% threshold targeted for subwatersheds are met in all the
RCMs, the spatial distribution of wetlands shows that they are highly concentrated in the
immediate vicinity of Lake Saint-Pierre and are scarce elsewhere in the study area. Many of the
wetlands are currently protected by existing protected areas (e.g. Nicolet MBS, Lavallière Bay,
île du Moine, Leon-Provencher ecological reserve), and it is important to pay particular attention
to the extensive wetlands located elsewhere in the St. Lawrence Lowlands (bogs in the TroisRivières, Nicolet-Yamaska and Bécancour RCMs; wooded swamps in the eastern part of the
study area).
In addition, the integrity of the wetlands is at risk because the 100 m buffer zone surrounding
them is totally naturally vegetated on 44% of sites. Special efforts should be made to reduce the
presence of human activity (agriculture, urban infrastructure) along wetlands, particularly in
44
areas identified as critical habitat for the Least Bittern. Similarly, riparian habitats adjacent to
watercourses are heavily disturbed by human activity, and conservation measures should be
implemented to improve the water quality and to offer quality riparian habitats for wildlife.
These efforts are particularly needed in the Nicolet-Yamaska and Pierre-De Saurel RCMs, where
watercourses are bordered by natural vegetation along only 30% of their banks.
Forest corridors
Of the 14 proposed corridor scenarios, only three have over 75% of forest cover across their total
area (6_8_P, 6_8_S and 9_10_S). Some proposed corridors seem to be functional, as is the case
for those located in the Trois-Rivières and Bécancour RCMs, and crop abandonment, especially
near the bottlenecks, could substantially increase the quality of the corridors. On the other hand,
several of the proposed corridors hardly seem suitable for birds, such as corridors 2_17 and
15_16, which are covered by annual crops on nearly half of the proposed routes. Few stepping
stones exist in less conducive environments, and the numerous bottlenecks may affect the
movement of birds. That being said, habitat connectivity might be deficient in several places. For
some of the more problematic corridors, such as those located in the D'Autray and Pierre-De
Saurel RCMs, isolated initiatives would probably not be enough to provide viable and functional
corridors for forest birds in areas of intense farming activity. Since considerable efforts would be
needed to improve the situation, an in-depth analysis would be needed to assess whether the
functionality of these corridors can be restored.
Even though the cover of several habitats meets the minimum area thresholds at various scales
(study area, RCMs, watersheds), it is impossible to determine whether the current landscape
can maintain sufficient numbers of breeding pairs in priority habitats because there are no
quantitative population targets for priority species in the BCR 13 conservation plan. In addition,
our study area covers only a portion of BCR 13, making it more difficult to interpret the
quantitative targets that could have been developed for a region larger than our study area.
10.2 REGIONAL ISSUES AND THREATS
The Lake Saint-Pierre region is located in the St. Lawrence Lowlands, the most populous
ecoregion of Quebec, where anthropogenic pressures are the highest. Therefore, several
development issues may cause conflict when it comes to protecting natural environments. Jobin
et al. (2007) studied the recent habitat dynamics in the St. Lawrence Lowlands for the period
1993–2001, and observed a significant conversion of perennial crops into annual crops. This
trend was particularly significant in the D'Autray, Pierre-De Saurel, Bécancour and NicoletYamaska RCMs. Agriculture intensification in the Lake Saint-Pierre flood plain was also noted
by Richard et al. (2011) for the period 1950–2000. Jobin et al. (2007) also noted that forested
areas declined during the 1993–2001 period in all RCMs of the study area, mainly due to the
increase in cultivated areas and, to a lesser degree, to urbanization. Savoie (2002) also noted a
loss of forest area in the Centre-du-Québec region for the same period.
Monitoring of wetland coverage in the Lake Saint-Pierre region shows that the area covered by
these habitats remained relatively stable between the 1990–1991 and 2000–2002 periods, but
that the spatio-temporal dynamic of the habitat classes was highly variable (Jean and Létourneau
2011). As such, several wetlands were converted into open water or were drained for farming
purposes on the south shore of the lake, while low marshes became high marshes dominated
45
by Reed Canarygrass (Phalaris arundinacea) and by wooded swamps in the Lavallière and
Saint-François bay areas. These changes would be related to large-scale changes in water levels
of the St. Lawrence River. On the other hand, the wetlands in the Montérégie region showed
a significant decrease (22% of the sectors considered) in size due to farming (GéoMont and
Environment Canada 2008) between 1964 and 2006, and several wetlands located in the Lake
Saint-Pierre agricultural plain might have suffered the same fate during this period. In fact,
wetlands continue to disappear in southern Quebec despite existing regulations (Queste 2011).
For example, the area used for cranberry production increased from 1000 ha in 1999 to 2500 ha
in 2009 in the Centre-du-Quebec region, where 80% of Quebec's cranberry growers are found
(Poirier 2010), and the remaining natural bogs in the landscape may be altered if this trend holds.
A visual analysis of satellite images available on Google Earth© shows recent conversions of
bogs into cranberry fields in areas located on the edge of the study area (municipalities of
Saint-Louis-de-Blandford, Manseau and Notre-Dame-de-Lourdes).
Industrial and urban development continues to modify the landscape in the Lake Saint-Pierre
region. Jobin et al. (2007) as well as Jean and Létourneau (2011) noted a sharp increase in
anthropogenic developments around the city of Trois-Rivières, at the expense of natural habitats.
A visual analysis of satellite images available on Google Earth© shows recent residential
developments in areas once covered by farmland or forests (e.g. Trois-Rivieres, Nicolet and
Sorel-Tracy). Finally, the possible extraction of shale gas in the St. Lawrence Valley, a region
targeted for this type of operation (MRNF 2012b), could change the landscape and affect the
availability and integrity of habitats available to nesting birds.
11.0 STEP 8 – CONSERVATION PLAN FOR THE STUDY AREA
The conservation actions included in the BCR 13 conservation plan (Fournier et al. 2010)
along with the analysis of the spatial distribution of priority habitats in the study area enable
us to target habitats and areas where protective measures would be required. The proposed
conservation plan for the pilot project is divided into three sections:
•
•
•
Priority habitat patches with spatial reference
Priority habitats without spatial reference
Landscape elements to consider for maintaining ecological processes
11.1 PRIORITY HABITAT PATCHES WITH SPATIAL REFERENCE
The priority habitats that should be maintained to provide functional and viable habitats for
priority species in this project were determined in the previous sections. These habitats all have
a spatial reference, which allows us to locate them in the study area.
Habitat polygons for avian species at risk
Some sites are high priority because of the known presence of nesting bird species at risk.
Notably the identified critical habitats for the Least Bittern and the known nesting sites for avian
species at risk: the Least Bittern (occurrences other than the critical habitats), the Short-eared
Owl, the Sedge Wren and Nelson's Sharp-tailed Sparrow. Admittedly, these priority sites are
based on our current knowledge of these species’ distribution, which parallels the more
46
traditional “hot spot” approach for site prioritization. However, protection of these sites is
of utmost importance because of the status of these species.
Point data on nesting avian species at risk
Point records on nesting sites frequented by avian species at risk are also available; notably, the
Chimney Swift, which nests in chimneys, and the Peregrine Falcon, whose known nesting sites
are located on anthropogenic structures.
Coarse filter polygons (prioritization of patches)
The scenario where the top 25 patches of each habitat class are prioritized is included in the
conservation plan. It should be noted again that all patches that meet the coarse filter criteria are,
a priori, important for nesting birds and should be considered (see sections 9.2.3 and 9.2.4).
Fine filter polygons
The habitat classes identified by the fine filter are bare soil in forested areas, forest disturbances
(burning, logging), sand/gravel pits and sandy shores with steep slopes.
Corridors
Forest corridors identified in section 9.2.2. Bottlenecks found in these corridors are also
illustrated.
Figure 31 shows the spatial distribution of these priority sites in the territory of the pilot project
as well as the location of existing protected areas. Priority terrestrial habitats include forests in
the northwest part of the study area and in the Trois-Rivières region, as well as in the eastern part
of the study area located south of the St. Lawrence River. Few priority sites are located in the
agricultural areas. Most priority sites located in aquatic areas are clustered on the south shore of
Lake Saint-Pierre and in the Berthier-Sorel archipelago.
11.2 PRIORITY HABITATS WITHOUT SPATIAL REFERENCE
Other habitat components described in the BCR 13 conservation plan and specific to the project's
priority species are not discernible on the land-use digital layers and therefore are not considered
in the coarse and fine filter criteria. These habitats are generally uncommon but essential for
some species. Appropriate conservation measures should be implemented to ensure that these
specific needs are met.
47
Figure 31 − Priority habitat patches with spatial reference
(see Figure 27 for the spatial distribution of all patches that meet the coarse filter criteria)
48
These measures should be supplemented by specific actions that promote safe nesting conditions
for these priority species. Table 17 lists some of these conservation measures and the species that
are targeted.
Table 17 − Conservation measures and species targeted for priority habitats without spatial reference
Conservation measures
Species targeted
Preserve trees and snags with large diameters (> 30 cm)
Install and maintain nesting boxes
Encourage the implementation of protective perimeters
proposed by the MRNF around nesting sites, both on
private and public lands
Promote the upkeep of gravel roofs in urban areas
Promote the upkeep of old farm buildings
Promote the conservation and upkeep of suitable
chimneys in urban areas
Peregrine Falcon
Common Nighthawk
Barn Swallow
Chimney Swift
Avoid disturbances in sand pits near nesting sites
Avoid disturbing areas of shallow water and aquatic
vegetation in Lake Saint-Pierre
Avoid the use of herbicides and promote the mechanical
maintenance of vegetation in power line ROWs
*
American Kestrel; Barred Owl; Brown Creeper;
Chimney Swift; Eastern Screech-Owl; Northern Flicker;
Northern Saw-whet Owl; Wood Duck
American Kestrel; Eastern Screech-Owl; Northern Sawwhet Owl; Purple Martin*; Wood Duck
Bank Swallow; Belted Kingfisher; Northern Roughwinged Swallow; Whip-poor-will
Greater Scaup; Lesser Scaup (foraging areas during
migration)
American Woodcock; Brown Thrasher; Eastern
Kingbird
This species nests almost exclusively in human-made nesting boxes (only a few breeding records in natural cavities exist for the 20th century
in eastern North America) (Brown 1997).
11.3
LANDSCAPE ELEMENTS TO CONSIDER FOR MAINTAINING
ECOLOGICAL PROCESSES
Landscape elements must be considered in the conservation plan in order to maintain ecological
processes and the integrity of habitats in the study area. General conservation measures that
reduce anthropogenic pressures on watercourses, waterbodies and wetlands should be
established.
Vegetated riparian habitats
The need to maintain vegetated riparian habitats along watercourses is raised in the BCR 13
conservation plan (Fournier et al. 2010) in order to maintain water quality for birds nesting or
feeding in these habitats. The effectiveness of riparian corridors in reducing diffuse agricultural
pollution is well documented in Quebec (Duchemin and Majdoub 2004; Gagnon and Gangbazo
2007; Duchemin and Hogue 2009), and various guidelines have been issued in this regard
(OMAFRA 2004; Bentrup 2008; Fondation de la faune du Québec and Union des producteurs
agricoles 2011). The information presented in section 9.2.1 clearly shows that watercourses
found in the study area are highly vulnerable to anthropogenic pressures (pollution, erosion,
etc.), and the proportion of watercourses with naturally vegetated riparian habitats and adjacent
49
buffer zone is far from the 75% threshold. Corrective measures are strongly needed to remedy
this situation.
Buffer zones around wetlands
The recommendations for riparian habitats also apply to wetlands, as the 100-m-wide buffer
zone around them is totally naturally vegetated on less than half of the sites. Wetlands are
therefore vulnerable to diffuse agricultural pollution, and additional efforts to protect shorelines
are required, especially in areas dominated by intensive farming activity where the drift and
runoff of pesticides and fertilizers into aquatic ecosystems can be harmful (Roy 2002; Lee et
al. 2003; Lovell and Sullivan 2006).
Toposequence of wetlands in Lake Saint-Pierre
The abundance and diversity of wetlands found in the Lake Saint-Pierre region led to this area
being designated as a Ramsar site and as a Biosphere Reserve. There is still an assemblage of
wetlands in certain areas that form a natural toposequence ranging from areas of shallow water
and aquatic beds to marshes, wet meadows, shrub swamps and wooded swamps. These are
remnants of coastal wetlands that were subject to the natural variations in water levels of the
St. Lawrence River and should be given special attention.
11.4 SPECIAL CONSIDERATIONS OF THE CONSERVATION PLAN
Certain general facts should be considered in order to guide conservation activities toward
habitats that can provide maximum benefits for the project's priority species. First, the area
covered by some habitat classes does not reach the minimum area threshold required to provide
a functional landscape for nesting birds in several of the RCMs. Maintaining forests and large
woodlands located in the agricultural matrix or in urban areas should be prioritized, as they
contribute to the diversity of the regional avifauna (Environment Canada 2007; Minor and
Urban 2010; Oliver et al. 2011), while prioritizing the patches identified in section 9.2.4.
Similarly, in areas where forest cover exceeds the 30% threshold, it would be appropriate to
maintain woodlots located in areas prone to human development, such as those located on the
outskirts of urban areas, in order to ensure that the forest cover remains above this threshold
(Environment Canada 2007).
Again, various factors can explain how these habitat patches were preserved in the landscape,
despite the existing anthropogenic pressures that may impact them at various levels (see
section 10.1). Some priority sites may be located in areas less prone to human development and
are protected “de facto” because they are not under immediate threats. It is therefore important to
quantify the anthropogenic pressures that may impact priority sites and concentrate conservation
efforts where they are most urgent. Second, the restoration or creation of habitats should be
considered in order to increase the size of these habitat classes where required. The landscape
analysis to identify areas to be restored in the study area remains to be done. The creation of
buffer zones around existing protected areas or critical habitats for species at risk should also be
considered, whereby priority sites that meet the coarse filter criteria are first on the list.
50
Furthermore, breeding site selection for several bird species acts at several spatial scales, and
landscape composition plays an important role in this selection. Several species will nest in
landscapes where their preferred nesting habitats are locally abundant. This is why conservation
efforts should be focused on preserving priority habitat patches in areas where these habitats are
already present, in order to provide an optimal landscape for nesting birds. For example, several
wetland-dependant species select their breeding sites in areas where wetlands abound (Brown
and Dinsmore 1986; Calmé 1998; Naugle et al. 2000; Fairbairn and Dinsmore 2001; Riffell et
al. 2003; Forcey et al. 2011). The protection of natural bogs located near exploited bogs should
therefore be prioritized (Environment Canada 2010b) because of their vulnerability to future
exploitation, as is currently the case with cranberry fields in the Centre-du-Québec region.
Similarly, it is well documented that grassland bird species prefer to nest in areas where
perennial crops dominate the matrix and avoid areas dominated by annual crops, forested and
anthropogenic areas (Hamer et al. 2006; Veech 2006; Renfrew and Ribic 2008; Jobin and
Falardeau 2010). As such, grassland birds are amongst the species groups where population
declines are the steepest in southern Quebec (North American Bird Conservation Initiative
Canada 2012). Therefore, it would be justified to preserve and increase the availability of
perennial crops in areas that are already well covered by these habitats in order to increase their
importance at the regional level, as in the Maskinongé and Bécancour RCMs. On the other hand,
preserving perennial crops in areas where these habitats are rare would most likely be less suited
because these habitat patches may become "sink habitats." A socioeconomic analysis and
modelling of future landscape scenarios would help determine the most suitable areas for these
species. An in-depth analysis is thus required before considering a large-scale conversion of
annual crops to perennial crops in areas that are now under intensive farming activity, such as
the Pierre-De Saurel RCM.
11.5 CONSERVATION PLAN SPECIFIC TO EACH RCM AND WATERSHED
The landscape analysis and the prioritization of coarse filter patches were done globally for the
entire study area so as not to divide habitat patches according to administrative boundaries that
are irrelevant from a biological point of view. However, the implementation of the conservation
plan's recommendations will require appropriate tools useful to local stakeholders. The
recommendations in the conservation plan are thus detailed for each RCM, i.e., the scale where
the territorial planning takes place, and for each watershed, i.e., the ecological division where
watershed-based organizations work towards the conservation of habitats at the regional scale.
Table 18 presents a summary of priority habitats and conservation measures required in each
RCM and watershed. However, in order to keep the report short, a detailed conservation plan
(including maps showing the spatial distribution of priority habitats) is presented only for the
Bécancour RCM as an example. The comprehensive conservation plan for each RCM and each
watershed is presented in the associated methodological report (Jobin et al. 2013).
51
Table 18 − Summary of the presence of protected areas, priority habitat types and conservation measures
required in each RCM and watershed
11.5.1
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
√
Proposed forest corridor
√
√
√
√
√
√
√
√
√
√
√
√
Woodlot and forests
√
√
√
√
√
√
√
√
√
√
√
Bare soil
√
√
√
√
√
√
√
√
√
√
√
√
Sandy shore with steep slope
√
Sand/gravel pit
√
Old field
√
√
Vegetated buffer zone (wetland)
√
√
√
√
√
√
√
√
√
Maintain and create
Vegetated riparian habitat
√
√
√
√
√
√
√
√
√
√
√
Fine filter polygon
Forest disturbance
√
√
√
√
√
√
Bog
√
CDPNQ polygon
√
Wet meadow
√
√
Forest swamp
√
√
Shrub swamp
√
√
√
√
√
Marsh
√
√
√
√
√
√
√
√
√
√
√
√
√
Forest
√
√
√
Perennial crop
Watershed
Bayonne
Maskinongé
Du Loup/Yamachiche
Saint-Maurice
Batiscan
Bécancour
Nicolet
Saint-François
Yamaska
Richelieu
√
√
√
√
√
√
Coarse filter polygon (25 best/habitat class)
SOS-POP nesting site
RCM
D'Autray
Maskinongé
Trois-Rivières
Bécancour
Nicolet-Yamaska
Pierre-de Saurel
Critical habitat for the LEBI
Existing protected area
Species at risk
√
√
√
√
√
√
√
√
√
√
√
√
√
Example of a detailed conservation plan: Bécancour RCM
Figures 32 and 33 show the spatial distribution of priority sites for the Bécancour RCM. Less
than 50% of this RCM, namely its western region, is included in the study area. Descriptive
results and conservation actions proposed below thus refer only to this area of the RCM. A large
watercourse, the Bécancour River, flows in this area. More than half of the study area of this
RCM is occupied by farmland, especially perennial crops, with several priority patches
exceeding the threshold value of 40 ha. This is the only RCM where perennial crops dominate
annual crops. Several old fields and shrublands cover > 5 ha, with some covering > 50 ha.
Forests occupy nearly 30% of the territory (desired threshold), and several forest patches cover
> 100 ha, with some covering > 1000 ha. Many of these patches have also been identified as
priority sites, and three forest corridors have been proposed. In addition, several forest
disturbances, potentially conducive to certain species of birds, are present.
None of the priority wetland patches selected is located along the St. Lawrence River. They are
almost exclusively located within and around the Léon-Provancher ecological reserve, except
from a few large bogs (> 20 ha) located in the centre and east of the RCM. Overall, the area
covered by wetlands in the RCM reached the threshold value of 10%. The presence of the
ecological reserve allows for the preservation of habitats used by species at risk, including
the Least Bittern and the Bald Eagle.
52
Figure 32 − Spatial distribution of priority habitats in the Bécancour RCM (including those amongst
the 25 best patches of each habitat class determined for the whole study area)
Figure 33 − Spatial distribution of all habitat patches that meet the coarse filter criteria in the Bécancour RCM
The proposed conservation actions are to:
•
•
•
ensure the protection of wetlands used by the Least Bittern and other marsh birds in
the Bécancour industrial park;
conserve large existing forests (to reach and maintain 30% of forest cover in the RCM),
with special attention paid to those identified as priority sites;
maintain existing forests in order to promote the movement of forest birds (preservation
of existing forest corridors and promotion of proposed corridors);
53
•
•
•
•
•
•
•
evaluate the possibility of improving the functionality of the proposed corridors through
targeted actions (e.g. planting, conversion of annual crops to perennial crops, crop
abandonment in less productive sectors);
conserve bogs located in the Saint-Sylvère region and north of Sainte-Marie-deBlandford;
maintain perennial crops in order to promote a regional concentration of forages and
pastures conducive to grassland birds, mainly around the Léon-Provancher ecological
reserve and in the eastern part of the RCM;
verify if birds (e.g. swallows, Belted Kingfisher) nest in the sand/gravel pits east of
Bécancour and, where appropriate, limit disturbances that these operations may cause
during the nesting season;
ensure that existing wetlands remain intact;
promote the creation and preservation of naturally vegetated riparian habitats along
watercourses;
promote the establishment and preservation of natural habitats in a 100 m buffer zone
surrounding wetlands.
11.6 LIMITS OF THE CONSERVATION PLAN
The proposals submitted are based on the best information currently available on land use of the
study region and on the knowledge of the breeding ecology of birds in Quebec. Some habitats
are highly dynamic, such as perennial crops and shrublands, and the layers of information that
can spatially represent them may have been produced several years ago. The regional issues
mentioned in section 10.2 may also have caused changes to the landscape since the creation of
the land use map. The proposed conservation plan is therefore intended to guide conservation
actions toward those sites deemed most important for birds and species at risk, but a validation
of the current status of the proposed sites, both in terms of their nature and their spatial limits,
is essential.
It should also be noted that the sites of interest identified in the project aim to preserve the
habitats of priority migratory birds and critical habitats of species at risk. Sites known to harbour
species at risk for which critical habitats are not identified are not specifically documented in this
conservation plan (e.g. the American Water-willow, Green Dragon and Wood Turtle), and
documents that present specific conservation actions for these species exist for most of them
(federal: recovery strategies, action plans, management plans; provincial: conservation plans).
Records of these species, however, were considered in the prioritization of coarse filter habitat
patches. In addition, other sites of interest for the conservation of flora and wildlife exist in the
region, such as exceptional forest ecosystems and wildlife habitats (e.g. muskrat habitat, heron
colony) identified by the MRN and the MDDEFP, and efforts to conserve these sites should be
carried out in conjunction with those identified in this conservation plan.
54
12.0 STEP 9 − IMPLEMENTATION OF THE CONSERVATION
PLAN: APPROACHES AND PROPOSALS
The priority sites and conservation actions proposed in the conservation plan focus on various
types of habitats in the Lake Saint-Pierre region, including many wetlands. Protected areas were
established in the region essentially to protect significant wetlands located around Lake SaintPierre. Priority sites were also identified in terrestrial environments (forests, shrublands,
perennial crops) that are essential for the nesting and dispersal of nesting birds, but their
protection is currently inadequate.
Several laws, regulations and policies allow for the protection of natural elements (e.g. Canada
Wildlife Act, An Act Respecting Threatened or Vulnerable Species, An Act Respecting the
Conservation and Development of Wildlife), and others whose primary objective is not directly
protecting habitats have their place in this protection (e.g. the Cultural Property Act). Moreover,
land development and habitat protection issues are defined during land use planning. The RCMs
have the power to integrate the protection of the natural environment into their land use and
development planning as do municipalities in their urban planning. There are also many
voluntary conservation options available to owners of private property who wish to protect their
land (Longtin 1996; Queste 2011). Finally, funding programs exist to support stewardship and
habitat protection activities, made possible by non-governmental organizations and other local
stakeholders (Environment Canada 2012; ROBVQ 2012).
The conservation priorities identified in the conservation plan can thus serve as a foundation to
guide local actions through various funding programs such as the Habitat Stewardship Program
for Species at Risk (HSP) or the EcoAction Community Funding Program. They can also
complement existing action plans such as the Eastern Habitat Joint Venture (EHJV) and the
St. Lawrence Action Plan (SLAP). As such, layers of digital information (shapefiles) showing
the spatial location of priority sites identified in this pilot project are made available to the public
who wish to incorporate them in their own land-use planning process. It is by securing existing
tools and consulting stakeholders in the field that important natural habitats for migratory birds
and species at risk can be protected.
13.0 ADDITIONAL INFORMATION
13.1 OVERALL ASSESSMENT OF THE PROJECT
13.1.1 Advantages of the landscape approach
Habitat conservation has traditionally taken a species-by-species approach or been based on sites
known to be important for biodiversity (hot spots). However, these approaches have limitations
because they generally ignore the potential of sites that have not been inventoried and often
include small areas only. In addition, the connectivity between these sites is usually not
considered. The need to account for species dispersal is an important challenge that requires a
different analysis of the landscape. The advent of the landscape ecology theory, linked to the
evolution of associated theoretical ecology concepts (island biogeography, metapopulations, etc.)
provides a more holistic and dynamic perception of the landscapes. Moreover, new technologies
(e.g. GIS and landscape analysis software such as FRAGSTATS) and access to geospatial data
55
(e.g. satellite images) now facilitate the implementation of the landscape approach. Knowledge
of the importance of the spatial context for wildlife conservation and access to better tools to
address environmental or land use issues on a large scale provides opportunities to perceive
and analyze habitat conservation in a new way.
Determining minimum area thresholds to select priority habitat polygons has the advantage of
locating high-potential sites for certain bird species even though no targeted on-site survey has
been performed. As such, there is a strong agreement between priority woodlots and wetlands
identified by Gratton (2010), CRECQ (2012), and ours because these prioritization exercises
were all based on a multicriteria analysis that included patch area as a selection criterion. Patches
located far from the immediate Lake Saint-Pierre area were thus selected, whereas the vast
majority of priority sites identified in previous bird conservation plans in the BCR 13 were along
Lake Saint-Pierre (Chapdelaine and Rail 2004; Aubry and Cotter 2007; Environnement Canada
2010a, 2010b; Lepage et al. 2010). Adding the connectivity analysis between woodlots, habitat
classes not previously considered (perennial crops, old fields), and criteria aiming to prioritize
habitat patches based on their intrinsic value (e.g. shape, interior habitat) and their ecological
role in the landscape (e.g. buffer zones around critical habitat and protected areas, forest
corridor) clearly demonstrate the added value of the landscape approach used in this pilot
project.
The landscape approach therefore goes well beyond the simple consideration of important
habitats for species or guilds of species and helps determine the functionality of the landscape
(habitat availability, connectivity) while considering the importance of the surrounding
environment (matrix). The logic model developed during this study is based on the landscape
ecology theory and can therefore be applied to various spatial scales (ecozone, BCR, RCM, etc.).
These scales determine the level of accuracy of the information and data necessary to study the
landscape of a given territory. As such, several ongoing research initiatives in Quebec are based
on the concepts of landscape ecology (Connexion Montérégie 2012; Nature-Action 2012;
Université de Montréal 2012). The approach initiated as part of the pilot project and the lessons
learned should encourage the implementation of similar projects in other regions where
anthropogenic pressures may affect the landscape.
13.1.2
Disadvantages of the landscape approach
The landscape approach requires a more comprehensive and thus more complex process than
traditional approaches used in conservation. It is therefore important to portray the landscape as
realistically as possible. Geomatic tools greatly facilitate the implementation of this approach but
require advanced equipment (high performance computers, appropriate and functional softwares)
and specialized expertise (GIS specialists, cartographers). We estimate that the pilot project
required the contribution of a full-time biologist/analyst for about 1.75 year and a GIS specialist
for about 10 months. Note that the production of the land-use map of the study area by
combining data from different sources needed a large involvement by the geomatic people (see
section 13.2). The availability of resources to accomplish such a project can therefore become
an issue.
Landscape ecology also requires the adoption of a multidisciplinary perspective and involves the
use of data (e.g. socio-economic data) or specialized tools (planning or modeling software) that
56
are less familiar than traditional ecology approaches. Available data and limited knowledge of
an area can therefore restrict the application of the landscape approach in a given territory.
13.1.3
Involvement of other CWS units, partners, etc.
The expertise of several knowledgeable people from the government, non-governmental
organizations, universities and co-workers from the CWS was required at various stages of the
project (acquisition and sharing of data, validation of various decisions). Other experts in land
use planning or in modelling could also be involved in such an integrative approach (data,
predictive models, planning strategies, scenarios, etc.). An update of data and monitoring
of actions taken are essential, and regional and local stakeholders, experts, and partners in
conservation must take part in such a project because they have good knowledge of the terrain
and regional development issues (and possibly more adequate or recent data). As outlined in
the logic model, they should be involved in the early stage of the project in order to ensure
the project is understood by those in the field, its adequate execution and that it is properly
implemented (Leitão and Ahern 2002; Thompson 2011). In addition, an upstream involvement
from such a vast array of partners allows a common understanding of what landscape ecology
is and the benefits of such an approach.
13.1.4
Steps to complete
Some of the steps in the logic model were omitted intentionally or have not been fully completed
because this was a pilot project with a limited timeframe. For example, the data collected was
limited to biological, geographical and physical data. The contribution of socio-economic data
such as the demographic characteristics of the territory, forthcoming urban expansion and land
development (agriculture, industry, etc.), as well as "heritage" data, such as Aboriginal land
claims, would have clarified the human portrait of the study area and major regional issues.
Coupled with a predictive model, the development of different conservation scenarios would
help assess the impacts (positive and negative) of anthropogenic developments, such as projects
subject to environmental assessments on the structure, composition and integrity of the
landscape.
The identification of key areas where restoration efforts would be justified to favour biodiversity
(breeding habitats, functional corridors) was not completed, but this analysis would
bring an added value to the conservation plan. It would also be necessary to validate the
recommendations of the conservation plan. In fact, the land use map does not provide an updated
portrait of the current habitat distribution across the landscape, and it will be essential to check
its validity with local partners before implementing the conservation plan. A few guidelines
accompany the conservation plan, and this aspect could be further explored to better describe
the conservation options available to regional stakeholders and help with its implementation.
13.2
13.2.1
PROBLEMS ENCOUNTERED AND RECOMMENDATIONS
Data
The production of this conservation plan is dependent on information found in the BCR plans
regarding the selection of priority bird species and habitats. Some recommendations had to be
validated by CWS experts before they were integrated into the landscape analysis. In addition,
57
a landscape ecology study requires data coming from a variety of sources. Associated with this
wealth of data are data management issues. It is essential to document each completed step, both
in terms of the people contacted and data processing and conversions. Metadata files must be
completed systematically in order to ensure that the information is documented and retained,
while metadata from external files should be consulted carefully before integrating them in the
analyses. In fact, some data were not used because metadata were missing or incomplete. Also,
consider the delay that may occur between a request for data acquisition from partners and
reception of these data, as well as subsequent processing required to convert them into the
appropriate format for the project.
The quality of geospatial data is another obstacle, as their inclusion may be difficult considering
their variability (resolution, date, projection, etc.). The production of the land use map by
integrating data representing different thematics and coming from varied sources was laborious.
The use of a single land-use map (e.g. classified images from AAFC) would have greatly
reduced the resources needed to have this base map. On the other hand, the quality of the data
associated with each habitat class coming from a single source would have been reduced. In
addition, some data on the same topic had conflicting information (e.g. protected areas limits),
which required additional research to clarify the information. Finally, the data were kept in a
common data server accessible to all members of the work team, which created certain problems
in terms of accessibility and management of the file versions. Hence the importance of properly
documenting all steps performed. The presence of a data manager would have facilitated this
process while allowing other team members to focus on other steps of the project.
The lack of quality data to represent certain topics such as old fields and gravel and sand pits was
problematic. Other habitat features associated with priority species could not be represented
spatially and were not included in the spatial analysis, such as the location of adequate chimneys
to breeding Chimney Swifts.
13.2.2
Softwares
The choice of the appropriate softwares requires a good understanding of how data will be
handled and processed in order to ensure that the softwares are compatible with the project
objectives. This helps determine the required formats (raster or vector) in order to limit the
number of conversions and reduce the risk of errors. Our choice of using Corridor Designer and
FRAGSTATS was dictated by their compatibility with ArcGIS and the wide selection of spatial
analyses they offer. However, these softwares require an adequate knowledge of geomatics tools
and a good understanding of landscape metrics. Certain situations, such as changes in software
versions, can lead to incompatibilities when programming scripts, as was the case with the HPP
and FRAGSTATS software. The frequency of software updates and technical support offered
should therefore be considered when choosing processing tools. Finally, the availability of high
performance computers able to handle the software and the large size of geospatial files must be
considered in project planning. At a minimum, we recommend using ArcGIS 9.1 (or an earlier
version) with the Spatial Analyst extension, a computer with a 2.2 GHz processor, 2 GB RAM,
2.4 GB of free space, and a 256 MB RAM graphic card.
58
13.2.3
Teamwork
The magnitude of this project required rigorous teamwork and a great amount of cooperation
among team members. The roles of all members on such a multidisciplinary team (biology,
geomatics, geography, planning, etc.) should be defined based on the strengths and expertise
of each member. Day-to-day discussions between team members allowed everyone to gain
knowledge that goes beyond the scope of the project. Indeed, the synergy associated with the
multiple expertise involved in the project certainly contributed to a more rigorous product than
if the project had been accomplished by a reduced team. Meetings were held regularly to share
information, and meeting notes were produced to document decisions. These notes became
essential for monitoring the project. In addition, it often proved very useful to refer to the
purpose and objectives of the project in order not to wander into unnecessary analyses.
14.0 CONCLUSION
The Lake Saint-Pierre pilot project gave us the opportunity to develop and test a methodology to
determine priority habitats for nesting birds in a portion of the BCR 13. Based on the landscape
ecology theory, the approach developed is more dynamic and inclusive than traditional
approaches in conservation (i.e. by "hot spot" or species) because it allows to work on a larger
scale and to analyze the landscape as a whole, taking into account the various components
(biological, geographical/physical, socio-economic, etc.) that characterize it. Priority sites
identified are no longer considered as separate entities, but rather as different components
that are more or less interconnected and part of a whole.
Compared to traditional approaches, the landscape approach, more complex in terms of data
collection and analysis, requires the establishment of a team with varied skills and highlights
the essential contribution of geomatics and various planning and mapping tools. Although the
investment of time and money may be larger, the results of such a landscape approach allow for
more informed decisions regarding conservation on vast territories by identifying priority sites
that would be overlooked by the traditional “hot spot” approach. The benefits of this more
holistic landscape approach make the challenge of its practical implementation worthwhile.
The pilot project was an opportunity to define a phased approach in order to achieve a
comprehensive conservation plan in the Lake Saint-Pierre region, to determine a suite of action
to achieve conservation priorities in the field, and most importantly, to develop the know-how
required to achieve the habitat protection objectives of the conservation plans for migratory birds
(BCR plans) and the recovery programs for species at risk. Results from our analyses, namely
maps of priority sites and the associated digital layers (shapefiles), are made available to the
public and local stakeholders who wish to incorporate them in their own land-use planning
process (e.g. when RCM development plans are revised). We hope that this project will draw
attention to this type of integrated approach and will stimulate discussion on its applicability
to bird conservation plans developed for the BCRs or any other conservation initiative such as
the St. Lawrence Action Plan or the Eastern Habitat Joint Venture.
59
15.0 LITERATURE CITED
Ahern, J. 2006. Theories, methods and strategies for sustainable landscape planning. Pages
119-131, In: From Landscape Research to landscape Planning: Aspects of Integration,
Education and Application. Tress, B., G. Tress, G. Fry, et P. Opdam, Editeurs. Springer.
The Netherlands.
Andren, H. 1994. Effects of habitat fragmentation on birds and mammals in landscapes
with different proportions of suitable habitat: a review. Oikos 71: 355-366.
Aubry, Y. and R. Cotter. 2007. Quebec Shorebird Conservation Plan. Environment Canada,
Canadian Wildlife Service, Quebec Region, Sainte-Foy. xvi + 196 pp.
Austen, M.J.W., C.M. Francis, D.M. Burke and M.S. Bradstreet. 2001. Landscape context
and fragmentation effects on forest birds in southern Ontario. Condor 103: 704-714.
Baum, K.A., K.J. Haynes, F.P. Dillemuth and J.T. Cronin. 2004. The matrix enhances
the effectiveness of corridors and stepping stones. Ecology 85: 2671-2676.
Beier, P. and R.F. Noss. 1998. Do habitat corridors provide connectivity? Conservation Biology
12: 1241-1252.
Beier, P., D.R. Majka and W.D. Spencer. 2008. Forks in the road : Choices in procedures
for designing wildland linkages. Conservation Biology 22: 836-851.
Bélanger, L. and M. Grenier. 1998. Importance et causes de la fragmentation forestière dans
les agroécosystèmes du sud du Québec. Série de rapports techniques no 327, Service
canadien de la faune, Environnement Canada, région du Québec, 39 p.
Bélanger, L. and M. Grenier. 2002. Agriculture intensification and forest fragmentation
in the St. Lawrence valley, Québec, Canada. Landscape Ecology 17: 495–507.
Bélanger, L. and M. Grenier. 2003. Atlas de conservation des terres humides. Environnement
Canada, Sevice canadien de la faune, région du Québec.
Bélanger, L., B. Jobin, G. Lacroix and Y. Bédard. 2006. Utilisation par l’avifaune des emprises
autoroutières du sud du Québec faisant l’objet d’une gestion écologique de la végétation.
Série de rapports techniques no 388, Service canadien de la faune, région du Québec,
Environnement Canada, Sainte-Foy, xii + 89 p. et annexes.
Bélisle, M. and A. Desrochers. 2002. Gap-crossing decisions by forest birds : an empirical basis
for parameterizing spatially-explicit, individual-based models. Landscape Ecology 17:
219-231.
Bennett, A.F. 1999. Linkages in the landscape : the role of corridors and connectivity in wildlife
conservation. IUCN, Gland, Switzerland and Cambridge, U.K.
Bentrup, G. 2008. Conservation buffers: design guidelines for buffers, corridors, and greenways.
Gen. Tech. Rep. SRS-109. Asheville, NC: Department of Agriculture, Forest Service,
Southern Research Station. 110 p. Site: http://nac.unl.edu/bufferguidelines/index.html
60
Brown, C. R. 1997. Purple Martin (Progne subis), The Birds of North America Online
(A. Poole, Ed.). Cornell Lab of Ornithology; Ithaca, USA.
Site: http://bna.birds.cornell.edu/bna/species/287
Brown, M. and J.J. Dinsmore. 1986. Implications of marsh size and isolation for marsh bird
management. Journal of Wildlife Management 50: 392-397.
Calmé, S. 1998. Les patrons de distribution des oiseaux des tourbières du Québec méridional.
Thèse de Doctorat. Département des Sciences du bois et de la forêt. Faculté de Foresterie
et de Géomatique, Université Laval, Québec. 93 p.
Chapdelaine, G. and J.-F. Rail. 2004. Quebec’s Waterbird Conservation Plan. Migratory Bird
Division, Canadian Wildlife Service, Quebec Region, Environment Canada, Sainte-Foy,
Quebec. 99 pp.
Connexion Montérégie. 2012. Biodiversity, Ecosystem Services, and Connectivity
in Montérégie. Site: www.connexionmonteregie.com/index.html
Corry, R.C. 2004. Characterizing fine-scale patterns of alternative agricultural landscapes
with landscape pattern indices. Landscape Ecology 20: 591-608.
CRECQ. 2012. Portrait des milieux humides du Centre-du-Québec. Conseil régional
de l’environnement du Centre-du-Québec, 135 p.
Debinski, D.M. 2006. Forest fragmentation and matrix effects : the matrix does matter. Journal
of Biogeography 33: 1791-1792.
De Smith, M., M. Goodchild and P. Longley. 2011. Geospatial Analysis – A Comprehensive
Guide, 3rd edition. Site: www.spatialanalysisonline.com/output/
Desrochers, A. and S.J. Hannon. 1997. Gap crossing decisions by forest songbirds during
the post-fledging period. Conservation Biology 11: 1204-1210.
Detenbeck, N., S.M. Galatowitsch, J. Atkinson and H. Ball. 1999. Evaluating perturbations
and developing restoration strategies for inland wetlands in the Great Lakes Basin.
Wetlands 19: 789-829.
Dettmers, R. 2003. Status and conservation of shrubland birds in the northeast US. Forest
Ecology and Management 185: 81-93.
Duchemin, M. and R. Majdoub. 2004. Les bandes végétales filtrantes : de la parcelle au bassin
versant. Vecteur Environnement 37: 36-50.
Duchemin, M. and R. Hogue. 2009. Reduction in agricultural non-point source pollution
in the first year following establishment of an integrated grass/tree filter strip system
in southern Quebec (Canada). Agriculture, Ecosystems and Environment 131: 85-97.
Duchesne, S. and L. Bélanger. 1997. Fragmentation forestière et corridors verts en paysage
agricole : 1. Revue des principales normes de conservation. Série de rapports techniques
no 288, Service canadien de la faune, région du Québec, Environnement Canada, SainteFoy, 68 p.
61
Duchesne, S., L. Bélanger and J.-P.L. Savard. 1998. Fragmentation forestière et corridors verts
en paysage agricole : 2. Effets de bordure et de discontinuité des boisés. Série de rapports
techniques no 318, Service canadien de la faune, région du Québec, Environnement
Canada, Sainte-Foy, 70 p.
Duchesne, S., L. Bélanger, M. Grenier and F. Hone. 1999. Guide de conservation des corridors
forestiers en milieu agricole. Environnement Canada, Service canadien de la faune,
région du Québec et Fondation Les oiseleurs du Québec inc., 60 p.
Environment Canada. 2004. How Much Habitat is Enough? A Framework for Guiding Habitat
Rehabilitation in Great Lakes Areas of Concern (Second Edition). Environment Canada,
Canadian Wildlife Service, Ontario Region. Downsview, Ontario. 80 p.
Environment Canada. 2007. Area-Sensitive Forest Birds in Urban Areas. Environment Canada,
Canadian Wildlife Service, Ontario Region, Toronto. 58 p.
Environnement Canada. 2010a. Plan de conservation des oiseaux terrestres du Québec :
volume 1, oiseaux des milieux forestiers. Environnement Canada, Service canadien
de la faune, région du Québec. Rapport préliminaire inédit.
Environnement Canada. 2010b. Plan de conservation des oiseaux terrestres du Québec :
volume 2, oiseaux des milieux agricoles, humides, urbains, arctiques, alpins et des
falaises. Environnement Canada, Service canadien de la faune, région du Québec.
Rapport préliminaire inédit.
Environment Canada. 2011. Recovery Strategy for the Least Bittern (Ixobrychus exilis)
in Canada [Proposed]. Species at Risk Act Recovery Strategy Series. Environment
Canada. Ottawa. v + 34 pp.
Site: www.registrelep.gc.ca/document/default_e.cfm?documentID=1291
Environment Canada 2012. Funding.
Site: www.ec.gc.ca/financement-funding/Default.asp?lang=En&n=3A49CF19-1
Fahrig, L. 1997. Relative effects of habitat loss and fragmentation on population extinction.
Journal of Wildlife Management 61: 603-610.
Fahrig, L. 2003. Effects of habitat fragmentation on biodiversity. Annual Review of Ecology,
Evolution, and Systematics 34: 487-515.
Fairbairn, S.E. and J.J. Dinsmore. 2001. Local and landscape-level influences on wetland bird
communities of the Prairie Pothole Region of Iowa, USA. Wetlands 21: 41-47.
Fischer, R.A. 2000. Width of riparian zones for birds. EMRRP Technical Notes Collection
(TN EMRRP-SI-09), U.S. Army Engineer Research and Development Center, Vicksburg,
MS.
Fondation de la faune du Québec and Union des producteurs agricoles. 2011. Manuel
d’accompagnement pour la mise en valeur de la biodiversité des cours d’eau en milieu
agricole. 122 p.
62
Forcey, G.M., W.E. Thogmartin, G.M. Linz, W.J. Bleier and P.C. McKann. 2011. Land use and
climate influences on waterbirds in the Prairie Potholes. Journal of Biogeography 38:
1694-1707.
Forman, R.T.T. 1995. Land mosaics: the ecology of landscapes and regions. Cambridge
University Press, Cambridge, England.
Fournier, F., B. Audet and S. Paradis. 2010. Bird conservation plan for the Lower Great
Lakes/St. Lawrence Plain Bird Conservation region (BCR 13) in Quebec. Draft Report –
Unpublished. Canadian Wildlife Service, Environment Canada. Québec, Quebec.
Gagnon, E., and G. Gangbazo. 2007. Efficacité des bandes riveraines : analyse de la
documentation scientifique et perspectives. Ministère du Développement durable,
de l’Environnement et des Parcs, Direction des politiques de l’eau, Québec. 17 p.
García-Feced, C., S. Saura and R. Elena-Rosselló. 2011. Improving landscape connectivity
in forest districts: A two-stage process for prioritizing agricultural patches for
reforestation. Forest Ecology and Management 261: 154-161.
Gauthier, J. and Y. Aubry. 1996. The breeding birds of Quebec: atlas of the breeding birds
of southern Québec. Association québécoise des groupes d'ornithologues, Province
of Quebec Society for the Protection of Birds, Canadian Wildlife Service. Quebec
Region, Montréal, Quebec.
GéoMont. 2008. Cartographie de base des milieux humides de la Montérégie. GéoMont
et Canards Illimités.
Site: http://foliogis.ducks.ca/qc/fr/monteregie/reg16_rapport_avril08.pdf
GéoMont and Environnement Canada. 2008. Évaluation des changements des terres humides
entre 1964 et 2006 – Secteur de la Montérégie. Agence géomatique Montérégienne
(GéoMont) et Environnement Canada, Service canadien de la faune. Rapport de projet.
45 p.
Gratton, L. 2010. Plan de conservation de la vallée du Saint-Laurent et du lac Champlain, région
du Québec. Conservation de la Nature Canada, Montréal, Quebec. 150 p.
Grenier, M., Demers, A.-M., Labrecque, S., Benoit, M., Fournier, R.A., and Drolet, B. 2007.
An object-based method to map wetland using RADARSAT-1 and Landsat ETM
images : test case on two sites in Québec, Canada. Canadian Journal of Remote Sensing,
vol. 33, suppl. 1, pp. S28-S45.
Gustafson, E.J. 1998. Quantifying landscape spatial pattern: what is the state of the art?
Ecosystems 1: 143-156.
Hamer, T.L., C.H. Flather and B.R. Noon. 2006. Factors associated with grassland bird species
richness : the relative roles of grassland area, landscape structure, and prey. Landscape
Ecology 21: 569-583.
Hargis, C.D., J.A. Bissonette and J.L. David. 1998. The behavior of landscape metrics
commonly used in the study of habitat fragmentation. Landscape Ecology 13: 167-186.
63
Hodges Jr., M.F. and D.G. Krementz. 1996. Neotropical migratory breeding bird communities
in riparian forests of different widths along the Altamaha River, Georgia. Wilson
Bulletin 108: 496-506.
Holzmueller, E.J., M.D. Gaskins and J.C. Mangun. 2011. A GIS approach to prioritizing
habitat for restoration using neotropical migrant songbird criteria. Environmental
Management 48: 150-157.
Huber, P.R., J.H. Thorne, N.E. Roth and M.M. McCoy. 2011. Assessing, ecological condition,
vulnerability, and restorability of a conservation network under alternative urban growth
policies. Natural Areas Journal 31: 234-245.
Institut de la statistique du Québec. 2010. Répertoire des établissements miniers du Québec –
Liste des établissements producteurs pour certaines régions et substances. 39 p.
Also available at: http://diff1.stat.gouv.qc.ca/rem/classes/accueil
Jaeger, J.A.G. 2000. Landscape division, splitting index, and effective mesh size: new measures
of landscape fragmentation. Landscape Ecology 15: 115-130.
Jean, M. and G. Létourneau. 2011. Changes to the wetlands of the St. Lawrence River from 1970
to 2002. Environment Canada, Science and Technology Branch, Quebec Water Quality
Monitoring and Surveillance Section, Technical Report No. 511, 293 p.
Jobin, B. and G. Falardeau. 2010. Habitat Associations of Grasshopper Sparrows in Southern
Québec. Northeastern Naturalist 17: 135–146.
Jobin, B., M. Grenier and P. Laporte. 2001. Disponibilité de l'habitat de nidification préférentiel
de la Pie-grièche migratrice au Québec. Série de rapports techniques no 377,
Environnement Canada, Service canadien de la faune, région du Québec, Sainte-Foy.
xii + 65 p. + 2 annexes.
Jobin, B., C. Latendresse, C. Maisonneuve, A. Sebbane and M. Grenier. 2007. Changements
de l’occupation du sol dans le sud du Québec pour la période 1993-2001. Série
de rapports techniques no 483, Environnement Canada, Service canadien de la faune,
région du Québec, Sainte-Foy, Québec. 112 p. + annexes.
Jobin B., R. Langevin, M. Allard, S. Labrecque, D. Dauphin, M. Benoit and P. Aquin. 2013.
Évaluation d’une approche d’analyse du paysage pour planifier la conservation
des habitats des oiseaux migrateurs et des espèces en péril dans l’écozone des Plaines
à forêts mixtes : étude de cas au lac Saint-Pierre – Rapport méthodologique.
Environnement Canada, Service canadien de la faune, région du Québec, Québec.
Rapport non publié. 196 p. et annexes.
Kampf, H. and F. Stavast. 2005. Committee of experts for the development of the Pan-European
ecological network: report on the implementation of the Pan-European ecological
network. The Netherlands. Ministry of Agriculture, Nature and Food Quality: Ede, NL.
Site: www.minlnv.nl/txmpub/files/?p_file_id=14151
Keller, C.M.E., C.S. Robbins and J.S. Hatfield. 1993. Avian communities in riparian forests
of different widths in Maryland and Delaware. Wetlands 13: 137-144.
64
Langevin, R. 1997. Guide de conservation des boisés en milieu agricole. Service canadien
de la faune, Environnement Canada, Sainte-Foy. 77 p.
Langevin, R. and L. Bélanger. 1994. Conservation des îlots boisés en paysage agricole. I.Revue de littérature et synthèse des connaissances sur leur utilisation par l'avifaune.
Série de rapports techniques no 221. Environnement Canada, Service canadien de
la faune, région du Québec. 87 p.
Langevin, R. and L. Bélanger. 1995. Conservation des îlots boisés en paysage agricole. II.Choix des critères et grille d'évaluation à des fins de conservation : Série de rapports
techniques no 224. Environnement Canada, Service canadien de la faune, région du
Québec. 44 p.
Latendresse, C., B. Jobin, A. Baril, C. Maisonneuve, C. Boutin and D. Côté. 2008. Dynamique
spatio-temporelle des habitats fauniques dans l’écorégion des Basses terres du fleuve
Saint-Laurent, 1950-1997. Série de rapports techniques no 494, Environnement Canada,
Service canadien de la faune, région du Québec, Québec, 83 p. + annexes.
Lee, K.H., T.M. Isenhart and R.C. Schultz. 2003. Sediment and nutrient removal in an
established multi-species riparian buffer. Journal of Soil and Water Conservation 58: 1-8.
Leitão, A.B. and J. Ahern. 2002. Applying landscape ecological concepts and metrics in
sustainable landscape planning. Landscape and Urban Planning 59: 65-93.
Lepage, C., D. Bordage, D. Dauphin, F. Bolduc and B. Audet. 2010. Plan de conservation de
la sauvagine du Québec – 2010. Environnement Canada, Service canadien de la faune,
région du Québec. Rapport préliminaire inédit.
Longtin, B. 1996. Options de conservation : Guide du propriétaire. Centre québécois du droit
de l’environnement. Montréal, Quebec. 100 p.
Lovell, S.T. and W.C. Sullivan. 2006. Environmental benefits of conservation buffers in the
United States: evidence, promise, and open questions. Agriculture, Ecosystems and
Environment 112: 249-260.
Maheu-Giroux M., S. de Blois and B. Jobin. 2006. Dynamique des paysages de quatre Réserves
nationales de faune du Québec : Suivi des habitats et des pressions périphériques.
Université McGill, Département de sciences végétales et Environnement Canada, Service
canadien de la faune, région du Québec. 67 p. + annexes.
Maheu-Giroux, M. and J. Belvisi. 2007. Volet biodiversité de l’Initiative nationale d’élaboration
de normes agroenvironnementales – Projet pilote de la portion québécoise de l’écorégion
des basses-terres du Saint-Laurent – Phase 1. Série technique de l'Initiative nationale
d'élaboration de normes agroenvironnementales rapport no 3-16. 243 p.
Majka, D., P. Beier and J. Jenness. 2007. CorridorDesigner ArcGIS Toolbox Tutorial,
Environmental Research, Development and Education for the New Economy (ERDENE)
initiative from Northern Arizona University, 25 p.
Mason, J., C. Moorman, G. Hess and K. Sinclair. 2007. Designing suburban greenways to
provide habitat for forest-breeding birds. Landscape and Urban Planning 80: 153-164.
65
McGarigal, K., S.A. Cushman, M.C. Neel and E. Ene. 2002. FRAGSTATS v3: Spatial Pattern
Analysis Program for Categorical Maps. Computer software program produced by the
authors at the University of Massachusetts, Amherst.
Site: www.umass.edu/landeco/research/fragstats/fragstats.html
McGarigal, K., B.W. Compton, S.D. Jackson, K. Rolih and E. Ene. 2005. Conservation
Assessment and Prioritization System (CAPS) Highlands Communities Initiative,
PHASE 1. Final Report. 22 p. Site: www.umass.edu/landeco/research/caps/caps.html
McPherson, M., C. Nielsen, and K. Proudlock. 2009. The Development of Tier 1 Generalized
Habitat-based Standards for Ecozones in Agricultural Regions of Canada. National
Agri-Environmental Standards Initiative Synthesis Report No. 3. Environment Canada.
Gatineau, Quebec. 137 p.
Ménard, S., M. Darveau, L. Imbeau and L. V. Lemelin. 2006. Méthode de classification des
milieux humides du Québec boréal à partir de la carte écoforestière du 3e inventaire
décennal. Québec, Canards Illimités Canada.
Ministère des Ressources naturelles et de la Faune du Québec (MRNF). 2012a. Registre foncier
du Québec. Site: www.registrefoncier.gouv.qc.ca/Sirf
Ministère des Ressources naturelles et de la Faune du Québec (MRNF). 2012b. Système
d’information géoscientifique pétrolier et gazier. Cartes des permis.
Site: http://sigpeg.mrnf.gouv.qc.ca/gpg/classes/igpg
Ministère du Développement durable, de l’Environnement, de la Faune et des Parcs (MDDEFP).
2002. Aires protégées.
Site: www.mddefp.gouv.qc.ca/biodiversite/aires_protegees/index.htm
Minor, E. and D. Urban. 2010. Forest bird communities across a gradient of urban development.
Urban Ecosystems 13: 51–71.
Municonsult. 2002. Réserve de la biosphère du lac Saint-Pierre, Habitats, ressources fauniques
et exploitation, 33 p.
NABCI International. 2012. Background, Vision, and Strategy: NABCI Declaration of Intent.
North American Bird Conservation Initiative.
Site: www.nabci.net/International/English/background.html
Nature-Action. 2009. Plan de conservation et de mise en valeur des boisés de la municipalité
régionale de comté de Pierre-De Saurel. Pour la MRC de Pierre-De Saurel. Rapport
final. 39 p.
Nature-Action. 2012. Nature-Action Québec. Les milieux naturels.
Site: www.nature-action.qc.ca/site/realisations/milieux-naturels
Naugle, D.E., R.R. Johnson, M.E. Estey and K.F. Higgins. 2000. A landscape approach to
conserving wetland bird habitat in the prairie pothole region of eastern South Dakota.
Wetlands 20 : 588-604.
66
Neave, E., D. Baldwin, and C. Nielsen. 2009. Tier 2 and 3 Standards – Developing Landscapespecific, Habitat-based Standards Using Multiple Lines of Evidence. National AgriEnvironmental Standards Initiative Synthesis Report No. 4. Environment Canada.
Gatineau, Quebec. 138 p.
North American Bird Conservation Initiative Canada. 2012. The State of Canada’s Birds, 2012.
Environment Canada, Ottawa, Canada. 36 p.
Oliver, A.J., C. Hong-Wa, J. Devonshire, K.R. Olea, G.F. Rivas and M.K. Gahl. 2011. Avifauna
richness enhanced in large, isolated urban parks. Landscape and Urban Planning 102:
215-225.
Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA). 2004. Best Management
Practices: Buffer Strips. Ontario Cattlemen's Association, Canada, Agriculture and AgriFood Canada, Ontario. Ministry of Agriculture, Food and Rural Affairs. 141 p.
Poirier, I. 2010. La canneberge au Québec et dans le Centre-du-Québec : Un modèle de
développement durable, à la conquête de nouveaux marchés. Ministère de l’Agriculture,
des Pêcheries et de l’Alimentation, Direction régionale du Centre-du-Québec,
Victoriaville, Quebec. 35 p.
Poulin, M., M. Bélisle and M. Cabeza. 2006. Within-site habitat configuration in reserve design:
A case study with a peatland bird. Biological Conservation 128: 55-66.
Qiu, Z. 2010. Prioritizing agricultural lands for conservation buffer placement using multiple
criteria. Journal of the American Water Resources Association 46: 944-956.
Queste, C., 2011. Les milieux humides dans le sud du Québec : entre destruction et protection.
Analyse critique et élaboration d’une stratégie de conservation. Rapport de stage présenté
à Nature Québec, à l’Université du Littoral Côte d’Opale, et à l’Université des Sciences
et Technologies de Lille 1, Nature Québec, Quebec, 44 p.
Rail, J.-F., M. Darveau, A. Desrochers and J. Huot. 1997. Territorial responses of boreal forest
birds to habitat gaps. Condor 99: 976-980.
Renfrew, R.B. and C.C. Ribic. 2008. Multi-scale models of grassland passerine abundance
in a fragmented system in Wisconsin. Landscape Ecology 23: 181-193.
Renfrew, R.B., C.A. Ribic and J.L. Nack. 2005. Edge avoidance by nesting grassland birds:
a futile strategy in a fragmented landscape. Auk 122: 618-636.
Richard, G., D. Côté, M. Mingelbier, B. Jobin, J. Morin and P. Brodeur. 2011. Utilisation
du sol dans la plaine inondable du lac Saint-Pierre (fleuve Saint-Laurent) durant
les périodes 1950, 1964 et 1997 : interprétation de photos aériennes, numérisation
et préparation d’une base de données géoréférencées. Rapport technique préparé pour
le ministère des Ressources naturelles et de la Faune et Environnement Canada. 42 p.
Riffell, S. K., B. E. Keas and T. M. Burton. 2003. Birds in North American Great Lakes coastal
wet meadows: is landscape context important? Landscape Ecology 18: 95-111.
67
ROBVQ. 2012. Recherche de financement : Document d’orientation à l’intention des OBV.
Regroupement des organismes de bassins versants du Québec. 28 p.
Rompré, G., Y. Boucher, L. Bélanger, S. Côté and D. Robinson. 2010. Conservation de la
biodiversité dans les paysages forestiers aménagés : utilisation des seuils critiques
d’habitat. The Forestry Chronicle 86: 572-579.
Roy, L. 2002. Les impacts environnementaux de l’agriculture sur le Saint-Laurent.
Le Naturaliste canadien 126: 67-77.
Saunders, D.A., R.J. Hobbs and C.R. Margules. 1991. Biological consequences of ecosystem
fragmentation: a review. Conservation Biology 5: 18-32.
Savoie, C. 2002. Le phénomène de déboisement : Évaluation par télédétection entre le début des
années 1990 et 1999-2000. Région Le Centre-du-Québec. Direction de l’environnement
et du développement durable, Ministère de l’Agriculture, des Pêcheries et de
l’Alimentation du Québec. 24 p.
Schlossberg, S. and D.I. King. 2008. Are shrubland birds edge specialists? Ecological
Applications 18: 1325-1330.
Schwenk, W.S. and T.M. Donovan. 2011. A multispecies framework for landscape conservation
planning. Conservation Biology 25: 1010-1021.
Shanahan, D.F., C. Miller, H.P. Possingham and R.A. Fuller. 2011. The influence of patch
area and connectivity on avian communities in urban revegetation. Biological
Conservation 144: 722-729.
Shao, G., D. Liu and G. Zhao. 2001. Relationships of image classification accuracy and variation
of landscape statistics. Canadian Journal of Remote Sensing 27: 33-43.
Spackman, S.C. and J.W. Hughes. 1995. Assessment of minimum stream corridor width
for biological conservation: species richness and distribution along mid-order streams
in Vermont, USA. Biological Conservation 71: 325-332.
Stauffer, D.F. and L.B. Best. 1980. Habitat selection by birds of riparian communities:
evaluating effects of habitat alterations. Journal of Wildlife Management 44: 1-15.
Steedman, R.J. 1987. Comparative analysis of stream degradation and rehabilitation
in the Toronto area. Ph.D. Thesis, University of Toronto. 172 p.
Stephens, S.E., D.N. Koons, J.J. Rotella and D.W. Willey. 2003. Effects of habitat fragmentation
on avian nesting success: a review of the evidence at multiple spatial scales. Biological
Conservation 115: 101–110.
Tate, D.P. 1998. Assessment of the biological integrity of forest bird communities – a Draft
Methodology and Field Test in the Severn Sound Area of Concern. Severn Sound RAP
Technical Report. Canadian Wildlife Service, Ontario Region.
Thompson, B.A. 2011. Planning for implementation: Landscape-level restoration planning
in an agricultural setting. Restoration Ecology 19: 5-13.
68
Tozer, D.C., E. Nol and K.F. Abraham. 2010. Effects of local and landscape-scale habitat
variables on abundance and reproductive success of wetland birds. Wetlands Ecology
and Management 18: 679-693.
Trani, M.K. and R.H. Giles Jr. 1999. An analysis of deforestation: metrics used to describe
pattern change. Forest Ecology and Management 114: 459-470.
Turner, M.G., R.H. Gardner and R.V. O'Neill. 2001. Landscape ecology in theory and practice:
pattern and process. Springer, New York. 404 p.
Uezu, A. and J.P. Metzger. 2011. Vanishing bird species in the Atlantic Forest: relative
importance of landscape configuration, forest structure and species characteristics.
Biodiversity and Conservation 20: 3627-3643.
Université de Montréal. 2012. Chaire en paysage et environnement.
Site: www.paysage.umontreal.ca/
Van der Sluis, T., M. Bloemmen and I.M. Bouwma. 2004. European corridors: Strategies for
corridor development for target species. ECNC, Tilburg, the Netherlands and Alterra.
The Netherlands. 32 p.
Veech, J.A. 2006. A comparison of landscapes occupied by increasing and decreasing
populations of grassland birds. Conservation Biology 20: 1422-1432.
Villard, M.A., M.K. Trzcinski and G. Merriam. 1999. Fragmentation effects on forest birds:
Relative influence of woodland cover configuration on landscape occupancy.
Conservation Biology 13: 774-783.
Watling, J.I., A.J. Nowakowski, M.A. Donnelly and J.L. Orrock. 2011. Meta-analysis reveals
the importance of matrix composition for animals in fragmented habitat. Global Ecology
and Biogeography 20: 209-217.
69
APPENDIX 1 − LIST OF 19 BIRD SPECIES NOT SELECTED
FOR THE PILOT PROJECT
1
*
English name
Latin name
Group1
Reason
Bald Eagle
Haliaeetus
leucocephalus
Land.
No known nesting site in the study area
Cerulean Warbler*
Dendroica cerulea
Land.
No known nesting site in the study area
Eastern Towhee
Pipilo
erythrophthalmus
Land.
Rare species in the region
Field Sparrow
Spizella pusilla
Land.
Rare species in the region
Golden-winged Warbler*
Vermivora
chrysoptera
Land.
No known nesting site in the study area
Grasshopper Sparrow
Ammodramus
savannarum
Land.
No known nesting site in the study area
Loggerhead Shrike*
Lanius ludovicianus
Land.
No known nesting site in the study area
Olive-sided Flycatcher*
Contopus cooperi
Land.
Rare species in the region
Willow Flycatcher
Empidonax traillii
Land.
No conservation issue in BCR 13 in Québec
Red-headed Woodpecker*
Melanerpes
erythrocephalus
Land.
Rare species in the region
Rose-breasted Grosbeak
Pheucticus
ludovicianus
Land.
Stewardship species
Yellow-throated Vireo
Vireo flavifrons
Land.
Rare species in the region
Common Loon
Gavia immer
Mar.
Stewardship species
Yellow Rail*
Coturnicops
noveboracensis
Mar.
Rare species in the region
American Black Duck
Anas rubripes
Wat.
Stewardship species
Brant
Branta bernicla
Wat.
Rare species in the region
Canada Goose (Atlantic Flyway
Resident)
Branta canadensis
maxima
Wat.
Overabundant species
Canada Goose (Atlantic))
Branta canadensis
canadensis
Wat.
Stewardship species
Snow Goose
Chen caerulescens
atlantica
Wat.
Overabundant species
Land. = Landbirds; Mar. = Marshbirds/Waterbirds; Wat. = Waterfowl
Species with an asterisk are listed at risk according to SARA or COSEWIC.
APPENDIX 2 − SPECIES ASSOCIATED WITH COARSE FILTER
HABITAT CLASSES
Priority habitat classes
Bog
Wet meadow
Shrub swamp
Marsh
Forest swamp
Wetlands
Forest
Decid, Mixt, Conif
English name
American Kestrel
Old field/Shrubland
Group
Landbirds
Perennial crop
Farmland
X
Baltimore Oriole
X
Bank Swallow
Barn Swallow
X
Barred Owl
X
Belted Kingfisher
Black-billed Cuckoo
Bobolink
X
X
X
Brown Creeper
X
Brown Thrasher
X
X
Canada Warbler
X
Chimney Swift
X
X
X
Common Nighthawk
Eastern Kingbird
Eastern Meadowlark
X
Eastern Screech-Owl
X
X
Long-eared Owl
Nelson's Sparrow
X
X
Northern Flicker
Northern Harrier
X
X
Eastern Wood-Pewee
Horned Lark
X
X
X
X
X
X
X
X
X
Northern Rough-winged Swallow
Northern Saw-whet Owl
X
Palm Warbler
X
Peregrine Falcon (anatum)
Purple Martin
Savannah Sparrow
X
X
X
X
X
X
X
X
X
Wilson's Phalarope
X
X
Sedge Wren
Short-eared Owl
X
Vesper Sparrow
X
Whip-poor-will
X
Wood Thrush
Shorebirds
X
American Woodcock
X
Killdeer
X
Upland Sandpiper
X
X
Wilson's Snipe
X
Marshbirds/
American Bittern
X
X
X
Waterbirds
Black Tern
X
X
Least Bittern
X
X
Sora
X
X
Common Tern
Virginia Rail
Waterfowl
Blue-winged Teal
X
X
X
X
X
Greater Scaup
Lesser Scaup
Wood Duck
X
Species not associated with a habitat class are considered in the fine filter criteria.
X
APPENDIX 3 − MATRICES USED TO CALCULATE THE EDGE CONTRAST INDEX
OF HABITAT PATCHES FOR GENERAL AND DETAILED LAND USE CLASSES
General classes
Open water
0.00
0.50
1.00
1.00
1.00
1.00
1.00
Open water
Wetland
Anthropogenic
Annual crop
Perennial crop
Old field
Forest
Wetland
Anthropogenic
Annual crop
Perennial crop
Old field
Forest
0.00
1.00
0.75
0.75
0.50
0.50
0.00
0.75
0.75
1.00
1.00
0.00
0.25
0.75
0.75
0.00
0.50
0.75
0.00
0.25
0.00
Anthropogeni
c
Anthropogeni
c – Other
Highway
ROW
Bare soil
0.00
0.50
0.50
0.00
Unidentified wetland
0.50
0.50
0.50
0.00
Unidentified swamp
0.50
0.50
0.75
0.50
0.50
0.50
0.75
0.50
0.50
0.25
0.50
0.25
0.50
0.50
0.50
0.50
0.00
0.00
0.25
0.50
0.00
0.25
0.25
0.00
0.50
0.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.00
0.00
0.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.25
0.25
0.25
0.25
0.00
0.25
0.00
0.75
0.75
0.50
0.00
0.75
0.75
0.75
0.75
0.25
0.00
Wet meadow
1.00
1.00
0.50
0.50
0.50
0.50
0.75
0.25
1.00
1.00
0.75
0.75
0.50
0.25
0.00
Old field/Shrubland
1.00
1.00
1.00
1.00
0.50
0.75
0.50
0.75
0.50
0.75
0.25
0.75
0.50
0.75
0.25
0.75
1.00
1.00
1.00
1.00
0.75
1.00
0.75
1.00
0.75
0.75
0.50
0.50
0.50
0.50
0.00
0.50
0.00
1.00
1.00
1.00
0.75
0.75
0.50
0.75
0.75
1.00
1.00
1.00
1.00
0.75
0.75
0.75
0.25
0.50
0.00
Forest – Deciduous
1.00
1.00
1.00
0.75
0.50
0.75
0.25
0.50
1.00
1.00
1.00
1.00
0.75
0.75
0.75
0.25
0.50
0.75
0.00
Forest – Mixed
1.00
1.00
1.00
0.75
0.50
0.75
0.25
0.50
1.00
1.00
1.00
1.00
0.75
0.75
0.75
0.25
0.50
0.75
0.00
0.00
Forest – Coniferous
1.00
1.00
1.00
0.75
0.50
0.75
0.25
0.50
1.00
1.00
1.00
1.00
0.75
0.75
0.75
0.25
0.50
0.75
0.00
0.00
Orchard
Forest disturbance
Forest –
Mixed
0.75
0.50
Forest –
Deciduous
0.75
0.75
Forest
disturbance
0.75
0.75
Perennial
crop
0.75
0.75
Annual crop
0.75
0.75
Unidentified
wetland
0.75
0.75
Bare soil
Annual crop
Marsh
0.75
1.00
Anthropo-genic – Other
Highway ROW
Shallow
water
1.00
1.00
Open water
1.00
Perennial crop
Shrub swamp
Forest swamp
Bog
Anthropo-genic
Forest –
Coniferous
Bog
0.00
Marsh
Orchard
Forest
swamp
0.00
Old
field/Shrubla
nd
Shrub swamp
Open water
Shallow water
Wet meadow
Unidentified
swamp
Detailed classes
0.00
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF

advertisement