Linking linear programming and spatial simulation models to predict landscape effects of forest management alternatives

Forest management planners require analytical tools to assess the effects of alternative strategies on the sometimes disparate benefits from forests such as timber production and wildlife habitat. We assessed the spatial patterns of alternative management strategies by linking two models that were developed for different purposes. We used a linear programming model (Spectrum) to optimize timber harvest schedules, then a simulation model (HARVEST) to project those schedules in a spatially explicit way and produce maps from which the spatial pattern of habitat could be calculated. We demonstrated the power of this approach by evaluating alternative plans developed for a national forest plan revision in Wisconsin, USA. The amount of forest interior habitat was inversely related to the amount of timber cut, and increased under the alternatives compared to the current plan. The amount of edge habitat was positively related to the amount of timber cut, and increased under all alternatives. The amount of mature northern hardwood interior and edge habitat increased for all alternatives, but mature pine habitat area varied. Mature age classes of all forest types increased, and young classes decreased under all alternatives. The average size of patches (defined by age class) generally decreased. These results are consistent with the design goals of each of the alternatives, but reveal that the spatial differences among the alternatives are modest. These complementary models are valuable for quantifying and comparing the spatial effects of alternative management strategies.     

Assessing the Effects of Management Alternatives on Habitat Suitability in a Forested Landscape of Northeastern China

Forest management often has cumulative, long-lasting effects on wildlife habitat suitability and the effects may be impractical to evaluate using landscape-scale field experiments. To understand such effects, we linked a spatially explicit landscape disturbance and succession model (LANDIS) with habitat suitability index (HSI) models to assess the effects of management alternatives on habitat suitability in a forested landscape of northeastern China. LANDIS was applied to simulate future forest landscape changes under four management alternatives (no cutting, clearcutting, selective cutting I and II) over a 200-year horizon. The simulation outputs were linked with HSI models for three wildlife species, the red squirrel (Sciurus vulgaris), the red deer (Cervus elaphus) and the hazel grouse (Bonasa bonasia). These species are chosen because they represent numerous species that have distinct habitat requirements in our study area. We assessed their habitat suitability based on the mean HSI values, which is a measure of the average habitat quality. Our simulation results showed that no one management scenario was the best for all species and various forest management scenarios would lead to conflicting wildlife habitat outcomes. How to choose a scenario is dependent on the trade-off of economical, ecological and social goals. Our modeling effort could provide decision makers with relative comparisons among management scenarios from the perspective of biodiversity conservation. The general simulation results were expected based on our knowledge of forest management and habitat relationships of the species, which confirmed that the coupled modeling approach correctly simulated the assumed relationships between the wildlife, forest composition, age structure, and spatial configuration of habitat. However, several emergent results revealed the unexpected outcomes that a management scenario may lead to.     

Forest planning in an Oregon case study: Defining the problem and attempting to meet goals with a spatial-analysis technique

Five major management goals were identified for the upper Grande Ronde River Basin on the Wallowa-Whitman National Forest in northeastern Oregon: to produce high-quality fish habitat, to maintain elk habitat, to restore and maintain forest conditions within the natural range of viability, and to contribute to community economic stability. From the broad goals, specific goals for stream temperature, habitat effectiveness index (HEI), habitat corridors, maintenance of land in late or old seral stages, and a nondeclining even flow of timber were selected. A case study was undertaken in a small watershed that is under typical societal constraints to determine whether one decisionsupport tool, SNAP II+, could evaluate the selected goals in a single planning exercise. Three riparian management strategies and two forest road scenarios were used. The exclusion of harvest and road-building from riparian zones in order to increase habitat protection decreased harvest levels and net present value but maintained preactivity stream temperatures. Other resources were generally maintained within prescribed management levels. Although the technique has limitations (e.g., it does not account for riparian zones in calculations of forage and cover for HEI, and it can use the maximum but not minimum acreage goal for some resources), it shows promise for evaluating management tradeoffs in watershed analysis.This is Paper 3069 of the Forest Research Laboratory, Oregon State University, Corvallis, Oregon.     

Simulating Spatial and Temporal Context of Forest Management Using Hypothetical Landscapes

/ Spatially explicit models that combine remote sensing with geographic information systems (GIS) offer great promise to land managers because they consider the arrangement of landscape elements in time and space. Their visual and geographic nature facilitate the comparison of alternative landscape designs. Among various activities associated with forest management, none cause greater concern than the impacts of timber harvesting on the composition, structure, and function of landscape ecosystems. A timber harvest allocation model (HARVEST) was used to simulate different intensities of timber harvest on 23,592-ha hypothetical landscapes with varying sizes of timber production areas and different initial stand age distributions. Our objectives were to: (1) determine the relative effects of the size of timber production areas, harvest intensity, method used to extract timber, and past timber harvest activity on the production of forest interior and edge; and (2) evaluate how past management (in the form of different initial stand age distributions) constrains future timber production options. Our simulations indicated that the total area of forest interior and the amount of forest edge were primarily influenced by the intensity of timber harvest and the size of openings created by harvest. The size of the largest block of interior forest was influenced most by the size of timber harvests, but the intensity of harvest was also significant, and the size of nontimber production areas was important when harvests were numerous and widely dispersed within timber management areas, as is often the case in managed forests. Stand age-class distributions produced by past harvest activity limited the amount of timber production primarily when group selection was used, but also limited clear-cutting when recent harvest levels were high.KEY WORDS: Simulation modeling; Timber harvest; Historical context; Spatial context; Landscape pattern; Forest interior; Forest edge     

Spatial modeling of harvest constraints on wood supply versus wildlife habitat objectives

We studied the effects of spatial and temporal timber harvesting constraints on competing objectives of sustaining wildlife habitat supply and meeting timber harvest objectives in a boreal mixedwood forest. A hierarchical modeling approach was taken, where strategic and tactical level models were used to project blocking and scheduling of harvest blocks. Harvest block size and proximity, together with short- and long-term temporal constraints, were adjusted in a factorial manner to allow creation of response-surface models. A new measure of the habitat mosaic was defined to describe the emergent pattern of habitat across the landscape. These models, together with multiple linear regression, were used to provide insight on convergence or divergence between spatial objectives. For example, green-up delay (defined as time required before a harvest block adjacent to a previously logged block can be scheduled for harvest) had an adverse effect on the amount of annual harvest area that could be allocated and blocked spatially, and habitat supply responded in an opposite direction to that of wood supply, where caribou, moose wintering, and marten habitat supply increased when harvest blocks were further apart, maximum block size smaller, and both a green-up delay and mesoscale stratification were applied. Although there was no "solution space" free of conflicts, the analysis suggests that application of the mesoscale stratification, together with a diversity of harvest block sizes and a between-harvest block proximity of 250 m, will perform relatively well with respect to wood supply objectives, and at the same time create a less fragmented landscape that better reflects natural forest patterns.     

Financial viability of reforesting reclaimed surface mined lands, the burden of site conversion costs, and carbon payments as reforestation incentives

US Federal law mandates that mined land be returned by mine operators to a condition capable of supporting its pre-mining use or a higher use. Previously forested lands have commonly been reclaimed to hayland/pasture or wildlife habitat, and most of these lands have been abandoned from management and rendered non-productive. This situation has left landowners in the position of converting these reclaimed mined lands to forests at a later date, if they choose to make them economically productive. Such land-use conversion, however, comes with a substantial up-front cost to the landowner, which makes the financial viability of such a conversion questionable. We examine the financial viability of reforestation of these previously reclaimed mine lands by calculating land expectation value (LEV) under a range of conditions that include forest type, site quality, and reforestation intensity. We find that conversion to white pine is viable on higher quality sites under low to moderate interest rates with low or high timber prices, but conversion to mixed hardwoods is only profitable under the high price scenario with low interest rates, and only on higher quality sites. We also consider the implications of a shift in reforestation burden from the landowner to the mine operator, and results suggest that including costs of reforestation as part of the mining operation creates a financially viable forest enterprise for landowners under all scenarios for both white pine and mixed hardwoods. Two forms of carbon payments that could encourage reforestation of previously reclaimed mined lands also are examined: an annual payment based upon the total accumulated carbon found on-site in a given year, and an annual payment based on only the increment of carbon storage each year. Our carbon payment results indicate that annual values of up to $5.17 per ton of carbon stored in hardwoods and $9.39 per ton of carbon stored in pines would be required to make reforestation profitable under the poorest conditions (high interest rates, low prices, and poor quality site) when the payment is based on accumulated on-site carbon, although lower values are required under more favorable scenarios. Payments that are based upon the annual increment of carbon must fall in the range of $8.66–$71.88 per ton of carbon stored in hardwoods and $0–$83.29 per ton of carbon stored in pines to make reforestation financially viable.     

Spatial Modeling of Harvest Constraints on Wood Supply Versus Wildlife Habitat Objectives

We studied the effects of spatial and temporal timber harvesting constraints on competing objectives of sustaining wildlife habitat supply and meeting timber harvest objectives in a boreal mixedwood forest. A hierarchical modeling approach was taken, where strategic and tactical level models were used to project blocking and scheduling of harvest blocks. Harvest block size and proximity, together with short- and long-term temporal constraints, were adjusted in a factorial manner to allow creation of response–surface models. A new measure of the habitat mosaic was defined to describe the emergent pattern of habitat across the landscape. These models, together with multiple linear regression, were used to provide insight on convergence or divergence between spatial objectives. For example, green-up delay (defined as time required before a harvest block adjacent to a previously logged block can be scheduled for harvest) had an adverse effect on the amount of annual harvest area that could be allocated and blocked spatially, and habitat supply responded in an opposite direction to that of wood supply, where caribou, moose wintering, and marten habitat supply increased when harvest blocks were further apart, maximum block size smaller, and both a green-up delay and mesoscale stratification were applied. Although there was no solution space free of conflicts, the analysis suggests that application of the mesoscale stratification, together with a diversity of harvest block sizes and a between-harvest block proximity of 250 m, will perform relatively well with respect to wood supply objectives, and at the same time create a less fragmented landscape that better reflects natural forest patterns.     

Opportunity costs of implementing forest plans

Intellectual concern with the National Forest Management Act of 1976 has followed a course emphasizing the planning aspects of the legislation associated with the development of forest plans. Once approved, however, forest plans must be implemented. Due to the complex nature of the ecological systems of interest, and the multiple and often conflicting desires of user clientele groups, the feasibility and costs of implementing forest plans require immediate investigation. For one timber sale on the Coconino National Forest in Arizona, forest plan constraints were applied and resulting resource outputs predicted using the terrestrial ecosystem analysis and modeling system (TEAMS), a computer-based decision support system developed at the School of Forestry, Northern Arizona University, With forest plan constraints for wildlife habitat, visual diversity, riparian area protection, and soil and slope harvesting restrictions, the maximum timber harvest obtainable was reduced 58% from the maximum obtainable without plan constraints.Former Graduate Student at Northern Arizona University.     

Habitat classification: A comparison using avian species and guilds

Results of breeding bird censuses in 1979 and 1980 were used to compare the relationships of both species and guilds to forest habitats in the White Mountains of New Hampshire. Several age classes of 11 forest cover types were studied: northern hard-woods (Fagus-Betula-Acer), spruce (Picea), spruce-fir (Picea-Abies), birth (Betula), swamp hardwoods (Acer-Pinus-Tsuga), pine (Pinus strobus andP. resinosa), balsam fir (Abies), aspen (Populus tremuloides andP. grandidentata), northern red oak (Quercus), oak-pine (Quercus-Pinus), and hemlock (Tsuga). All types were even-aged; only northern hardwoods had an additional uneven-aged condition. Forest cover types were also pooled to consider generalized habitats: hardwoods, mixed forest, or softwoods. Results of ordinations based on censuses of 74 bird species indicate that foraging guilds are more related to general cover types than are nesting substrate guilds, but bird species reflect habitat differences to a greater degree than do either guild scheme. Also, considerable overlap occurs in bird species distribution between hardwoods and mixed forests; softwoods show little overlap with other types. Discriminant function and classification analyses revealed that bird species composition can be used to correctly classify general forest habitats more accurately (83.8%) than either foraging (63.2%) or nesting substrate guilds (58.4%). These results indicate that, of the habitats studied, avian species compositions are more characteristic than are foraging or nesting substrate guild composition, which tend to be similar across forest habitats.     

Changes in Upland Wildlife Habitat on Farmland in Illinois 1920–1987

/ An index of upland wildlife habitat was developed to investigate patterns and changes in habitat over time, using four years (1920, 1940, 1964, 1987) and the state of Illinois as an example. The index was composed of two subdivisions that described, at the county level, the quantity of wildlife habitat and a third subdivision that described farming disturbances that impacted the quality of the habitat. Data came from the US Census of Agriculture. The first subdivision that reflected quantity of habitat was called the wildlife habitat subdivision and was the sum of percentage woodland on farms, percentage farmland in nonrow crops, and percentage farmland in set-aside programs. The second subdivision that reflected the quantity of habitat was termed the soil-related features subdivision and was the sum of the percentage of farmland that was not highly erodible, the percentage of farmland in soil-protecting crops, and the percentage of farmland in conservation tillage. The third subdivision, reflecting the quality of the habitat, was the farming disturbance subdivision and was the sum of the percentage of grazing and the percentage of land on which fertilizers, pesticides, and herbicides were applied. Overall, major decreases occurred between 1920 and 1987 in the subdivisions reflecting the quantity of wildlife habitat and a major increase occurred in the subdivision associated with farming disturbance, reflecting the intensification of agriculture in the state. However, there was variability throughout the state, with some counties being more favorable to wildlife (as measured by the subdivisions) than others. Most of the changes within the state for the subdivisions reflecting quantity of upland wildlife habitat occurred during 1940 while changes in the farming disturbance subdivision (reflecting habitat quality) occurred in 1964. By 1987, the western and southern parts of Illinois were the most favorable for wildlife as reflected in all three subdivisions. Upland wildlife harvest indices were related to the subdivisions in 1964 and 1987, when harvest indices were available. Cottontail and northern bobwhite harvests were higher in counties with higher amounts of the wildlife habitat subdivision in both years. Cottontail harvest was also higher in counties with lower levels of the farming disturbance subdivision in 1964 and higher levels of soil-related features subdivision in 1987. Indices at the county level have the potential to be used in a multiscale analysis to investigate the impact of policy changes on large-scale areas of the Midwest and to develop regional perspectives of the impacts of agriculture on upland wildlife and their habitats.KEY WORDS: Upland wildlife; Habitat; Agriculture; Illinois     

Shrub-Steppe Early Succession Following Juniper Cutting and Prescribed Fire

Pinus-Juniperus L. (Piñon-juniper) woodlands of the western United States have expanded in area nearly 10-fold since the late 1800’s. Juniperus occidentalis ssp. occidentalis Hook. (western juniper) dominance in sagebrush steppe has several negative consequences, including reductions in herbaceous production and diversity, decreased wildlife habitat, and higher erosion and runoff potentials. Prescribed fire and mechanical tree removal are the main methods used to control J. occidentalis and restore sagebrush steppe. However, mature woodlands become difficult to prescribe burn because of the lack of understory fuels. We evaluated partial cutting of the woodlands (cutting 25–50% of the trees) to increase surface fuels, followed by prescribed fire treatments in late successional J. occidentalis woodlands of southwest Idaho to assess understory recovery. The study was conducted in two different plant associations and evaluated what percentage of the woodland required preparatory cutting to eliminate remaining J. occidentalis by prescribed fire, determined the impacts of fire to understory species, and examined early post-fire successional dynamics. The study demonstrated that late successional J. occidentalis woodlands can be burned after pre-cutting only a portion of the trees. Early succession in the cut-and-burn treatments were dominated by native annual and perennial forbs, in part due to high mortality of perennial bunchgrasses. By the third year after fire the number of establishing perennial grass seedlings indicated that both associations would achieve full herbaceous recovery. Cutting-prescribed fire combinations are an effective means for controlling encroaching late successional J. occidentalis and restoring herbaceous plant communities. However, land managers should recognize that there are potential problems associated with cutting-prescribed fire applications when invasive weeds are present.     

Riparian Zone Management in the Pacific Northwest: Who's Cutting What?

Oncorhynchus sp.), regional governments now restrict timber harvest in riparian forests. I summarize and assess the riparian zone management guidelines of the states of California, Oregon, and Washington (USA) and the province of British Columbia (Canada). Only Oregon and British Columbia protect fish-bearing streams with “no-harvest” zones, and only the wider (20–50 m) no-harvest zones for larger fish-bearing streams in British Columbia are likely to maintain near-natural linkages between riparian and stream ecosystems. All four jurisdictions protect most streams with “management zones” of variable width, in which timber harvest activities are restricted. All the management zone guidelines permit the harvest of the largest conifers from riparian forests and will, if applied over a series of timber harvest rotations (60–80 years), result in the continued removal of potential sources of large woody debris from the region's watersheds. All four jurisdictions require additional protection for streams and watersheds that are severely degraded or (in the United States) contain threatened or endangered species. The governments of the PNW have taken a “manage until degraded, then protect” approach to riparian forest management that is unlikely to maintain or restore the full suite of riparian-stream linkages necessary for lotic ecosystems to function naturally at the stream, watershed, basin, or regional scale.     

Linking Multiple Accounts with GIS as Decision Support System to Resolve Forestry/Wildlife Conflicts

With shrinking forest resources and increasing demands for timber, conflicts between forestry and wildlife become more contentious and more frequent. New decision support methods are needed to address complex multiple land use conflicts. Multiple accounts methods linked with GIS and production models enable us to address trade-offs between timber and non-timber values, thus facilitating the evaluation and comparison of different management scenarios in a rapid and spatially referenced manner. The approach is documented in a case study in the trans-boundary zone between two national parks in British Columbia, where caribou/logging conflicts are widespread.     

FOREST HARVESTING AND WATER: THE LAKE STATES EXPERIENCE1

ABSTRACT: The impact of forests on water has_- been a subject of argument for more than a century. It still is; and many studies conform that there is no single right answer in the debate. In the Lake States, clearcutting natural peatlands will not change annual stream-flow nor will it seriously impact water quality if logging is done on frozen soils. However, clearcutting will cause water tables to fluctuate more, ranging from 9 cm higher to 19 cm lower than in peatlands with mature forests. Clearcutting upland hardwoods or conifers will increase annual strearnflow by 9 to 20 cm (a 30- to 80-percent increase). Streamfiow returns to preharvest levels in 12 to 15 years. Annual peak flows are at least doubled and snowmelt flood-peak increases may persist for 15 years. Water quality is not widely impacted, but operating logging equipment in stream channels will cause channel clogging by filamentous algae and loss of fish habitat. Permanent changes from forest to agricultural and urban land use on two-thirds or more of a watershed will significantly increase the size of flood peaks in the 2- to 30-year return interval storm or snowmelt.     

EFFECTS OF BASIN-SCALE TIMBER HARVEST ON WATER YIELD AND PEAK STREAMFLOW1

ABSTRACT: Streamflow changes resulting from clearcut harvest of lodgepole pine (Pinus contorta) on a 2145 hectare drainage basin are evaluated by the paired watershed technique. Thirty years of continuous daily streamflow records were used in the analysis, including 10 pre-harvest and 20 post-harvest years of data. Regression analysis was used to estimate the effects of timber harvest on annual water yield and annual peak discharge. Removal of 14 million board feet of lodgepole pine (Pinus contorta) from about 526 hectares (25 percent of the basin) produced an average of 14.7 cm additional water yield per year, or an increase of 52 percent. Mean annual daily maximum discharge also increased by 1.6 cubic meters per second or 66 percent. Increases occurred primarily during the period of May through August with little or no change in wintertime streamflows. Results suggest that clearcutting conifers in relatively large watersheds (> 2000 ha) may produce significant increases in water yield and flooding. Implications of altered streamflow regimes are important for assessing the future ecological integrity of stream ecosystems subject to large-scale timber harvest and other disturbances that remove a substantial proportion of the forest cover.     

Simulating secondary succession of elk forage values in a managed forest landscape,western Washington

Modern timber management practices often influence forage production for elk (Cervus elaphus) on broad temporal and spatial scales in forested landscapes. We incorporated site-specific information on postharvesting forest succession and forage characteristics in a simulation model to evaluate past and future influences of forest management practices on forage values for elk in a commercially managed Douglas fir (Pseudotsuga menziesii, PSME)-western hemlock (Tsuga heterophylla, TSHE) forest in western Washington. We evaluated future effects of: (1) clear-cut logging 0, 20, and 40% of harvestable stands every five years; (2) thinning 20-year-old Douglas fir forests; and (3) reducing the harvesting cycle from 60 to 45 years. Reconstruction of historical patterns of vegetation succession indicated that forage values peaked in the 1960s and declined from the 1970s to the present, but recent values still were higher than may have existed in the unmanaged landscape in 1945. Increased forest harvesting rates had little short-term influence on forage trends because harvestable stands were scarce. Simulations of forest thinning also produced negligible benefits because thinning did not improve forage productivity appreciably at the stand level. Simulations of reduced harvesting cycles shortened the duration of declining forage values from approximately 30 to 15 years. We concluded that simulation models are useful tools for examining landscape responses of forage production to forest management strategies, but the options examined provided little potential for improving elk forages in the immediate future.     

Using a GIS model to assess terrestrial salamander response to alternative forest management plans.

A GIS model predicting the spatial distribution of terrestrial salamander abundance based on topography and forest age was developed using parameters derived from the literature. The model was tested by sampling salamander abundance across the full range of site conditions used in the model. A regression of the predictions of our GIS model against these sample data showed that the model has a modest but significant ability to predict both salamander abundance and mass per unit area. The model was used to assess the impacts of alternative management plans for the Hoosier National Forest (Indiana, USA) on salamanders. These plans differed in the spatial delineation of management areas where timber harvest was permitted, and the intensity of timber harvest within those management areas. The spatial pattern of forest openings produced by alternative forest management scenarios based on these plans was projected over 150 years using a timber-harvest simulator (HARVEST). We generated a predictive map of salamander abundance for each scenario over time, and summarized each map by calculating mean salamander abundance and the mean colonization distance (average distance from map cells with low predicted abundance to those with relatively high abundance). Projected salamander abundance was affected more by harvest rate (area harvested each decade) than by the management area boundaries. The alternatives had a varying effect on the mean distance salamanders would have to travel to colonize regenerating stands. Our GIS modeling approach is an example of a spatial analytical tool that could help resource management planners to evaluate the potential ecological impact of management alternatives.     

Cumulative silvicultural impacts on watersheds: A hydrologic and regulatory dilemma

Because of the nature of watersheds, the hydrologic and erosional impacts of logging and related road-building activities may move offsite, affecting areas downslope and downstream from the operation. The degree to which this occurs depends on the interaction of many variables, including soils, bedrock geology, vegetation, the timing and size of storm events, logging technology, and operator performance. In parts of northwestern California, these variables combine to produce significant water quality degradation, with resulting damage to anadromous fish habitat.Examination of recent aerial photographs, combined with a review of public records, shows that many timber harvest operations were concentrated in a single 83 km² watershed in the lower Klamath River Basin within the past decade. The resulting soil disturbance in this case seems likely to result in cumulative off-site water quality degradation in the lower portion of the Basin.In California, both state and federal laws require consideration of possible cumulative effects of multiple timber harvest operations. In spite of recent reforms that have given the state a larger role in regulating forest practices on private land, each timber harvest plan is still evaluated in isolation from other plans in the same watershed. A process of collaborative state-private watershed planning with increased input of geologic information offers the best long-term approach to the problem of assessing cumulative effects of multiple timber harvest operations. Such a reform could ultimately emerge from the ongoing water quality planning process under Section 208 of the amended Federal Water Pollution Control Act.     

The remaining islands with primary rain forest: A global resource

Islands with large tracts of primary rain forest constitute a distinct complex of terrestrial and marine ecosystems with global significance because of high levels of biological richness or, in the case of oceanic islands, the evolutionary distinctiveness of the organisms that they support. The myriad of small habitat units makes these settings particularly prone to fragmentation and thus problematic for the conservation of biological diversity. Most of the remaining examples of islands with primary rain forest are in the Pacific Rim, particularly in Indonesia, and there are threats from intensive timber harvesting, mining, tourism, and dismemberment of traditional convervation systems. Prospects for inventorying, monitoring, and protecting the remaining islands with relatively pristine successional mosaics of humid forest and shallow marine habitats within a framework global monitoring are explored. Recommendations are made for a program of expanded use of the biosphere reserve designation for rain forest islands and adjacent marine areas.