Changes In Forest Area Research Articles

Linking Spatial Pattern and Ecological Responses in Human‐Modified Landscapes: The Effects of Deforestation and Forest Fragmentation on Biodiversity

AbstractStudies of forest loss and fragmentation provide clear examples of the linkages between ecological pattern and process. Reductions in forest area lead to higher within‐patch extinction rates, the eventual loss of area‐sensitive species, and declines in species richness and diversity. Forest loss also results in increased isolation of remnants, lower among‐patch immigration rates, and less ‘rescue’ from surrounding populations. Specific responses, however, are sometimes counterintuitive because they depend on life‐history tradeoffs that influence population dynamics and species co‐existence in heterogeneous landscapes, not just forest remnants. Thus, while fragmentation generally favours r‐selected, generalist strategies, such as high dispersal and a wide niche breadth, ecological outcomes may be confounded by species‐specific responses to conditions in the human‐dominated matrix and the ways in which forest edges shape cross‐landscape movements. Given that pressures on global forestlands continue to intensify due to growing population sizes, economic pressures, and needs for space and resources, successfully maintaining or restoring species will necessitate a combination of short‐ and long‐term actions that address both habitat protection and restoration. Doing so will require an interdisciplinary approach that gives adequate attention to the manners by which forest loss and fragmentation affect population dynamics through changes in forest area, isolation, habitat quality, matrix properties, and edge effects as well as the synergistic interactions of fragmentation with climate change, human‐altered disturbance regimes, species interactions and other drivers of species population declines.

Quantifying scaling effects on satellite‐derived forest area estimates for the conterminous USA

We quantified the scaling effects on forest area estimates for the conterminous USA using regression analysis and the National Land Cover Dataset 30 m satellite‐derived maps in 2001 and 1992. The original data were aggregated to: (1) broad cover types (forest vs. non‐forest); and (2) coarser resolutions (1 km and 10 km). Standard errors of the model estimates were 2.3% and 4.9% at 1 km and 10 km resolutions, respectively. Our model improved the accuracies for 1 km by 0.6% (12 556 km2) in 2001 and 1.9% (43 198 km2) in 1992, compared to the forest estimates before the adjustments. Forest area observed from Moderate Resolution Imaging Spectroradiometer (MODIS) 2001 1 km land‐cover map for the conterminous USA might differ by 80 811 km2 from what would be observed if MODIS was available at 30 m. Of this difference, 58% (46 870 km2) could be a relatively small net improvement, equivalent to 1444 Tg (or 1.5%) of total non‐soil forest CO2 stocks. With increasing attention to accurate monitoring and evaluation of forest area changes for different regions of the globe, our results could facilitate the removal of bias from large‐scale estimates based on remote sensors with coarse resolutions.

Effects of land cover changes induced by large physical disturbances on hydrological responses in Central Taiwan

This study analyzes the significant impacts of typhoons and earthquakes on land cover change and hydrological response. The occurrence of landslides following typhoons and earthquakes is a major indicator of natural disturbance. The hydrological response of the Chenyulan watershed to land use change was assessed from 1996 to 2005. Land use changes revealed by seven remote images corresponded to typhoons and a catastrophic earthquake in central Taiwan. Hydrological response is discussed as the change in quantities and statistical distributions of hydrological components. The land cover change results indicate that the proportion of landslide relative to total area increased to 6.1% after the Chi-Chi earthquake, representing the largest increase during the study period. The study watershed is dominated by forest land cover. Comparisons of hydrological components reveal that the disturbance significantly affects base flow and direct runoff. The hydrological modeling results demonstrate that the change in forest area correlates with the variation of base flow and direct runoff. Base flow and direct runoff are sensitive to land use in discussions of distinction. The proposed approach quantifies the effect of typhoons and earthquakes on land cover changes.

Biomass carbon sinks in Japanese forests: 1966-2012

The role of forests in absorbing atmospheric carbon has been recognized under the Kyoto Protocol, which allows signatory countries to use forests as a mitigation option. Although several studies have estimated carbon stock changes in Japanese forests, most only estimate changes up to 1995 or ignore carbon stock changes in natural forests. This study is the first attempt to estimate carbon stock changes in Japanese forests from 1966 to 2012, to coincide with the final year of the Kyoto Protocol's first commitment period. Forest area and growing stock data were analysed. Then, two models for predicting the change in forest area and growing stock were developed. Results showed that most natural forest loss resulted from conversion to plantation forest. The total above-ground and below-ground carbon stock in Japanese forests has been estimated to have increased from 1114.8 TgC in 1966 to 2076.0 TgC in 2012, representing an increase of 20.9 TgC year -1 over the same period. During the first commitment period of the Kyoto Protocol (2008-2012), annual carbon sinks were estimated at 20.1 TgC, of which ∼76.9 per cent were sequestered in plantation forests. Of the 20.1 TgC year -1 , eligible carbon sinks are estimated at 10.2 TgC or ∼78.7 per cent of the maximum or capped amount as allowed under the Marrakesh Accord. Although further effort is needed so that the capped amount of 13 TgC year -1 could be achieved, this study suggests that carbon sinks through forest management could be used to offset industrial carbon emissions.

Estimating Area of Forest Change by Random Sampling of Change Strata Mapped Using Satellite Imagery

Abstract There is worldwide interest in monitoring changes in forest area for carbon budgeting and assessing biodiversity changes. For cost-effective monitoring over large areas, automated mapping is necessary. However, the accuracy of automated mapping is rarely high enough to permit accurate estimation of changes in forest area when there is small change. In this article, we show how accurate area statistics of small change can be achieved by combining automated change mapping with stratified random sampling. The stratified random sampling can be made efficient by exhaustive sampling of a stratum with contiguous areas of apparent change greater than or equal to some lower limit (5 ha in the case presented here) and splitting the forest-to-forest stratum into large and small spectral change. Efficiency of the method is high when the exhaustively sampled stratum contains a high proportion of all real change. A regional example from New Zealand of mapping change in forest area is given. Acceptable accuracy (afforestation is estimated to ±7% of reported afforestation) from acceptable effort (10,702 sample points) is obtained, representing an efficiency gain of 78% (in terms of sample size) over simple random sampling. The method can be applied to national estimation of forest change to the same accuracy with an efficiency gain of approximately 70%. It provides a more rapid alternative to multipurpose forest inventories, which depend on intensive and expensive field data, and has significant implications for carbon accounting at regional and national scales, where rapid production of usable estimates are required for carbon exchange markets.

Impact of changing wood demand, climate and land use on European forest resources and carbon stocks during the 21st century

AbstractWe used the European Forest Information Scenario Model (EFISCEN) to project the development of forest resources for 15 European countries from 2000 to 2100 under a broad range of climate scenarios, which were based on the a1fi, a2, b1 and b2 storylines of the Special Report on Emissions Scenarios of the Intergovernmental Panel on Climate Change. Each climate scenario was associated with consistent land‐use change and wood demand assumptions. Climate change‐induced growth changes were incorporated into the calculations by scaling inventory‐based stem growth in EFISCEN by net primary productivity estimated from the Lund–Potsdam–Jena dynamic global vegetation model. The impact of changes in wood demand, climate and forest area were studied separately, and in combination, in order to assess their respective effects. For all climate scenarios under consideration, climate change resulted in increased forest growth, especially in Northern Europe. In Southern Europe, higher precipitation in spring and the projected increased water‐use efficiency in response to rising atmospheric CO2concentrations mitigated the effects of increasing summer drought. Climate change enhanced carbon sequestration in tree biomass. The climate change‐induced increase in tree growth led to a faster increase in growing stocks compared with the simulation using current climate. As productivity decreased in higher stocked forests, the positive impact of climate change began to level off during the second half of the 21st century in the scenarios where wood demand was low. Afforestation measures had the potential to increase growing stock and annual increment; however, large areas were needed to obtain notable effects. Despite noticeable differences in the growth response between the climate scenarios, changes in wood demand proved to be the crucial driving force in forest resource development. Tree carbon stocks increased by 33–114% between 2000 and 2100 when only changes in wood demand were regarded. Climate change added another 23–31% increase, while changes in forest area accounted for an additional increase of 2–40%. Our results highlight potential future pathways of forest resource development resulting from different scenarios of wood demand, land use and climate changes, and stress the importance of resource utilization in the European forest carbon balance.

Changes in Forest Area Along Stream Networks in an Agricultural Catchment of the Great Barrier Reef Lagoon

Scenes from the series of multispectral sensors on the Landsat satellites were used to map recent changes (between 1972 and 2004) in forest cover within and adjacent to stream networks of an intensively farmed region of the southern Great Barrier Reef catchment (Australia). Unsupervised ISODATA classifications of Tasseled-Cap transformed data (at 57 m ground resolution) mapped forest and cleared areas within 150 m of Pisoneer catchment waterways with 72.2% overall accuracy (K(hat) = 0.469), when adjusted for the size of each class. Although the user's accuracy was higher for the forest class (82.1 +/- 8.4% at alpha = 0.05), large errors of commission (34.2 +/- 8.3%) substantially affected map accuracy for the cleared class. The main reasons for misclassification include: (1) failure to discriminate narrowly vegetated riparian strips; (2) misregistration of scenes; and (3) spectral similarity of ground cover. Error matrix probabilities were used to adjust the mapped area of classes, resulting in a decline of forest cover by 12.3% and increase of clearing by 18.5% (22.4 km(2) change; 95% confidence interval: 14.3-29.6 km(2)) between 1972 and 2004. Despite the mapping errors, Landsat data were able to identify broad patterns of land cover change that were verified from aerial photography. Most of the forest losses occurred in open forest to woodland habitat dominated by Eucalyptus, Corymbia, and Lophostemon species, which were largely replaced by sugarcane cropping. Melaleuca communities were similarly affected, though they have a much smaller distribution in the catchment.

Assessing changes of forest area and shrub encroachment in a mire ecosystem using digital surface models and CIR aerial images

This paper presents an approach to assess increase and decrease (2002–1997) of forest area and other wooded areas in a mire biotope in the Pre-alpine zone of Central Switzerland using logistic regression models and airborne remote sensing data (CIR aerial images, DSM derived from them). The present study was carried out in the framework of the Swiss Mire Protection Program, where increase and decrease of forest areas are key issues. In a first step, automatic DSMs were generated using an image matching approach from CIR aerial images of 1997 and 2002. In a second step, the DSMs were co-registered and normalized using LiDAR data. Tree layers from both years of various levels of detail were then generated combining canopy covers derived from normalized DSMs with a multi-resolution segmentation and a fuzzy classification. On the basis of these tree layers, fractional tree/shrub covers were calculated using explanatory variables derived from these DSMs only. Bias was estimated by analysing the distribution of the fractional model differences. The corrected models reveal a decrease of tree/shrub probability which indicates a decrease of forest and other wooded areas between 1997 and 2002. The models also indicate real shrub encroachment in open mire. The detection of shrub encroachment may be helpful for selective logging purposes for sustainable mire habitat management. The study stresses the importance of high-resolution and high-quality DSMs and highlights the potential of fractional covers for ecological modeling.

Effects of species composition, land surface cover, CO2 concentration and climate on isoprene emissions from European forests

Emissions of isoprene from terrestrial vegetation are known to affect atmospheric chemical properties, like its oxidation capacity or the concentration of tropospheric ozone. The latter is of concern, since besides being a potent greenhouse gas, O(3) is toxic for humans, animals, and plants even at relatively low concentrations. Isoprene-emitting forests in the vicinity of NO(x) pollution sources (like cities) can contribute considerably to O(3) formation, and to the peak concentrations observed during hot summer weather. The biogenic contribution to O(3) concentrations is generally thought to increase in a future, warmer climate--pushing values beyond health thresholds possibly even more frequently and over larger areas--given that emissions of isoprene are highly temperature-dependent but also because of the CO(2) fertilisation of forest productivity and leaf growth. Most projections of future emissions, however, do not include the possible CO(2)-inhibition of leaf isoprene metabolism. We explore the regional distribution of emissions from European woody vegetation, using a mechanistic isoprene-dynamic vegetation model framework. We investigate the interactive effects of climate and CO(2) concentration on forest productivity, species composition, and isoprene emissions for the periods 1981-2000 and 2081-2100. Our projection of future emissions includes a direct CO(2)-isoprene inhibition. Across the model domain, we show that this direct effect has the potential to offset the stimulation of emissions that could be expected from warmer temperatures and from the increased productivity and leaf area of emitting vegetation. Changes in forest species composition that may result from climate change can play a substantial additional role in a region's future emissions. Changes in forest area or area planted in woody biofuels in general are not noticeable in the overall European forest isoprene budget, but--as was the case for changes in species composition--may substantially affect future projections in some regions of the continent.

Hydrological impact assessment of afforestation and change in tree-species composition – A regional case study for the Federal State of Brandenburg (Germany)

Policy changes are often implemented without the assessment of their full environmental impact. We investigate the hydrological effects of changes in forest area triggered by shifts in European agricultural policy, and changes in species composition caused by decisions of regional forest authorities. The scenarios were modelled for the Federal State of Brandenburg (Germany) on a 50 m grid scale using the SWIM (Soil Water Integrated Model) model utilising spatially-explicit land-use patterns. The results suggest a negative impact of afforestation on abandoned arable land (9.4% total state area) on the regional water balance, causing an increase in mean annual evapotranspiration of 3.7% at 100% afforestation when compared to no afforestation. The relatively small annual change covers a much more pronounced seasonal effect, leading to an increase in evapotranspiration by 25.1% in spring with a strong feedback to other hydrological components. In contrast, a change in species composition in existing forest (29.2% total state area) from predominantly Scots Pine to Common Oak decreased the evapotranspiration by 3.4%, accompanied by a much weaker but apparent seasonal pattern. The changes show a high spatial heterogeneity that is masked by a linear mean response for the total state area.

Effects of Forest Regrowth and Urbanization on Ecosystem Carbon Storage in a Rural–Urban Gradient in the Southeastern United States

Forest regrowth after cropland abandonment and urban sprawl are two counteracting processes that have influenced carbon (C) sequestration in the southeastern United States in recent decades. In this study, we examined patterns of land-use/land-cover change and their effect on ecosystem C storage in three west Georgia counties (Muscogee, Harris, and Meriwether) that form a rural–urban gradient. Using time series Landsat imagery data including MSS for 1974, TM for 1983 and 1991, and ETM for 2002, we estimate that from 1974 to 2002, urban land use in the area has increased more than 380% (that is, 184 km2). Most newly urbanized land (63%) has been converted from forestland. Conversely, cropland and pasture area has decreased by over 59% (that is, 380 km2). Most of the cropland area was converted to forest. As a result, the net change in forest area was small over the past 29 years. Based on Landsat imagery and agricultural census records, we reconstructed an annual gridded data set of land-cover change for the three counties for the period 1850 to 2002. These data sets were then used as input to the Terrestrial Ecosystem Model (TEM) to simulate land-use effects on C fluxes and storage for the study area. Simulated results suggest that C uptake by forest regrowth (approximately 23.0 g C m−2 y−1) was slightly greater than the amount of C released due to deforestation (approximately 18.4 g C m−2 y−1), thus making the three counties a weak C sink. However, the relative importance of different deforestation processes in this area changed significantly through time. Although agricultural deforestation was generally the most important C-release process, the amount of C release attributable to urbanization has increased over time. Since 1990, urbanization has accounted for 29% of total C loss from the study area. We conclude that balancing urban development and forest protection is critically important for C management and policy making in the southeastern United States.

Baselines for land-use change in the tropics: application to avoided deforestation projects

Although forest conservation activities, particularly in the tropics, offer significant potential for mitigating carbon (C) emissions, these types of activities have faced obstacles in the policy arena caused by the difficulty in determining key elements of the project cycle, particularly the baseline. A baseline for forest conservation has two main components: the projected land-use change and the corresponding carbon stocks in applicable pools in vegetation and soil, with land-use change being the most difficult to address analytically. In this paper we focus on developing and comparing three models, ranging from relatively simple extrapolations of past trends in land use based on simple drivers such as population growth to more complex extrapolations of past trends using spatially explicit models of land-use change driven by biophysical and socioeconomic factors. The three models used for making baseline projections of tropical deforestation at the regional scale are: the Forest Area Change (FAC) model, the Land Use and Carbon Sequestration (LUCS) model, and the Geographical Modeling (GEOMOD) model. The models were used to project deforestation in six tropical regions that featured different ecological and socioeconomic conditions, population dynamics, and uses of the land: (1) northern Belize; (2) Santa Cruz State, Bolivia; (3) Paraná State, Brazil; (4) Campeche, Mexico; (5) Chiapas, Mexico; and (6) Michoacán, Mexico. A comparison of all model outputs across all six regions shows that each model produced quite different deforestation baselines. In general, the simplest FAC model, applied at the national administrative-unit scale, projected the highest amount of forest loss (four out of six regions) and the LUCS model the least amount of loss (four out of five regions). Based on simulations of GEOMOD, we found that readily observable physical and biological factors as well as distance to areas of past disturbance were each about twice as important as either sociological/demographic or economic/infrastructure factors (less observable) in explaining empirical land-use patterns. We propose from the lessons learned, a methodology comprised of three main steps and six tasks can be used to begin developing credible baselines. We also propose that the baselines be projected over a 10-year period because, although projections beyond 10 years are feasible, they are likely to be unrealistic for policy purposes. In the first step, an historic land-use change and deforestation estimate is made by determining the analytic domain (size of the region relative to the size of proposed project), obtaining historic data, analyzing candidate baseline drivers, and identifying three to four major drivers. In the second step, a baseline of where deforestation is likely to occur–a potential land-use change (PLUC) map—is produced using a spatial model such as GEOMOD that uses the key drivers from step one. Then rates of deforestation are projected over a 10-year baseline period based on one of the three models. Using the PLUC maps, projected rates of deforestation, and carbon stock estimates, baseline projections are developed that can be used for project GHG accounting and crediting purposes: The final step proposes that, at agreed interval (e.g., about 10 years), the baseline assumptions about baseline drivers be re-assessed. This step reviews the viability of the 10-year baseline in light of changes in one or more key baseline drivers (e.g., new roads, new communities, new protected area, etc.). The potential land-use change map and estimates of rates of deforestation could be re-done at the agreed interval, allowing the deforestation rates and changes in spatial drivers to be incorporated into a defense of the existing baseline, or the derivation of a new baseline projection.

The effect of Appalachian mountaintop mining on interior forest

Southern Appalachian forests are predominantly interior because they are spatially extensive with little disturbance imposed by other uses of the land. Appalachian mountaintop mining increased substantially during the 1990s, posing a threat to the interior character of the forest. We used spatial convolution to identify interior forest at multiple scales on circa 1992 and 2001 land-cover maps of the Southern Appalachians. Our analyses show that interior forest loss was 1.75–5.0 times greater than the direct forest loss attributable to mountaintop mining. Mountaintop mining in the southern Appalachians has reduced forest interior area more extensively than the reduction that would be expected based on changes in overall forest area alone. The loss of Southern Appalachian interior forest is of global significance because of the worldwide rarity of large expanses of temperate deciduous forest.

A dynamic bio-economic model for analyzing deforestation and degradation: An application to woodlands in Uganda

This study is an effort at developing a dynamic non-linear programming bio-economic model capable of analyzing deforestation and degradation processes in the Ugandan woodlands. The model is unique in capturing the entire systems behavior affecting deforestation and degradation, approximated through changes in forest area and stand density. Thus the model incorporates biological factors, such as growth, the socio-economic factors and related interactions between them. A matrix growth sub-model for the woodlands, accounting for diameter increment, mortality, recruitment and harvest of trees, was imbedded in the overall model to account for the stock changes over the planning horizon. The socio-economic factors and relations affecting the woodland deforestation and degradation processes are addressed through incorporation of three sets of activities, namely, crop cultivation, livestock raising and wood harvesting for firewood and charcoal. Demand and supply relations for these activities are linked through behavioral, structural and accounting equations. The production, consumption, and sale decisions are assumed to be made simultaneously by the households. Though, the model can be successfully used at different levels and planning horizons, a village is the decision making unit over the period 2003–2020 for the purpose of this study. The net present value of cash flows is taken as a proxy measure for utility that is sought to be maximized under six alternate scenarios, including the base scenario. Under the base scenario, 46.7% woodland clearing for cultivation and 42.1% reduction in biomass density were observed over the model planning horizon. Under the scenarios with improved agricultural yields and increased charcoal prices and taxes, the clearing of available woodland was predicted at 98.3%. Biomass density was observed to decline by about 11.9% and 15.5% with the imposition of wood harvest and charcoal quotas. It was observed that none of the policy interventions in this study resulted in reduced degradation and deforestation, which posits a dilemma to the policy makers and implementers. However, we conclude that a well implemented system for taxes or quota restrictions would enhance sustainable resource use in the remaining woodland areas. These interventions are highly limited by implementation and enforcement problems that need to be properly addressed by policy makers and planners.

Changes in forest area in Africa 19902005

SUMMARY The forest area in Africa is estimated at 635 million hectares, equivalent to 21 percent of the total land area, or about 0.7 hectares per person. It ranges from undisturbed primary forests (an estimated 9 percent) to modified natural forests (87 percent), semi-natural forests (1 percent) and forest plantations (3 percent). The net loss of forests remains alarmingly high at an estimated 4.0 million hectares per annum between 2000 and 2005. However, the rate of loss is decreasing when compared to the 1990s and is lower than previously assessed. These and other findings of the recently concluded Global Forest Resources Assessment 2005 provide a detailed picture of the current status and trends related to forests in Africa. The findings also show that the rate of afforestation is increasing, as is the area of forest designated for the conservation of biological diversity. Yet, African forests remain under threat from population growth and unsustainable forestry and agricultural practices.

Projected changes in mineral soil carbon of European forests, 1990–2100

Forests are a major land use in Europe, and European forest soils contain about the same amount of carbon as is found in tree biomass. Changes in the size of the forest soil carbon pool could have significant impacts on the European carbon budget. We present the first assessment of future changes in European forest soil organic carbon (SOC) stocks using a dedicated process-based SOC model and state-of-the-art databases of driving variables. Soil carbon change was calculated for Europe using the Rothamsted Carbon model using climate data from four climate models, forced by four Intergovernmental Panel on Climate Change (IPCC) emissions scenarios (SRES). Changes in litter input to the soil due to forest management, projected changes in net primary production (NPP), forest age-class structure, and changes in forest area were taken into account. Results are presented for mineral soil only. Under some climate scenarios carbon in forest soils will increase slightly (0.1 to 4.6 Pg) in Europe over the 21st Century, whilst for one scenario, forest SOC stocks are predicted to decrease by 0.3 Pg. Different trends are seen in different regions. Climate change will tend to speed decomposition, whereas increases in litter input due to increasing NPP and changing age-class structure will slow the loss of SOC. Increases in forest area could further enhance the total soil carbon stock of European forests. Whilst climate change will be a key driver of change in forest soil carbon, changes in ageclass structure and land-use change are estimated to have greater effects. Key words: Soil organic carbon, Europe, climate change, forest management, land-use change, Rothamsted Carbon model, EFISCEN model, LPJ model

An economic model of international wood supply, forest stock and forest area change

Wood supply, the link between roundwood removals and forest resources, is an important component of forest sector models. This paper develops a model of international wood supply within the structure of the spatial equilibrium Global Forest Products Model. The wood supply model determines, for each country, the annual forest harvest, the annual change of forest stock and the annual change of forest area. The results suggest that global forest area would decline by 477 million ha between 1999 and 2030, with the largest decline in Asia and Africa. However, global forest stock would increase by 25 billion m3, with the largest increase in Europe, and North and Central America. Higher global harvests and lower prices were predicted than those predicted in the past with exogenous timber supply assumptions.

Deforestation: correlations, possible causes and some implications

Changes in national forest areas during 1990-2000 are contrasted with other variables to illustrate correlations and provoke discussion about possible causes. Twenty-five statistically-significant correlations (including rural population, life expectancy, GDP, literacy, commerce, agriculture, poverty and inflation) are illustrated and a statistical model suggests that good governance, alternative employment opportunities, and payments for environmental services may be effective in combating deforestation. The data suggest that a global forest convention may need to be supported by substantial and carefully-targeted development assistance to foster good governance.

Impact of socio-economic factors on the changes in forest areas in China

This paper finds that China is in the later stages of the U-shape forest resources curve hypothesis. Panel data for 12 years, 1990 through 2001, for 30 provinces of Mainland China are used in the analysis. We develop a national forest areas model, regional models and provincial models for Mainland China. Panel data analysis is used for national and regional model estimations and time series analysis is used for provincial model estimations. It was found that per capita GDP (in 1990 value) has a far greater impact on the changes in forest areas than any other factor. Among the 30 provinces, 21, including Southern provinces, are in the later stages of forest area changes, but some remote provinces, including Inner Mongolia, Hainan, Gansu, and Xinjiang, and some advanced coastal provinces, including Tianjin, Shangdong, and Jiangsu, are still in the earlier stages of the changes in forest areas, i.e., economic development still puts pressure on forest.

Forest landscape pattern in the KwaZulu–Natal midlands, South Africa: 50 years of change or stasis?

Abstract Understanding patterns and processes of habitat change is essential for managing and conserving forest fragments in anthropogenically altered landscapes. Digitized aerial photographs from 1944 and 1996 were examined for changes to the indigenous forest landscape in the Karkloof‐Balgowan archipelago in KwaZulu–Natal, South Africa. Attributes relating to proximate land‐use, patch shape, isolation and position in the landscape were used to determine putative causes of forest change. The total change in forest area was −5.7% (forest covered 6739 ha in 1996). This is contrasted with previous reports for the period 1880–1940 that estimated change in total forest area of up to −80%. Attrition was the predominant process of forest transformation between 1944 and 1996. Despite little overall change in forest area, 786 mostly small (<0.5 ha) forest patches were lost from the landscape, leaving 1277 forest patches in 1996. An increase in patch isolation, but no change in patch cohesion accompanied the changes in forest area. Ignoring patches that were eliminated, 514 patches decreased in area. This was partly a function of patch size, but the conversion of natural grassland to commercial plantation forestry in the matrix also influenced forest decline. Their small size and irregular shape caused forest patches in the region to be vulnerable to edge effects. Core area declined in a negative exponential way with increasing edge width and the total area of edge habitat exceeded that of core habitat at an edge width of only 50 m. Nevertheless, total core area decreased by only 2% (65 ha) between 1944 and 1996 because most of the eliminated patches were small and contained no core area. The large Karkloof forest (1649 ha) is a conservation priority for forest interior species, but the ecological role and biodiversity value of small forest patches should not be overlooked.