Primary databases for forest ecosystem management-examples from Ontario and possibilities for Canada: NatGRID

P R I M A R Y DATABASES F O R FOREST E C O S Y S T E M M A N A G E M E N T E X A M P L E S F R O M ONTARIO AND POSSIBILITIES F O R CANADA: NatGRID

DANIEL W. McKENNEY Canadian Forest Service - Ontario, Natural Resources Canada, P.O. Box 490, 1219 Queen Street East, Sault Ste. Marie, Ontario, P6A 5M7, Canada

BRENDAN G. MACKEY Department of Geography, Faculty of Science, The Australian National University, Canberra, ACT, 0200, Australia

and RICHARD A. SIMS Canadian Forest Service - Ontario, Natural Resources Canada, P.O. Box 490, 1219 Queen Street East, Sault Ste. Marie, Ontario, P6A 5M7, Canada

Abstract. This paper identifies some scientific impediments to ecosystem management and describes bio-physical databases required to help systematically and empirically address the ecological sustainability challenge. Examples are drawn from ongoing work in Ontario. This work has implications for efforts in ecological land classification, landscape ecology, more efficient locating of research and monitoring plots, wildlife management and ultimately trade-off analyses. We conclude with the recommendation that the key primary databases, as currently evolving for Ontario, could and should be developed nationally, thereby creating a "NatGRID database", i.e., Nationally Georeferenced Resource Information for Decision-making. NatGRID could be used to help address, in a more quantitative manner, fundamental questions regarding ecological sustainability and trade-offs in forest management.

1. Introduction

The pursuit of ecological sustainability has generally spurred on much activity in forestry circles in Canada. In the late 1980s the federal government established the "Green Plan", an initiative designed to develop a more comprehensive approach to environmental decision making (Government of Canada, 1990). In the Canadian Forest Service, the Green Plan has funded research in a broad range of topics from Ecological Reserves to Decision Support Systems. In 1992, the Canadian Council of Forest Ministers facilitated the production of a National Forest Strategy entitled Sustainable Forests: a Canadian Commitment (Anonymous, 1992a). This document commits governments to report on progress in achieving "sustainable" forestry including the status of forest biodiversity. This strategy was developed prior to the recent United Nations conference on Environment and Development (UNCED) and provided a basis for Canada's position on forest related issues. An International Convention on Biological Diversity was also signed by a number of countries at the UNCED meeting. As a signatory, Canada is developing a Environmental Monitoring and Assessment 39: 399-415, 1996. (~ 1996 Kluwer Academic Publishers. Printed in the Netherlands.

400

DANIEL W. McKENNEY ET AL.

biodiversity strategy through a Biodiversity Convention Office. A number of related workshops have been held to support policy development in this area, and without exception, they have recommended better coordination of ecological databases (e.g., Biodiversity Convention Office, 1994; McKenney et al. 1994). This paper identifies six impediments to operationalizing ecosystem management. Overcoming these impediments can be linked to the development and use of particular types of ecological data. Work in Ontario is described that provides an approach to integrate these data. Finally, some suggestions are made to pursue this type of integrated database development Canada-wide.

2. Impediments to Operationalizing Ecosystem Management Implementation of ecosystem management remains an elusive operational challenge, and one that remains subject to a wide range of interpretations (e.g., see Booth et al., 1993; Marshall et al., 1993; Woodley, 1993). Six major impediments to the implementation of ecosystem management are noted below. 2.1. LACK OF AN ANALYTICAL FRAMEWORK

In most "forest planning" models, non-wood values are generally articulated as some kind of constraint on timber production. This is a practical approach but is inadequate for ecosystem management which must be centred on the concept of multiple-use forestry. Forest economics has evolved to include multiple-uses in principle (i.e., priced and unpriced values) (Bowes and Krutilla, 1989); however, much effort is required to operationalize the theory for real-world decision-making. In principle, wood production should be considered as only one of the goods and services that can be generated from a forest. To be complete, the ecosystem planning framework needs to directly address corollaries of timber versus ecosystem management questions. For example, instead of only examining the opportunity cost of forgoing wood in favour of habitat protection, the question should also be addressed in terms of "what is the ecological opportunity cost of forgoing habitat protection in order to extract wood from a landscape?" 2.2.

INADEQUATE FOREST SURVEY METHODS

Standard forest resource survey/inventory methods are primarily aimed at generating information about the geographic distribution of wood volumes for commercial tree species. The conservation of biodiversity (a prime objective of ecosystem management) demands that attention be given to the distribution of a much wider range of species (e.g., non-commercial plants, wildlife and even microorganisms in a landscape; see Hunter, 1990; Crow et al., 1994). While standard forest resource inventories provide some of the required data, even forest ecosystem classification surveys (e.g., Sims et al., 1989) which sample all of the vascular plants in a set of

PRIMARYDATABASESFOR ECOSYSTEMMANAGEMENT

401

plots fall short of what is required. The mobility of vertebrate animals within forests and adjacent habitats makes even simple presence/absence lists inadequate. 2.3.

POORLY QUANTIFIED HABITAT DEFINITION AND MAPPING

A key problem for wildlife conservation revolves around precisely identifying and quantifying those components of the total environment which constitute an organism's habitat. Assuming that this can be done, then a next required step involves accurately mapping where this habitat occurs in the landscape. Given the complexities of resource partitioning amongst individuals, populations and species, the resultant spatial patterns can be extremely complicated. Moreover, such patterns often do not coincide with forest stand boundaries, as differentiated using timber inventory standards, or air-photo interpretation techniques which seek to apply those standards. 2.4. POOR UNDERSTANDING OF GENETIC DIVERSITY

The maintenance of within-species genetic diversity is another important goal of ecosystem management. In Canada, evidence is mounting that while many boreal species are widespread, genetic diversity is expressed through the process of local/regional adaptation (D. Joyce, Ontario Ministry of Natural Resources, pers. comm.). Thus populations of species develop adaptive gene complexes in response to local environmental conditions, particularly climate (Rehfeldt, 1990; Ledig, 1993). Scientific understanding of the spatial variation in genetic diversity for both common and rare species is generally poor. 2.5. DIFFICULTIES IN EXTRAPOLATING BIOLOGICAL SITE DATA

Ecosystem management generally requires resource information at a landscape scale. This is a problem since many surveys about biodiversity are taken at an irregular scatter of plots. Clearly, budget constraints prevent complete surveys of biological resources in all locations so survey data must be extrapolated. Survey results can be useful for spatial extension, however they need to be placed in context with dominant landscape features. Standard extrapolation methods based on air-photo interpretation are generally inadequate. These techniques need to be complemented by emerging computer-based methodologies such as those described later in this paper. 2.6. THE INABILITY TO ACCOUNT FOR SYSTEM DYNAMICS

Forest ecosystems are dynamic. Plants grow through time and respond to fluctuating environmental conditions, with commensurate changes in growth and yield, stand composition and structure, and distribution and availability of wildlife habitat.

402

DANIEL W. McKENNEY ET AL.

Ecosystem management must recognize these dynamics and how they vary spatially. Moreover, resource planners require a functional capability to investigate, evaluate and test alternative management and trade-off scenarios using a variety of simulation approaches. Hence, capacities are needed to generate predictions and anticipated management consequences, as well as static descriptions of forest pattern and process. This can only be accomplished by adopting a process-oriented, computer-based modelling and simulation approach.

3. Key Biophysical Data for Ecosystem Management Our thesis is that underpinning these particular six impediments is one common deficiency, namely, the lack of appropriate, accurately geo-referenced primary biophysical data to support the required spatial modelling. Forest ecosystem process models (e.g., Band et al., 1991, 1993), canopy gap models (e.g., Shugart et al., 1992), habitat supply models (e.g., Lindenmayer et al., 1991) and statistical models for spatially extending site data (e.g., Mackey, 1994) all require a remarkably similar set of primary input data. Of critical importance are estimates, or at least indices, of the main forcing functions of biological systems. These are represented by the thermal, radiation, moisture and nutrient regimes. These regimes provide a conceptual framework for modelling the distribution and availability of heat, light, water and nutrients - the primary physical determinants of biotic response. Quantifying these Primary Environmental Regimes (PERs) requires processing of primary climatic, terrain and substrate (soil/geology) attribute data. Data about forest canopy characteristics are also critical for more accurate representation of the hydrological cycle and soil water status in particular, and finer-scaled simulation of below-canopy micro-climatic conditions. PERs provide the landscape template against which forest dynamics occur. For example, the reponse of forest stands to disturbance is mediated by the prevailing physical environmental conditions. Disturbance data are still required, but prediction of future system response requires that landscape constraints be factored in. The role of PERs in studying the composition and productivity of forest ecosystems has long been recognized. In developing ecoregionalizations for Ontario, Hills (1961) based his analysis on macro-climatic gradients and their interaction with substrate and topography. However, he lacked the technology to directly quantify their effect on the biota. Instead he used the vegetation as an environmental "phytometer" (see Burger, 1993). With the introduction of new computer-based technologies, the situation has now changed. It is now possible to directly model the space/time variation in energy, moisture and mineral nutrient regimes and use this information to directly examine the response of particular biological phenomena to environmental conditions. The collation of existing biological survey data provides the basis to empirically calibrate the response of taxa to PERs. Better estimates of climate can now be

PRIMARY DATABASES FOR ECOSYSTEM MANAGEMENT

403

calculated ex situ at survey plots using computer models (i.e., at points between weather monitoring stations). Statistical analysis of these data can now be used to quantify the potential climatic niche of any taxa (e.g., Mackey and Sims, 1993). This framework provides the ability to undertake predictive spatial modelling of biological and ecological phenomena at a range of scales. For example, the potential climatic domain of a taxa can be mapped by a statistical function coupled to gridded estimates of the climatic variables (e.g., Mackey, 1994; Mackey et al., 1994a, see also Nix, 1986). Not only can species distributions be analyzed in this way but any biological unit can be targeted for analysis. In summary, selected climatic, terrain, substrate (soil/geology), and land cover attribute data are required to model the spatial distribution, and seasonal variation, in the PERs. Site data from biological surveys enables the environmental determinants of plant and animal response to be examined. 3.1. E X A M P L E S FROM ONTARIO

The approach being pursued (i.e., coupling statistical models of ecological response to geo-referenced/spatial databases) is very different from traditional ecosystem planning, A standard approach has been to use aerial photography to delineate existing forest timber stands, digitize the results as input to a geographic information system (GIS), and subsequently relate other ecological or biological phenomena to these timber stands. Fortunately, much original biological survey data remains available and thus can be linked to estimates of the PERs. On-going collaborative efforts in Ontario are aimed at the development of extensive (province-wide) primary databases to support environmental modelling and trade-off analyses (e.g., see Mackey and McKenney, 1994; Sims and Mackey, 1994). The compilation of these databases is currently providing a basis for the extrapolation, across entire landscapes, of data and knowledge gained from various current and historical field-based biological surveys and monitoring activities throughout Ontario. As suggested above, other environmental data are important and often critical. In Figure 1, biological competition, natural disturbance regimes (e.g., insect and fire), and direct human influences (e.g., harvesting) are also identified as important causal agents of forest pattern. But it is the PERs which ultimately constrain the biological composition and productivity potential of any given site. Consequently, these data have been given priority for current database development activities. The primary databases developed for Ontario include: 1) The development o f a province-wide Digital Elevation Model (DEM), using methods developed by Hutchinson (1989), and now available within the ARC/ INFO TM GIS package. A DEM is a regular grid or array of latitude, longitude and elevation (x, y, z) that represents the topography of the landscape. Topography plays a role in all four PERs. It exerts control on the distribution and availability of water, radiation, heat, and indirectly affects mineral nutrient regimes through erosion and leaching. Elevation also affects temperature and precipitation at a

404

DANIEL W. McKENNEY ET AL.

Pdmary Environmental Reglm~ Thermal Radiation Moisture Mineral nutrient f

Disturbance Regimes Fh'e Insects I

Disease

L Catastr~

I I E)dantEcological Response I ~

BiologicalRegimes

Physiognomic s t r u c t u r e / ~ Taxonomic composition, l" I abundance, viability } I Process rates J I

Competition Dependencies Geneticvariation

t AnthropogenlcRegimes Land use history Land use impact Extant land cover IrdrastnJclure Fig. 1. Determinants of ecological response.

meso-scale. There is now an enhanced capability for modelling topography, water catchments, and many terrain attributes at a number of scales, from a 100 m grid resolution to 1 km or greater across the entire province, and at 20 m resolution within the productive forest zone where digital National Topographic Series (1:50 000) or Ontario Base Maps (1 ::20 000) have been completed. Attributes such as slope, aspect, catchment areas, topographic wetness indices and flow paths, can be derived from DEMs (see Moore et al., 1991; Mackey et al., 1994b). 2) The development of climate surfaces for the entire province. These currently enable estimates of some 40 long-term mean monthly climatic variables (e.g., annual mean temperature; monthly mean, minimum, maximum temperature and precipitation; growing season start and end; degree days during the growing season; etc.) for any location where latitude, longitude and elevation (x, y, z) are known (Mackey et al., 1995, in press). A mathematical procedure using thin plate smoothing splines is used to interpolate an Ontario-wide network of long-term mean monthly weather station data (Hutchinson, 1987). The procedure estimates a bivariate function based upon position and elevation. Standard errors associated with the predicted values are generated. By coupling these surfaces to the DEM, gridded estimates of the climate variables can be generated and input into a GIS database. This process is illustrated in Figure 2 which shows gridded values for the province of growing season length (i.e., number of days above a base temperature). Additional efforts are aimed at creating surfaces of potential evaporation, winter severity and variances from the long term means. By overlaying several of these surfaces and using spatial statistics, various climatic classifications of the province have been derived. These techniques have the advantage of identifying areas that

405

PRIMARY DATABASES FOR ECOSYSTEM MANAGEMENT

11

~139

138 48

114o

-

ss

1159

-

68

I lo9

-

78

~179

-

88

i1189

-

98

199

-

288

[209

-

218

1219

-

228

1229

-

238

1239

-

248

Fig. 2. Growingseason lengthin Ontario resolvedat a 1 km resolution.

are geographically separate but climatically similar (see for details Mackey et al., in press). 3) The compilation of various growth and yield, and Forest Ecosystem Classification survey data, and other biological site data. To date, in collaboration with the Ontario Ministry of Natural Resources (OMNR), almost 6000 existing forest plot records have been collated (see Figure 3; McKenney et al., 1995")'. This includes accurately geo-referencing the individual plots (i.e., appending z, y, z co-ordinates) and examining the similarity of the surveys to determine if they can be combined. Ongoing collaborations are facilitating the compilation of other site data (i.e., wildlife data, forest insect and disease surveys, herpetofauna surveys, forest health surveys). Partners include the OMNR, the Canadian Wildlife Service, the Ontario Natural Heritage Information Centre, the University of Guelph and other CFS scientists. These survey records are being reanalyzed using the climate

406

D A N I E L W. M c K E N N E Y

ET AL.

Distribution of F E e plots across Ontario

Distribution of Growth and Yield plots across Ontario

..

---,,-,-w'--7

Fig. 3. Distribution in Ontario of forest ecosystem classification (FEC) plots (top), and growth and yield plots (bottom).

PRIMARY DATABASES FOR ECOSYSTEM MANAGEMENT

407

models so that predictive functions of potential distributions and productivity can be derived. 4) The provincial forest resource (wood) inventory geo-referenced at (approximately) a 10 km x 10 km scale in many parts of the province. Wood inventory data is derived from the Ontario Forest Resource Inventory (FRI). Aerial photographs were used to derive stand summary information for major tree species, including such parameters as composition, site class, stocking, age class and area. OMNR is currently in the process of digitizing the FRI data down to the stand level across the province. 5) Provincial surficial geology, bedrock geology and soils information. Substrate data are required as inputs to finer-scaled modelling of the soil/water and nutrient balances in a landscape. Various existing thematic mapped data are currently being digitized by various government agencies (mainly the Ontario Geological Survey) including provincial coverage of bedrock and surficial geology, the Ontario Land Inventory, and the Northern Ontario Engineering and Geological Terrain Survey (NOEGTS) maps. Considerable effort is being exerted to re-analyze these data in conjunction with the new terrain and climatic models. 6) Existing land cover for the entire province via remotely sensed satellite imagery. Land cover data are ultimately required: a) to provide a spatial template when predicting potential distributions, b) as input to habitat supply models, c) to provide data about land use disturbance, d) to modify the evaporation component of water-balance calculations, and e) as a spatial index of biotic response at a system level. Whilst landcover data are clearly required at a range of scales, a preliminary coverage has been obtained from the (NOAHH) AVHRR satellite sensor at a I km resolution (e.g., Palko et al., 1993). This matches the resolution of the meso-scaled climate data now developed for the province. In addition, OMNR is currently undertaking more detailed thematic land cover mapping, based upon a supervised classification of Landsat Thematic Mapper imagery (Anonymous, 1992b), for portions of the productive forest landbase in northern Ontario. 7) Digitized coverage of disturbance history on the land base including land-use history, fire history, and insect and disease outbreaks. These efforts are ongoing by various researchers and will provide, among other uses, a computer-based inventory of remoteness and naturalness for the province, based upon the method developed by Lesslie et al. (1988). 8) The development of more detailed prototype studies for operational-level applications. Underpinning the province-wide efforts is the development of similar databases at a finer resolution for the Rinker Lake Research area in northwestern Ontario (Sims and Mackey, 1994; Sims et al., 1994; Sims, 1994). This work is supporting a significant field validation project aimed at operational (e.g., about 1:20 000 scale) rather than strategic resource management issues. Various field studies and finer-scaled spatial modelling will enable evaluation of the error associated with the broader-scaled, province-wide analysis.

408

DANIEL W, McKENNEY ET AL.

The database development described above provides the capability to address previously intractable ecosystem management problems in a more quantitative, explicit and repeatable manner. The modelling framework is computer-based and can be used to support policy and operational problems in either an economic or non-economic context (impediment 1). The integration and spatial extension of existing surveys for forest sites in Ontario, including data from Forest Ecosystem Classification plots, growth and yield plots, Forest Insect and Disease monitoring plots, and wildlife surveys, have not been available in a spatial (GIS) context previously and has already spawned a number of specific applications which are currently being completed and reported on. These efforts are adding value to previously-collected biological survey data through reanalysis and spatial extension of the results (impediments 2 and 5), helping to quantify and map habitat of selected animal species (impediment 3) and creating opportunities to characterize genetic diversity at the population level through links to genecological studies (impediment 4). The methods adopted here also overcome most of the long-standing impediments to regional and landscape process modelling, namely the difficulty of generating reliable spatial estimates of PERs (impediment 6). To facilitate adoption of results quickly, decision support systems (DSS) are being developed that utilize the primary data for particular problems. For example, a DSS has been developed (SEEDWHERE) to assist in making seed and nursery stock transfer decisions in support of genetic resource management (Mackey and McKenney, 1995). At any given point in time, Ontario has between 8 and 13 billion seeds in storage. Decisions are being made about where to collect seed, where to use existing seedlots, how far to move seed from point of origin, etc. It is recognized that populations of tree species are genetically adapted to local or regional environmental conditions; forest regeneration efforts can be at risk if seed is planted in areas where the environmental, particularly climatic, conditions are dissimilar. Negative effects can include lower growth rates and increased probability of mortality and frost damage, and susceptibility to insects and diseases. Historically, seed zones have been established based largely on an intuitive understanding of meso-scale climatic gradients. Using the primary databases described above (i.e., climate surfaces and the DEM), SEEDWHERE generates similarity indices of the climate between regeneration sites, seed collection sites and/or existing seedlot sources. Genecological studies are underway to identify the actual index value beyond which it is unwise to move seed (D. Joyce, pers. comm.). Another DSS under construction which draws upon the primary databases is EDIS - "Environmental Domain Interrogation System" (Mackey et al., in prep). EDIS is being developed to support assessments of the representativeness of forest plots and parks against given environmental classifications (e.g., the climatic classification described above). EDIS has already been used to assess the representativeness of Forest Ecosystem Classification (FEC) and Growth and Yield (G&Y) plots in Ontario (McKenney et al., 1995a). The EDIS analysis has helped to identify significant gaps in the current plot inventory and where new plots would be

PRIMARY DATABASES FOR ECOSYSTEM MANAGEMENT

409

best located to sample the entire range of "environmental" space in the province. These analyses are helping in the determination of sampling priorities for ongoing provincial FEC and G&Y programs.

4. NatGRID - Towards the Development of National Geo-Referenced Information for Decision Makers

As already noted, traditional forest inventory has focussed on measuring attributes that are concerned mainly with wood supply. Sustainable development of Canada's forest resources ultimately requires the development and application of knowledgebased resource management systems that provide context to both ecological and wood resources. Reliable, spatially oriented computer-based decision support systems assist with this objective. Data are required about the distribution and abundance of many plant and animal species (not just commercial trees), their habitats, and the driving forces of ecological systems - climate, terrain, substrate, disturbance. Many of these attributes are not readily amenable to conventional resource inventory techniques but are essential for developing predictive, in addition to descriptive, capabilities. The need to examine environmental impacts across space/time scales further reduces the utility of traditional, cartographic-based resource inventories. If Canada is to move towards an ecosystem approach to natural resource management, an integrative framework is required. The approach and new technologies discussed above provide a basis for this framework. A spatially-referenced national inventory could provide the framework for ecosystem resource planning, monitoring, reporting and research. The approach is multi-disciplinary, drawing upon established partnerships and, as described above, provides the basis for integrating a range of existing datasets. We propose that a specified set of geo-referenced resource data be established with national coverage to support ecosystem management and the sustainable development of Canada's forest resources. This would create a NatGRID - a Nationally G__eoreferenced R_.esource Information for Decision-making database. The role for NatGRID would be to "provide a planning framework for assessing

decision options and trade-offs between wood production and biodiversity conservation", which is a central principle of ecosystem management and the crux of ecologically sustainable development in forestry. Four key steps in the development of a NatGRID are: 1. Generate nation-wide spatial data sets: 9 Digital elevation and terrain models (DEMs), including distributed models of catchment hydrology, based upon existing national coverage of National Topographic Series 1:250 000 mapsheets, digitized and archived by the Canada Centre for Mapping, Natural Resources Canada, Ottawa.

410

DANIEL W. McKENNEY El" AL.

Meso-scaled climate surfaces of Canada, based upon the existing national network of long-term climate recording stations, as archived and updated by Atmospheric Environment Service, Ottawa. 9 Remotely sensed landcover classifications of Canada (e.g., AVHRR, LANDSAT, SPOT, RADARSAT), as available from various agencies within Canada, but primarily the Canada Centre for Remote Sensing, Natural Resources Canada, Ottawa. 9 GIS coverage of geological and soils cover for Canada, as available from various agencies within Canada, but as a principal source, the Geological Survey of Canada, Natural Resources Canada, Ottawa and Agriculture Canada, Ottawa (e.g., Shields et al., 1991). 9

2. Compile and reanalyze existing biological (plant and animal) survey data; currently, most of the ecological (especially forest vegetation and soils) inventories are dispersed throughout the country and are archived or stored by various government agencies, including the Canadian Forest Service and several provincial forestry ministries. A concerted effort would be required to bring many of these together, ensure that they are properly and accurately geocoded, and, depending upon the individual surveys, bring the plot databases into a relatively common format, so that they could be used as part of a Canada- wide forest plot database. 3. Identify new surveys required to fill gaps through assessments of the representativeness of existing biological datasets. 4. Develop Decision Support Systems that enable partners such as provincial agencies and industry to make immediate use of project deliverables for forestry applications. The notion of NatGRID is practical and essential for several reasons: 1. Ecological systems transcend administrative boundaries, thus broad landscape perspectives are required for context (even outside areas where forests currently exist). 2. NatGRID makes use of existing datasets and adds value to biological survey data by providing a framework for the potential spatial extension of site data. Numerous biological survey datasets already exist and have already been mentioned. See Table I for additional illustrative examples. The size, location, exact state and other properties of these various datasets makes it more difficult to estimate the time involved in obtaining and integrating these into a common format/framework. As a rough estimate, 4 to 6 person-years (PYs) over 3 years are required, but an initial survey of the condition of these databases should lead to a more realistic estimate. 3. Data generated by NatGRID would be valuable in fulfilling Canada's obligations to the international Climate Change and Biodiversity Conventions by providing a mechanism for reporting and numerous scientific investigations.

411

PRIMARY DATABASES FOR ECOSYSTEM MANAGEMENT

TABLE I Example biological data sets. PRIMARY DATA Insect

EXAMPLE SOURCES*

Mayflies, Stoneflies and Caddisflies of Manitoba Forest Insect and Disease Information System (FIDS) Forest Insect Pest Parasite Database

Canada

Forest Depletion Estimation System (Insect-caused)

Canada Canada & Alaska Canada & Worldwide

Wildlife Serum Bank (Ungulates)

Canada

Residues in Arctic Wildlife (Seal, Polar Bear, Caribou) Yukon Land Use Planning GIS (Wildlife Habitats) He~etology Database

Canadian Arctic Yukon Canada & Worldwide

Mammal Collection Database

Tree

Canada

Forest Pest Infestation Maps

General Invertebrates Database

Plant

Manitoba & North America

BC & Yukon

Freshwater Planktonic Invertebrates Database

Animal

GEOGRAPHIC SCOPE OF S U R V E Y

Worldwide

Algae Herbarium Program Database

North America & Worldwide

Lichens Program Database

North America & Worldwide

Mosses Program Database

North America & Worldwide

Vaxcular Plants Program Database

North America & Worldwide

Acid Rain Impact Assessment Database

Eastern Canada

Acid Rain National Early Warning System (ARNEWS)

Canada

Forest Biomass Inventory

Canada

Microcomputer Permanent Forest Sample Plot Catalogue Prairie-NWT Forestry Data Bank and Archives

**AB, BC, MB, NB, PQ & SK Prairie Provinces & NWT

Yukon Land Use Planning GIS (Forest Inventory)

Yukon

Yukon Timber Production

Yukon

Forest Productivity Database

**AB, BC, MB. NS & SA

*Sources from the Statistics Canada draft report; Databases For Envkonmental Repotting: Federal Government Departments, excluding Environment Canada, January, 1991. AB: BC: MB:

Alberta British Columbia Manitoba

NB: NS: SK:

New Brunswick Nova Scotia Saskatchewan

412

DANIEL W. McKENNEY ET AL.

4. The development of a national DEM and climate surfaces is feasible in a short period of time. Based upon experiences and accomplishments in Ontario, the DEM work would likely take in order of 10 to 15 PYs, and could be completed at the 1:250 000 scale within about 3 years with this level of effort. The development of country-wide climate surfaces could also be completed from existing data within about 3 years, with 3 to 5 PYs. 5. Remotely sensed landcover classifications of the country already exist at roughly a 1 km grid resolution. In addition, finer-scaled analyses (e.g., Landsat TM) have been completed for selected parts of the country. We believe that a concentrated effort could provide national coverages based upon TM imagery within a short period of time; our understanding is that the biggest impediment here is the cost of the imagery, not necessarily the technology or effort required to complete this work. 6. Geological and soils cover have already been digitized for some parts of the country. Some effort would be required to compile and coordinate the existing datasets, and to obtain, for other areas, those map coverages which are not available in digital form. 7. Adoption of generic DSSs that utilize the primary data would be relatively straightforward given our experiences in Ontario, and the generally high level of information technology skills held, and being developed, by resource managers in Canada. In summary, for a relatively modest investment, significant advances towards a more quantitative basis for ecosystem management could be attained in a short period of time: 15 to 30 PYs, plus capital and coordination efforts over approximately 3-5 years of $1.5-2.0 million in total.

5. Concluding Comments Ecosystem management involves not just descriptions of current inventories of ecological values but also predictive capabilities about forest dynamics. These dynamics will vary in space and time as a function of primary environmental regimes and disturbance regimes. The efforts described above illustrate that spatially distributed databases of key primary biophysical attributes represent a significant contribution to many research programs and operational planning problems, including: ecological land and forest ecosystem classification, landscape ecology, bio-monitoring, wildlife habitat modelling, timber production planning and resource economic trade-off analyses. Among the other benefits already noted, NatGRID would lead to a more quantitative basis for defining those ecological regionalizations which clearly transcend administrative boundaries. In addition, NatGRID would provide a strong foundation for better coordination of Canadian ecological characteristics/ranges within an international context; similar activities to develop detailed DEMs and climate

PRIMARYDATABASESFOR ECOSYSTEMMANAGEMENT

413

surfaces are already underway in the USA and M e x i c o (S. Jenson, U.S. Geological Survey, pers. comm.). NatGRID could help to address ecosystem managementrelated questions at many levels o f jurisdiction (international, national, provincial/regional, municipal). Canada has an excellent past record in ecological survey, information technology and forest inventory; however, there is now an opportunity to build on past successes and develop a new, nation-wide approach to the problem o f ecosystem management. B y making more explicit the links between biotic systems and the environmental determinants, predictive as well as descriptive capabilities are possible for a much wider range o f ecosystem components than was previously possible. Also, rather than being limited to one predetermined set o f boundaries, ecological regionalizations can be enhanced or, if required, generated, to address particular ecosystem m a n a g e m e n t problems.

Acknowledgements T h e opinions expressed here are those o f the authors and not the organizations we work for. The Ontario work described here was financially supported by the Canadian Forest Service, Ontario Ministry o f Natural Resources, Sustainable Forestry Initiative, the Northern Ontario Development Agreement (NODA) and the Green Plan. Initial feasibility/background investigations related to NatGRID were supported by the Canadian Forest Service's Ecological Land Classification initiative.

References Anonymous: 1992a, 'National forest strategy: sustainable forests: a Canadian commitment', Canadian Council of Forest Ministers, Ottawa, Ontario, 28 pp. Anonymous: 1992b, 'Development of a spatial database of red and white pine old-growth forest in Ontario-West', Ont. For. Fragment. & Biodiv. Project, Rep. No. 5., Ont. For. Res. Inst., Ont. Min. Nat. Resour., Sault Ste. Marie, prepared by Spectranalysis Inc., Oakville, Ontario, 41 pp. + mapsheets. Band, L.E., Peterson, D.L., Running, S.W., Coughlan, J., Lammers, R., Dungan, J. and Nemani, R.: 1991, 'Forest ecosystem processes at the watershed scale: basis for distributed simulation', Ecol. Model. 56, 151-176. Band, L.E., Patterson, P., Nemani, R. and Running, S.W.: 1993, 'Forest ecosystem processes at the watershed scale: incorporating hillslope hydrology', Agric. and For. Meteorol. 63, 93-126. Biodiversity Convention Office: 1994, 'Canadian biodiversity strategy', Draft, Ottawa, Ontario. Booth, D.L., Boulter, D.W.K., Neave, D.J., Rotherham, A.A. and D.A. Welsh.: 1993, 'Natural forest landscape management: a strategy for Canada', Forestry Chronicle 69, 141-145. Bowes, M.D. and Krutilla, J.V.: 1989, Multiple-Use Management: The Economics of Public Forest Lands, Resources for the Future, Johns Hopkins University Press, Washington, 350 pp. Burger, D.: 1993, 'Revised site regions of Ontario: concepts, methodology and utility', Research Report No. 129, Ont. For. Res. Inst., Ont. Min. Nat. Resour., Sault Ste. Marie, Ontario, 24 pp. Crow, T.R., Haney, A. and Waller, D.M.: 1994, 'Report on the scientific roundtable on biological diversity convened by the Chequamegon and Nicolet National Forests', Gen. Tech. Rep. No. NC166, US Dep. Agric., For. Serv., North Central For. Exper. Stat., St. Paul, Minnesota, 54 pp.

414

DANIEL W. McKENNEY ET AL.

Government of Canada: 1990, Canada's Green Plan, Department of Supply and Services, Ottawa, Ontario, 174 pp. Hunter Jr., M.L.: 1990, Wildlife, Forests and Forestry: Principles of Managing Forestsfor Biological Diversity, Prentice Hall Inc., Englewood Cliffs, New Jersey, 370 pp. Hutchinson, M.F.: 1989, 'A new procedure for gridding elevation and stream line data with automatic removal of spurious pits', J. HydroL 106, 211-232. Hutchinson, M.E: 1987, 'Methods for generation of weather sequences', In: A.H. Bunting (ed.), Agricultural Environments: Characterisation, Classification and Mapping, CAB International, Wallingford, UK, pp. 149-157. Ledig, ET.: 1993, 'Secret extinctions: the loss of genetic diversity in forest ecosystems', In: M.A. Fenger, E.H. Miller, J.A. Johnson and E.J.R. Williams (eds.), Our Living Legacy: Proceedings of a Symposium on Biological Diversity, Royal British Columbia Museum, Victoria, British Columbia, pp. 127-140. Lindenmayer, D.B., Nix, H.A., McMahon, J.P., Hutchinson, M.E and Tanton, M.T.: 1991, 'The conservation of Leadbeater's Possum, Gymnobelideus leadbeaten McCoy, a case study of the use of bioclimatic modelling', J. Biogeogr. 18, 371-383. Lesslie R., Mackey, B.G. and Preece, K.M.: 1988, 'A computer-based method for evaluation of wilderness', Environmental Conservation 15, 225-232. Mackey, B.G.: 1994, 'Predicting the potential disturbance of rainforest structural characteristics', Journal of Veg. Science 5, 43-54. Mackey, B.G. and Sims, R.A.: 1993, 'A climatic analysis of selected boreal tree species, and potential responses to global climate change', World Res. Rev. 5(4), 469-487. Mackey, B.G. and McKenney, D.W.: 1994, 'The bio-environmental indices project: an overview', NODA/NFP Tech. Rep. No. 1, Can. For. Serv. - Ont. Region, Sault Ste. Marie, Ontario, 5 pp. Mackey, B.G., Sims, R.A., Baldwin, K.A. and Moore, I.D.: 1994a, 'Spatial analysis of boreal forest ecosystems: results from the Rinker Lake case study', In: Proceedings Internat. Conf. on Integrating GIS and Environmental Modelling, Sept. 26-30, 1993, Breckenrdge, Colorado, (in press). Mackey, B.G., McKenney, D.W., Sims, R.A., Widdifield, C., Lawrence, K. and Szcyrek, N.: 1994b, 'A new digital elevation model of Ontario', NODAJNFP Tech. Rep. No. TR-6, Can. For. Serv. Ontario, Sault Ste. Marie, Ontario, 26 pp. Mackey, B.G. and McKenney D.W.: 1995, 'SEEDWHERE: a computer tool for tree planting and ecological restoration', NODA/NFP Tech. Note No. 12, Can. For. Serv. - Ontario, Sault Ste. Marie, Ontario. Mackey, B.G., McKenney, D.W. and Nippers, B.: 1995, 'EDIS -Environmental Domain Interrogation System: a decision support tool for assessing representativeness', NODA/NFP Tech. Rep., Can. For. Serv. - Ontario, Sault Ste. Marie, Ontario, (in press). Mackey, B.G., McKenney, D.W., Yang, Y.Q., McMahon, J.P. and Hutchinson, M.F.: 1995, 'A climatic analysis of Hills site regions for the province of Ontario, Can. J. For. Res. (in press). Marshall, I.B., Hirvonen, H. and Wiken, E.: 1993, 'National and regional scale measures of Canada's ecosystem health', In: S. Woodley, J. Kay, and G. Francis (eds.), Ecological Integrity and the Management of Ecosystems, Heritage Resources Centre, Univ. Waterloo & Can. Parks Serv., St. Lucie Press, Ottawa, Ontario, pp. 117-129. McKenney, D.W., Mackey, B.G., McKee, J.E. and Zavitz, B.L.: 1995, 'A representativeness assessment of forest ecosystem and growth and yield plots in Ontario', NODA/NFP Tech. Rep., Can. For. Serv. - Ontario, Sault Ste. Marie, Ontario, (in press). McKenney D.W., Sims R.A., Mackey, B.G. and Campbell, K.L. (eds.): 1994, Towards a Set of Biodiversity Indicators for Canadian Forests: Proceedings of a Forest Biodiversity Indicators Workshop, Natural Resources Canada, Can. For. Serv. - Ontario, Sault Ste. Marie, Ontario, 133 pp. Moore, I.D., Grayson, R.B. and Ladson, A.R.: 1991, 'Digital terrain modelling: a review of hydrological, geomorphological and biological applications', Hydrol. Process 5, 3-30. Nix, H.A.: 1986, 'A biogeographic analysis of Australian elapid snakes', In: R. Longmore (ed.), Atlas of Elapid Snakes of Australia, Australian Flora and Fauna Series No. 7, Australian Government Publications Service, Canberra, Australia, pp. 4-15.

PRIMARYDATABASESFOR ECOSYSTEMMANAGEMENT

415

Palko, S., Lowe, J.J. and Pokrant, H.T.: 1993, 'Canada's new, seamless forest cover data bases', In: Proceedings, GIS 93, Vancouver, British Columbia, pp. 987-990. Rehfeldt, G.E.: 1990, 'Genetic differentiation among populations of Pinus ponderosa from the Upper Colorado River Basin', Botanical Gazette 151(1), 125-137. Shields, J.A., Tarnocai, C., Valentine, K.W.G. and MacDonald, K.B.: 1991, 'Soil landscapes of Canada: procedures manual and user's handbook', Publ. No. 1868/E, Agriculture Canada, Communications Branch, Ottawa, Ontario, 74 pp. Shugart, H.H., Smith, T.M. and Post, W.M.: 1992, 'The potential for application of individual-based simulation models for assessing the effects of global change', Ann. Rev. Ecol. System 23, 15. Sims, R.A., Towill, W.D., Baldwin, K.A. and Wickware, G.M.: 1989, 'Field guide to the forest ecosystem classification for northwestern Ontario', For. Can., Ont. Reg., Sault Ste. Marie, and Ont. Min. Nat. Res., Thunder Bay, Ontario, 191 pp. Sims, R.A. and Mackey, B.G.: 1994, 'Development of spatially-based ecosystem models for the Rinker Lake Research Area in northwestern Ontario's boreal forest', In: Proceedings, GIS 94, Volume 2, Vancouver, British Columbia, pp. 665-674. Sims, R.A.: 1995, 'Ecologically-based land and site classifications: frameworks for the description of ecological resources', In: T.J. Beechey (ed.), Workshop Proceedings, Coping with Ecological Representation, Ann. Gen. Mtg., Can. Counc. Ecol. Areas, Aug. 10-13, 1993, Windsor, Ontario, (in press). Sims, R.A., Mackey, B.G. and Baldwin, K.A.: 1995, 'Stand and landscape level applications of a forest ecosystem classification for northwestern Ontario, Canada', Annal. Sci. For. 6. Woodley, S.: 1993, 'Assessing ecosystem health in protected areas', In: S.F. Poser, W.J. Crins and T.J. Beechey (eds.), Size and Integrity Standards for Natural Heritage Areas in Ontario, Proceedings of a Seminar, Parks and Natur. Heritage Policy Br., Ont. Min. Natural Resour., Huntsville, Ontario, pp. 93-99.