Remote sensing and forest inventory for wildlife habitat assessment

Researchers and managers undertaking wildlife habitat assessments commonly require spatially explicit environmental map layers such as those derived from forest inventory and remote sensing. However, end users of geospatial products must often make choices regarding the source and level of detail required for characterizing habitat elements, with few published resources available for guidance. We appraised three environmental data sources that represent options often available to researchers and managers in wildlife ecological studies: (i) a pre-existing forest inventory; (ii) a general-purpose, single-attribute remote sensing land cover map; and (iii) a specific-purpose, multi-attribute remote sensing database. The three information sources were evaluated with two complementary analyses: the first designed to appraise levels of map quality (assessed on the basis of accuracy, vagueness, completion, consistency, level of measurement, and detail) and the second designed to assess their relative capacity to explain patterns of grizzly bear (Ursus arctos) telemetry locations across a 100,000-km² study area in west-central Alberta, Canada. We found the forest inventory database to be reasonably functional in its ability to support resource selection analysis in regions where coverage was available, but overall, the data suffered from quality issues related to completeness accuracy, and consistency. The general-purpose remote sensing land cover product ranked higher in terms of overall map quality, but demonstrated a lower capacity for explaining observed patterns of grizzly bear habitat use. We found the best results using the specific-purpose, multi-attribute remote sensing database, and recommend that similar information sources be used as the foundation for wildlife habitat studies whenever possible, particularly those involving large areas that span jurisdictional boundaries.     

Modeling changes in habitat conditions in northern hardwoods forests of the Adirondack Mountains of New York

Interest in enhancing the management of forest timber and wildlife resources has stimulated the development of modeling techniques to predict changes in timber and wildlife habitat characteristics. We assess the potential of combining models which predict forest structure with those which assess habitat quality to predict changes in forest habitat characteristics. We measured variables important to simulating forest growth and evaluating wildlife habitat in even-aged stands making up an 80 year chronosequence, and developed mathematical relationships between these sets of variables. The regression relationships were generally excellent (r² or R² > 0.63), and most habitat variables were strongly related to stand age. In earlier work, we incorporated these models into FORET, a model that successfully simulates succession in hardwoods forests of the eastern USA. In the present study, we adapted FORET to accommodate the growing season and species composition of northern hardwoods forests of the Adirondack Mountains of New York, creating a new model called FORDACK. We compared stand measurements with output generated by the model. FORDACK accurately simulated changes in structural characteristics of vegetation and predicted which species would be present in very low densities, but did not successfully predict the relative densities of the dominant species. Predictions were closest to actual conditions from 20 to 50 years after regeneration. The model performed well in predicting trends in habitat variables such as canopy height and closure. Departures of simulated conditions from the actual were greatest early (less than 20 years) and late in stand development (more than 50 years). Poor performance in predicting species composition may be related to excluding management history and site characteristics from the model. The approach exemplified by FORDACK shows good potential for predicting structural characteristics of wildlife habitat.     

A spatial analysis of forest management and its contribution to maintaining the extent of shrubland habitat in southern New England, United States

Conservation organizations in the northeastern United States (US) recommend forest clearcutting to create shrubland habitat, which is required by many wildlife species with declining populations. The planning of habitat management programs is hampered by a lack of information on the current extent of shrubland habitat and the current rate of forest clearcutting that creates shrubland habitat. We addressed these information gaps by using a combination of automated and manual approaches to determine the extent and spatial configuration of shrubland habitat and recent forest clearcuts. We focused on the state of Rhode Island because (a) it is representative of the northeastern US in terms of the prevalence of private ownership of forests, and the ongoing decline in the populations of many shrubland wildlife species; (b) federal, state and private conservation groups are actively promoting clearcuts to create shrubland habitat; (c) many state-wide GIS databases are available; and (d) the spatial extent of the state made our results both generalizable and politically relevant. Our fine-scale mapping allowed a detailed analysis of shrubland distribution in conjunction with other available GIS layers that facilitates identification of priority areas for habitat management. We found that the extent of upland shrubland in non-coastal areas is decreasing by at least 1.5% annually. Considering the lack of consensus about conservation targets for the amount of shrubland, we propose that conservation organizations attempt to stabilize rather than expand the extent of shrubland habitat. This approach would provide an opportunity to assess whether the current extent of shrubland is sufficient to maintain reduced but stable wildlife populations that require this habitat. We propose a coordinated forest management program with targets for increased forest management on conservation lands. We found that the average patch size of shrubland created by recent clearcuts is large enough for most shrubland bird species, but too small for the New England cottontail (Sylvilagus transitionalis), which has been proposed for threatened and endangered status.     

A Process for Improving Wildlife Habitat Models for Assessing Forest Ecosystem Health

Vertebrate wildlife will probably continue to be a primary surrogate for assessing biological diversity in forested ecosystems. However, assessment tools such as wildlife-habitat models generally have proved to be poor predictors of wildlife population responses to landscape-scale changes in forest ecosystems. Forest ecosystem assessment therefore will require improved models. To improve modeling capabilities, scientists must clarify the primary determinants of wildlife habitat selection, which is a behavioral process that links wildlife populations with ecosystem processes. Wildlife populations respond to functional redundancies caused by multiple interactions among landforms, soils, and vegetation. Therefore, probing wildlife habitat selection responses to attributes of landforms, soils, and vegetation should result in improved wildlife-habitat models. In this paper, radiotelemetry data from a study on northern spotted owls (Strix occidentalis caurina) are used to illustrate how remote sensing and geographic information systems (GIs) analysis might clarify basic determinants of habitat selection.     

A landscape and policy perspective on forest conversion: Long-tailed tit (<Emphasis Type="Italic">Aegithalos caudatus</Emphasis>) and the allocation of deciduous forests in southern Sweden

As resources allocated specifically for conservation are limited, there is a need to ensure conservation policy initiatives lead to effective conservation outcomes. In this study, we investigated the potential conservation benefits from alternative spatial allocations of old deciduous stands to a landscape dominated by coniferous production forests owned primarily by non-industrial private forest owners. As a target species, we used the long-tailed tit (Aegithalos caudatus), a species associated with deciduous forests and known to be sensitive to isolation. We used a previously published model based on empirical data on the occurrence of this species, to assess the probability of occurrence of the bird in a 4,000 km² area in southern Sweden for which we possess detailed spatial GIS data (kNN data) of tree species composition and age. We assessed alternative scenarios where old deciduous forest was allocated with or without respect to distance from existing old deciduous forests. Due to the long-tailed tit’s habitat requirement increasing the amount of old deciduous forests close to existing habitats was the most effective strategy. However, the potential advantages of this strategy may in fact be overturned in favor of the other scenarios if ownership structures and probable uptake rates of policy initiatives are also considered. If a policy initiative is targeted toward owners with properties in close proximity to existing suitable habitat, when compared to if all forest owners are targeted, a higher proportion of owners is needed to participate in order to achieve the same degree of habitat creation for the species. Here, we discuss the potential benefits for effective conservation policy formulation from integrating spatially explicit datasets and detailed ecological knowledge with land-ownership structures and policy uptake scenarios.     

Can habitat prediction models contribute to the restoration and conservation of the threatened tree Abies pinsapo Boiss. in Southern Spain?

We examined the association between habitat variables and the relative impacts of topographic microclimates as a valuable tool for restoration and conservation of Abies pinsapo in southern Spain. We used presence–absence data from A. pinsapo and 79 environmental variables and biomod species distribution models to describe the current and future species habitat across the Sierra de las Nieves Natural Park (southern Spain). A. pinsapo habitat was most strongly associated with microtopographic (solar incidence) and temperature variables, indicating climate-driven changes in microhabitat use. Most of the temperature variation among the study site was attributable to topographic microclimates rather than regional temperature differences, such that differences in microhabitat associations occurred principally between north- and south-facing slopes within the same region. The current potential distribution suggests that around 8.7% (56.44 km²) of the study area is highly suitable for A. pinsapo, with 9.7% (62.84 km²) being moderately suitable. Under different global circulation models and climate change scenarios, the net decrease in suitable habitat is predicted to be 93% of the current distribution by 2040, disappearing altogether by 2099. Our findings also show a sharp reduction of potential restoration areas (1.8% of the current areas). Microclimatic variation generated by the topography offers the microclimate-driven locations of habitat suitability which could shape species’ distribution restoration actions and their responses to environmental change. The approach presented here can provide a rapid assessment of the future conservation status of other important forest tree species in Spain, improving our understanding of the vulnerability of endangered species under climate change, and can be an effective tool for biodiversity conservation, restoration, and management.

     

Comparing regional forest policy scenarios in terms of predicted suitable habitats for the Siberian flying squirrel (Pteromys volans)

Forecasting the potential impacts of forest policies on species of special conservation value is a prerequisite for safeguarding forest biodiversity. In this study, regional forest policy scenarios were compared in terms of predicted habitats suitable for the Siberian flying squirrel (Pteromys volans). To derive both patch- and landscape-scale models to predict species presence in a forest stand, species occurrence data from a systematic field survey covering the whole distribution area of the flying squirrel in Finland and Multi-Source National Forest Inventory data were combined. Then, the Finnish forestry model MELA and the derived occupancy models were applied to predict the quantity of suitable habitats for flying squirrels in three different 50-year policy scenarios. The results confirm that increasing the utilization of felling potential from the level of business-as-usual to the level stated as policy targets in regional forest programs decreases the amount of suitable habitat in the future. However, regional forest programs had a less drastic impact on habitats than maximum sustainable removal, except in two regions. It should be noted that the occupancy models seemed to fail on sites that experts deem to be most suitable for the species. Obviously, there are other factors than forest management affecting presence.     

Habitat assessment for forest dwelling species using LiDAR remote sensing: Capercaillie in the Alps

Large-scale information on habitat suitability is indispensable for planning management actions to further endangered species with large-spatial requirements. So far, remote sensing based habitat variables mostly included environmental and land cover data derived from passive sensors, but lacked information on vegetation structure. This is a serious constraint for the management of endangered species with specific structural requirements. Light detection and ranging (LiDAR), in contrast to passive remote sensing techniques, may bridge this gap in structural information at the landscape scale. We investigated the potential of LiDAR data to quantify habitat suitability for capercaillie (Tetrao urogallus), an endangered forest grouse in Central Europe, in a forest reserve of 17.7 km². We used continuous variables of horizontal and vertical stand structure from first and last pulse LiDAR data and presence–absence information from field work to model habitat suitability with generalized linear models (GLM). The two final habitat suitability models explained the observed presence–absence pattern moderately well (AUC of 0.71 and 0.77) with horizontal structure explaining better than vertical structure. Relative tree canopy cover was the most important variable with intermediate values indicating highest habitat suitability. As such, LiDAR allowed us to translate the results from habitat modeling at the landscape scale to effective management recommendations at the local scale at a level of detail that hitherto was unavailable for large areas. LiDAR thus enabled us to integrate individual habitat preferences at the scale of entire populations and thus offers great potential for effective habitat monitoring and management of endangered species.     

Modelling and mapping the suitability of European forest formations at 1-km resolution

Proactive forest conservation planning requires spatially accurate information about the potential distribution of tree species. The most cost-efficient way to obtain this information is habitat suitability modelling i.e. predicting the potential distribution of biota as a function of environmental factors. Here, we used the bootstrap-aggregating machine-learning ensemble classifier Random Forest (RF) to derive a 1-km resolution European forest formation suitability map. The statistical model use as inputs more than 6,000 field data forest inventory plots and a large set of environmental variables. The field data plots were classified into different forest formations using the forest category classification scheme of the European Environmental Agency. The ten most dominant forest categories excluding plantations were chosen for the analysis. Model results have an overall accuracy of 76%. Between categories scores were unbalanced and Mesophitic deciduous forests were found to be the least correctly classified forest category. The model’s variable ranking scores are used to discuss relationship between forest category/environmental factors and to gain insight into the model’s limits and strengths for map applicability. The European forest suitability map is now available for further applications in forest conservation and climate change issues.     

不同森林管理预案下友好林业局森林景观的动态变化

森林作为最大的陆地生态系统,是生物圈中重要的一个环节,它不仅能够为人类提供木材,而且在涵养水源、保持水土、防风固沙、维护大气成分平衡等方面均发挥着重要的生态作用(代力民等,2005).     

Using multitemporal satellite imagery to characterize forest wildlife habitat: The case of ruffed grouse

Like many similar forest species, ruffed grouse (Bonasa umbellus; hereafter grouse) populations in the central and southern Appalachians (CSA) are strongly affected by forest composition at the landscape scale. Because these populations are in decline, managers require accurate forest maps to understand how stand level characteristics affect the survival and reproductive potentials of individual birds to design management strategies that improve grouse abundance. However, traditional mapping techniques are often labor-intensive and cost-prohibitive. We used a normalized difference vegetation index (NDVI) from each of 8 Landsat images and the digital elevation model (DEM)-derived variables of elevation and aspect in discriminant analyses to classify 7 study areas to 3 overstory classes (evergreen, hardwoods, and oak) and distinguish evergreen and deciduous understories in the CSA, 2000–2002. Overall accuracy was 82.08%, varying from 83.59% for oak to 79.79% for hardwoods overstories. Periods with large phenological differences among classes, particularly early and late spring, were most useful for discriminating overstory vegetation types. Alternatively, winter NDVI in combination with elevation was critical for differentiating evergreen and deciduous understories. Multitemporal image sets used in concert with DEMs provided a cost-effective alternative to hyperspectral sensors for improving wildlife habitat classification accuracy with Landsat imagery. This allowed for enhanced understanding of grouse-habitat relationships and habitat affects on grouse populations that allowed for improved management. With the incorporation of simple adjustments for local forest plant species phenology into the model, it may be used to better classify wildlife habitat of similar species in areas with comparable forest communities and topography. Multitemporal images can also be used to differentiate grassland communities, monitor wetlands, and serve as baseline data for detecting changes in land use over longer temporal scales, making their use in forest wildlife habitat studies cost-justifiable.     

Integrating landscape connectivity in broad-scale forest planning through a new graph-based habitat availability methodology: application to capercaillie (Tetrao urogallus) in Catalonia (NE Spain)

The loss of connectivity of forest landscapes is seriously hindering dispersal of many forest-dwelling species, which may be critical for their viability and conservation. In this context, explicitly incorporating connectivity considerations is an important challenge in current forest planning and management, but as yet there is a lack of operative methods for appropriate decision making in this respect. We describe a new methodology based on graph structures and a habitat availability index (integral index of connectivity) that integrates forest attributes (like habitat quality) and network connectivity in a single measure. We apply this methodology to examine the connectivity of the highly fragmented habitat of capercaillie (Tetrao urogallus) in Catalonia (NE Spain), where the threatened status of this forest bird species calls for landscape-level forest planning solutions. We analyse data on the distribution of capercaillie forest habitat at 1 km spatial resolution obtained from the recent Catalan Breeding Bird Atlas. We determine the functionally connected regions existing within its habitat distribution and identify the forest habitat areas that are more important for the maintenance of overall landscape connectivity for this species. Based on these results, we provide recommendations on certain critical public forests where management oriented to the conservation of capercaillie habitat is more necessary. These results highlight the potential and practical interest of the proposed methodology for successfully integrating landscape connectivity in broad scale forest planning.     

Evaluating the potential of sentinel-2, landsat-8, and irs satellite images in tree species classification of hyrcanian forest of iran using random forest

ABSTRACT

Spatially explicit information on tree species composition of any forest provides valuable information to forest managers as well as to nature conservationists. In this study, the potential of three spaceborne sensors: (1) Landsat-8, (2) Sentinel-2, and (3) IRS-Pansharpened were compared by applying Random Forest (RF) classification algorithms to classify the three most common tree species: Pinus taeda, Alnus spp., and Populus spp., in Hyrcanian forest of Iran. Three RF models with optimized parameters of mtry and ntree were used for the classification of trees species. Based on our Overall Accuracy (OA) and Kappa Coefficient (KC) analysis, IRS-Pansharpened data showed the highest accuracy (OA = 84.9% and KC = 79.7%), followed by Landsat-8 (OA = 78.2% and KC = 70.6%), and Sentinel-2 (OA = 77% and KC = 70%). According to the Mean Decrease in Accuracy (MDA) criterion delivered as an output of RF, the near-IR spectral band was found on the top rank (high variable importance) as compared to all other spectral bands for tree species classification. The findings of this study can be used by the researcher, forest managers, economists and policy and decision makers in the context of sustainable forest management of Hyrcanian forest resources.     

Spatial forest valuation: The role of location in determining attitudes toward payment for ecosystem services policies

Forests provide many benefits (e.g., esthetics, water purification, habitat, carbon sequestration) to humans when they are conserved. However, forest conservation efforts may conflict with other uses, particularly timber harvest. Incentives for forest conservation can be established through payment for ecosystem services (PES) policies. PES policies rely on forest valuation studies which traditionally have valued the whole forest, or characteristics of a forest. Methods that allow variability in estimated forest values over space can provide valuation information at a finer scale and aid in the optimization of PES policies. These spatially explicit values provide information critical to many policy decisions, particularly conservation prioritization and planning.Using moderate resolution satellite remote sensing (i.e., LandSat) and a forest use survey conducted in Kam Cha i, Thailand, methods to estimate forest product valuation for spatially heterogeneous plots are demonstrated. These estimates are produced using geographic information system (GIS) multicriteria decision analysis (MCDA). The MCDA method combines forest use data with associated travel costs and vegetation cover estimates to provide spatially explicit estimates of forest values. This information is then used to compare local use values against potential carbon credit values estimated from the literature. The results show that a majority of the forest plots have higher PES values than for local forest products. Some do not, however, which suggests a possible cause for resistance against potential PES policies implemented in these areas.     

Effects of administrative land-use and technical land-form constraints on timber production at the landscape level

Abstract

This study quantified, across a landscape in Eastern Finland, the influence of administrative land-use and technical land-form constraints on timber production. Spatially explicit data about the nature conservation areas, land use plans and steep slopes were integrated with Multi-source National Forest Inventory (MS-NFI) data. The Finnish forestry model MELA was used in the calculations related to updating forest data and estimating different scenarios of timber production with and without constraints. In the study area, the annual volume of maximum sustainable cutting removal defined for the next 30 years was decreased by one-third due to restrictions. The presented approach could be used, for example, to assess timber availability at the landscape level. Future challenges include ensuring the compatibility of spatially explicit data obtained from different sources, identifying the feasibility of forest management operations in the restricted area, and incorporating near-nature forest management operations in the forest planning system in order to estimate the timber production.     

Effects of management regime on the abundance and nest survival of shrubland birds in wildlife openings in northern New England, USA

Many shrubland bird species are declining in eastern North America and as a result forest managers have used a variety of techniques to provide breeding habitat for these species. The maintenance of permanent “wildlife openings” using prescribed burns or mechanical treatments is a widely used approach for providing habitat for these species, but there have been no studies of the effects of treatment regime on bird abundance and nest survival in managed wildlife openings. We studied shrubland birds in wildlife openings on the White Mountain National Forest (WMNF) in New Hampshire and Maine, USA, during 2003 and 2004. We analyzed bird abundance and nest survival in relation to treatment type (burned versus mowed), treatment frequency, time since treatment, and patch area. We found that wildlife openings provided habitat for shrubland birds that are not present in mature forest. There was relatively modest support for models of focal bird species abundance as a function of treatment regime variables, despite pronounced effects of treatment on habitat conditions. This probably was attributable to the combined effects of complex site histories and bird site fidelity. Overall nest success (52%) was comparable to other types of early-successional habitats in the region, but there were few supported relationships between nest survival and treatment variables. We conclude that wildlife openings provide quality habitat for shrubland birds of high conservation interest as long as managers ensure treatment intervals are long enough to permit the development of woody vegetation characteristic of the later stages of this sere. Also, wildlife openings should be large enough to accommodate the territory sizes of all target species, which was ≥1.2 ha in this study.     

Modeling wildfire risk to northern spotted owl (Strix occidentalis caurina) habitat in Central Oregon,USA

Natural disturbances including wildfire, insects and disease are a growing threat to the remaining late successional forests in the Pacific Northwest, USA. These forests are a cornerstone of the region's ecological diversity and provide essential habitat to a number of rare terrestrial and aquatic species including the endangered northern spotted owl (Strix occidentalis caurina). Wildfires in particular have reduced the amount of late successional forests over the past decade, prompting land managers to expand investments in forest management in an attempt to slow losses and mitigate wildfire risk. Much of the emphasis is focused specifically on late successional reserves established under the Northwest Forest Plan to provide habitat for spotted owls. In this paper, we demonstrate a probabilistic risk analysis system for quantifying wildfire threats to spotted owl habitat and comparing the efficacy of fuel treatment scenarios. We used wildfire simulation methods to calculate spatially explicit probabilities of habitat loss for fuel treatment scenarios on a 70,245 ha study area in Central Oregon, USA. We simulated 1000 wildfires with randomly located ignitions and weather conditions that replicated a recent large fire within the study area. A flame length threshold for each spotted owl habitat stand was determined using the forest vegetation simulator and used to predict the proportion of fires that resulted in habitat loss. Wildfire modeling revealed a strong spatial pattern in burn probability created by natural fuel breaks (lakes and lava flows). We observed a non-linear decrease in the probability of habitat loss with increasing treatment area. Fuels treatments on a relatively minor percentage of the forested landscape (20%) resulted in a 44% decrease in the probability of spotted owl habitat loss averaged over all habitat stands. The modeling system advances the application of quantitative and probabilistic risk assessment for habitat and species conservation planning.     

Temporal mapping of deforestation and forest degradation in Nepal: Applications to forest conservation

Deforestation and forest degradation are associated and progressive processes resulting in the conversion of forest area into a mosaic of mature forest fragments, pasture, and degraded habitat. Monitoring of forest landscape spatial structures has been recommended to detect degenerative trends in forest conditions. GIS and remote sensing play an important role in the generation of such data to identify degraded and deforested areas as well as potential areas for conservation. In this study we analyzed forest degradation and deforestation trends in Chitwan district in Nepal, which contains key habitat elements for wildlife in the region. An artificial neural network was used to predict forest canopy density in five classes using Landsat images of the year 2001. Forest canopy density was predicted with 82% overall accuracy. Except riverine forest, forest area of all other forest types was reduced. Terai Shorea robusta forest, which has high commercial value, showed a loss of 23% between 1976 and 1989 and an overall loss of 15% forest covers between the year 1976 and 2001. Deforestation and forest degradation disproportionately reduced the sizes of the different forest types, a finding that has important management implications. The maps presented in this article could be useful to prioritize limited resources for conservation.     

Point process models for mixed sessile forest stands

• Background   

Growth modelling of complex stands calls for the use of spatially explicit single-tree models. Such models require spatially explicit tree locations as the initial state to run simulations. Given the cost of such data, virtual forest stands, where tree locations are simulated, are generally used as the initial state.