Modelling interactions amongst people and forest resources at the landscape scale

FLORES, the Forest Land Oriented Resource Envisioning System, is a framework to facilitate quantitative modelling of ecological, economic and social issues at the landscape scale. This issue ofSmall-scale Forest Economics, Management and Policy describes the evolution of FLORES from a concept to a series of models calibrated for diverse locations, and documents the lessons learned. The idea to construct and use landscape-scale models of the forest frontier, based on simulating household decisions and land use at a spatial scale close to the field level, arose from a desire to add rigour to land-use policy research at CIFOR, the Center for International Forestry Research (Vanclay 1995). This simulation modelling approach to addressing interdisciplinary issues, where people are strongly interacting with forest resources, became known as FLORES, the Forest Land Oriented Resource Envisioning System (Vanclay 1998). Muetzelfeldtet al. (1998) constructed a simple prototype of a FLORES model to illustrate the concept and demonstrate the ability of a system-dynamics modelling environment to animate such a model (Muetzelfeldt and Taylor 1997, 2001, Muetzelfeldt and Massheder 2003). In 1999, FLORES became a reality, when 50 scientists from diverse disciplines met in Bukittinggi, Indonesia to construct the first FLORES model styled on this prototype (CIFOR 1999, Vanclayet al. 2003). The research reported in this special issue was made possible by generous financial support from the Department for International Development (UK), the European Community, the Asian Development Bank (under RETA 5812) and the Center for International Forestry Research. The views expressed herein are those of the authors alone. We would like to thank the following people who have assisted in the production of this special issue by acting as anonymous referees for these and other papers: Andy Warner, Brendan Moran, Bruno Verbist, Chris Dake, Chris Legg, Doug Sheil, Euan Mason, Francois Bousquet, Frank Vanclay, Geoff Slaughter, James Gambiza, John Herbohn, John Poulsen, Jungho Suh, Mike Spilsbury, Paul Phillips, Paul van Gardingen, Phil Norman, Philip Nyhus, Roger Wheate, Ross Sigley, Steve Harrison and Tom Evans. Thanks also to the editorial staff, especially Steve Harrison, John Herbohn and Jungho Suh, for making this special issue possible.     

Carbon pools and productivity in a 1-km heterogeneous forest and peatland mosaic in Minnesota,USA

Determining the magnitude of carbon (C) storage in forests and peatlands is an important step towards predicting how regional carbon balance will respond to climate change. However, spatial heterogeneity of dominant forest and peatland cover types can inhibit accurate C storage estimates. We evaluated ecosystem C pools and productivity in the Marcell Experimental Forest (MEF), in northern Minnesota, USA, using a network of plots that were evenly spaced across a heterogeneous 1-km² mosaic composed of a mix of upland forests and peatlands. Using a nested plot design, we estimated the standing C stock of vegetation, coarse detrital wood and soil pools. We also estimated aboveground net primary production (ANPP) as well as coarse root production. Additionally we evaluated how vegetation cover types within the study area differed in C storage. The total ecosystem C pool did not vary significantly among upland areas dominated by aspen (160 ± 13 Mg C ha⁻¹), mixed hardwoods (153 ± 19 Mg C ha⁻¹), and conifers (197 ± 23 Mg C ha⁻¹). Live vegetation accounted for approximately 50% of the total ecosystem C pool in these upland areas, and soil (including forest floor) accounted for another 35–40%, with remaining C stored as detrital wood. Compared to upland areas, total C stored in peatlands was much greater, 1286 ± 125 Mg C ha⁻¹, with 90–99% of that C found in peat soils that ranged from 1 to 5 m in depth. Forested areas ranged from 2.6 to 2.9 Mg C ha⁻¹ in ANPP, which was highest in conifer-dominated upland areas. In alder-dominated and black spruce-dominated peatland areas, ANPP averaged 2.8 Mg C ha⁻¹, and in open peatlands, ANPP averaged 1.5 Mg C ha⁻¹. In treed areas of forest and peatlands, our estimates of coarse root production ranged from 0.1 to 0.2 Mg C ha⁻¹. Despite the lower production in open peatlands, all peatlands have acted as long-term C sinks over hundreds to thousands of years and store significantly more C per unit area than is stored in uplands. Despite occupying only 13% of our study area, peatlands store almost 50% of the C contained within it. Because C storage in peatlands depends largely on climatic drivers, the impact of climate changes on peatlands may have important ramifications for C budgets of the western Great Lakes region.     

Modelling long-term impacts of environmental change on mid- and high-latitude European forests and options for adaptive forest management

The process based model SMART–SUMO–WATBAL was applied to 166 intensive monitoring forest plots of mid- and high-latitude Europe to evaluate the effects of expected future changes in carbon dioxide concentration, temperature, precipitation and nitrogen deposition on forest growth (net annual increment). These results were used in the large-scale forest scenario model EFISCEN (European Forest Information SCENario model) to upscale impacts of environmental change and to combine these results with adapted forest management. Because of the few plots available, Mediterranean countries were excluded from analyses. Results are presented for 109 million ha in 23 European countries.     

喀斯特地区低丘缓坡林地适宜性评价

为合理确定低丘缓坡土地开发为林地的适宜性,进而确定开发的空间和时序,选择了地形地貌、土壤和开发利用条件3类10个指标,建立了适用于喀斯特地区低丘缓坡林地适宜性评价指标体系;采用专家打分和层次分析法确定指标权重,并应用GIS技术对喀斯特地区低丘缓坡林地适宜性进行了评价。结果表明:高度适宜区面积为831.01 hm2,适宜区面积3 680.43 hm2,基本适宜区面积10 668.29 hm2,不适宜区面积785.56 hm2。坡度和相对高度是影响林地适宜性等级和空间分布的主导因素。     

Reporting the results of forest monitoring--an evaluation of the European forest monitoring programme

An evaluation of the pan-European forest monitoring programmeset up in the 1980s to examine the effects of atmospheric pollutionon forest condition is presented. This is based on a questionnairesurvey of national stakeholders and an examination of the qualityof scientific information contained in the programme reports.The relevance of the results to different types of stakeholderis discussed. Annual reports from the programme have focussedon scientific understanding, but have only partially satisfiedpolicy needs. Basic principles for an effective communicationstrategy to improve the content, accessibility and usabilityof the monitoring results are put forward. Future monitoringrequires a proper analysis of policy-relevant indicators andmore effort to get appropriate information to policy makerswhen they need it. Modelling and decision support tools shouldbe seen as vital parts of knowledge transfer.     

Predicting ash dieback severity and environmental suitability for the disease in forest stands

Ash dieback, caused by the ascomycete fungus Hymenoscyphus fraxineus, has been rapidly expanding across Europe during the last two decades, posing a considerable threat to native ash populations. In this study, we applied regression-based models trained by field data, in conjunction with geographic information systems, to produce spatial predictions of ash dieback severity and environmental suitability for the disease in Czech forests. A model of actual ash dieback severity relates disease extent to silvicultural and environmental characteristics of forest stands and their neighbourhood, while a model of environmental suitability for the disease quantifies the relative susceptibility of sites to the disease, independent of the current silvicultural characteristics. The final predictive maps suggested that fertile lowlands and humid areas bordering Poland and Slovakia were the most endangered regions. Areas at the lowest risk of damage were concentrated in dry areas and in highland and mountain areas in the western part of the country, usually with poor soils on acid bedrock. Predictions of actual disease severity are an effective tool for guiding the current management of infested stands whereas predicting environmental suitability is useful for making long-term strategic decisions, e.g. identifying areas where future ash regeneration and cultivation may be unsuccessful.     

Influence of fire on woody vegetation of savanna and forest formations in the Cerrado biome

Wildfires have environmental,economic,and social impacts,and can shape the landscape and benefit ecosystems such as the Cerrado.This study evaluated the diversity,similarity,and floristic and structural differences of woody savanna and forest formations of the Cerrado,when affected and not affected by fire.Twenty-eight 25-400m2 plots were randomly allocated and divided into Burnt Cerrado and Unburnt Cerrado,Burnt Forest,and Unburnt Forest,and divided into three levels of inclusion according to d...     

Modelling chemical changes of tidal waters emerging from a mangrove forest at Cananeia, Brazil

Chemical changes in anoxic tidal waters, emerging from the mangrove forest of Cananeia, (48° W, 25° S; Brazil), were interpreted using a stoichiometric approach. Significant depletion of magnesium and sulphates were observed and were apparently due to magnesium-iron replacement in octahedral sheets of clay minerals, and sulphate reduction, respectively. Iron release from clays, after bacterial-mediated reduction, apparently reacted with sulphides produced by sulphate reduction to form nono- and di-sulphides. Production of 41.5% of FeS and 58.5% of FeS₂satisfactorily explained all the stoichiometric relations between concentration changes of magnesium, sulphates, sulphides, iron and alkalinity expected by the model. As a consequence, the decomposition rate of organic matter in these prevailing anoxic conditions, would mainly stem from sulphate reduction (90% against 10% for ferrireduction). The simultaneous bacterial processes of uptake and release of ammonium included in the model would explain why the mangrove forest does not export significant amounts of nitrogen toward the adjacent lagoon.     

Trade in forest products between European Union and the Central and Eastern European access candidates

Patterns in forest products trade between European Union and Central and Eastern European access candidates were studied. The results indicated that both production and trade of forest products in Central and Eastern European countries in transition have considerably increased during the 1990s. The European Union has been the most important trading partner for CEE countries, even though the EU share from the external trade has been quite stable. The gravity models estimated explained approximately 66% of the variation in volume of bilateral trade flows in EU and CEE countries in 1997. Results indicated that trade between EU and CEE countries was below the level that would be expected on the basis of income and distance. A model separating the trade flows according to direction suggested that this is due to the low intensity of West–East trade, which probably originates in low levels of consumption of higher value-added products in CEE countries. East–West trade did not significantly differ from the average pattern.     

Modelling impacts of changes in carbon dioxide concentration,climate and nitrogen deposition on carbon sequestration by European forests and forest soils

Changes in the Earth's atmosphere are expected to influence the growth, and therefore, carbon accumulation of European forests. We identify three major changes: (1) a rise in carbon dioxide concentration, (2) climate change, resulting in higher temperatures and changes in precipitation and (3) a decrease in nitrogen deposition. We adjusted and applied the hydrological model Watbal, the soil model SMART2 and the vegetation model SUMO2 to asses the effect of expected changes in the period 1990 up to 2070 on the carbon accumulation in trees and soils of 166 European forest plots. The models were parameterized using measured soil and vegetation parameters and site-specific changes in temperature, precipitation and nitrogen deposition. The carbon dioxide concentration was assumed to rise uniformly across Europe. The results were compared to a reference scenario consisting of a constant CO₂ concentration and deposition scenario. The temperature and precipitation scenario was a repetition of the period between 1960 and 1990. All scenarios were compared to the reference scenario for biomass growth and carbon sequestration for both the soil and the trees.     

Modelling the interactions between moisture and nutrients in the control of forest growth

Three major approaches to stand-level forest growth-and-yield modelling are briefly described: an empirical predictive approach; a process-based explanatory approach; and a predictive approach which combines both the empirical and the process approaches (the ‘hybrid simulation’ approach). The degree to which models representing these three approaches include an explicit representation of moisture and nutrients is reviewed, and the relationship between modelling objective, time-scale, and the inclusion of representation of moisture and nutrients is explored. There is a brief consideration of the nutrient and moisture-related processes that might be represented in a process-based, rotation-length, explanatory model, and the major processes and parameters that are affected by changes in these factors. The paper concludes with a suggestion as to how the inherent complexity of a process-based approach to simulating moisture might be simplified for use in a predictive hybrid simulation model.

The way in which moisture and nutrients are represented in a forest stand or ecosystem model should be determined by the objectives of the model. There is no single ‘ideal’ approach, but for both explanatory and predictive models used over longer time-scales (one or more rotations), both moisture and nutrients should be explicitly represented. Because of the different time-scales of variation in moisture and nutrient parameters, and because management has a greater potential to affect nutrient parameters than moisture parameters over rotation-length time-scales, moisture can be treated in a more highly aggregated manner than nutrients in long-time-scale models. In contrast, nutrients can probably be omitted from shorter-time-scale models without significantly impairing their performance unless they are to be used to simulate events that effect nutrient availability significantly (e.g., fertilization). The relative importance of representing moisture and nutrients will also depend on the type of site and on the adaptations of the species being simulated. Simulation of moisture should probably take procedence on dry sites and in climates which have significant growing-season moisture-deficits, whereas a detailed simulation of nutrients may be essential for accurate growth-and-yield prediction on fresh to moist sites and in humid climates. Any simulation of the effects of long-term climatic change on forest productivity should include the interactive effects of both moisture and nutrients.     

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.     

Using ecological forest site mapping for long-term habitat suitability assessments in wildlife conservation—Demonstrated for capercaillie (Tetrao urogallus)

Although key factors for vegetation composition and structure, site and soil condition have received little attention as predictors of habitat suitability in wildlife ecology to date. Using the example of capercaillie (Tetrao urogallus), an indicator species for open, well-structured forest habitats, we evaluated the potential use of ecological forest site mapping for the identification of areas where the preferred vegetation structures are supported by the prevailing soil conditions. These are sites that we, therefore, expected to be of long-term relevance to the species.     

Scenario analysis of the impacts of forest management and climate change on the European forest sector carbon budget

Analysis of the impacts of forest management and climate change on the European forest sector carbon budget between 1990 and 2050 are presented in this article. Forest inventory based carbon budgeting with large scale scenario modelling was used. Altogether 27 countries and 128.5 million hectare of forests are included in the analysis. Two forest management and climate scenarios were applied. In Business as Usual (BaU) scenario national fellings remained at the 1990 level while in Multifunctional (MultiF) scenario fellings increased 0.5–1% per year until 2020, 4 million hectare afforestation program took place between 1990 and 2020 and forest management paid more attention to current trends towards more nature oriented management. Mean annual temperature increased 2.5 °C and annual precipitation 5–15% between 1990 and 2050 in changing climate scenario. Total amount of carbon in 1990 was 12 869 Tg, of which 94% in tree biomass and forest soil, and 6% in wood products in use. In 1995–2000, when BaU scenario was applied under current climatic conditions, net primary production was 409 Tg C year⁻¹, net ecosystem production 164 Tg C year⁻¹, net biome production 84.5 Tg C year⁻¹, and net sequestration of the whole system 87.4 Tg C year⁻¹ which was equal to 7–8% of carbon emissions from fossil fuel combustion in 1990. Carbon stocks in tree biomass, soil and wood products increased in all applied management and climate scenarios, but slower after 2010–2020 than that before. This was due to ageing of forests and higher carbon densities per unit of forest land. Differences in carbon sequestration were very small between applied management scenarios, implying that forest management should be changed more than in this study if aim is to influence carbon sequestration. Applied climate scenarios increased carbon stocks and net carbon sequestration compared to current climatic conditions.     

Spatially explicit assessment of forest road suitability for timber extraction and hauling in Switzerland

Efficient forest management, and wood production in particular, requires a forest road network of appropriate density and bearing capacity. The road network affects the choice of a suitable extraction method and the length of the transport route from the forest, while the road standard defines the truck type that can be used.

We evaluate the forest road network’s economic suitability for harvesting operations in the entire Swiss forest, an area of about 13,000 km² covering a range of topographies, based on the Swiss National Forest Inventory’s (NFI) forest road dataset. This dataset is based on information from an interview survey with the local forest services and includes all forest roads in Switzerland capable of carrying trucks. Extraction options and hauling routes are analysed together; thus, the entire logging process is examined.

Model results include maps of the most suitable extraction method; extraction costs; hauling costs; and a suitability map based on a combination of the results. While the larger part of the Swiss forest is classified as “suitable” for economic harvesting operations, significant portions also fall into the “limited suitability” and “not suitable” categories. Our analysis provides an objective, country-wide, spatially explicit assessment of timber accessibility. The resulting suitability map helps identify areas where timber harvesting is economic using the current forest road network, and where it is not. The model results can be used in road network planning and management, for example, by comparing road-network re-design scenarios, and compared to the spatial distribution of available wood volume.

     

广州南香山水库森林水文季节变化研究

为了提高森林水文研究过程中数据采集的精确性,应用现代测绘技术,包括水准测量、全站仪三维坐标测量、GNSS 卫星测量及测深仪的水深测量等,研究了广州南香山水库的库容变化规律,反演水库上游森林径流。研究结果表明：南香山一年内的降雨量分布出现类似双波浪形的变化,波峰出现在5月份和8月份,波谷出现在1月和10月;降雨量与库容增量存在线性关系,相关系数为0.9152;地表径流的大小主要与植被类型,土壤水分的饱和度及地形有密切关系。相比传统的测量方法,现代测绘技术更精确地反映了广州南香山地区森林水文的变化规律。