WP4: Mapping of ecosystem services with high resolution remote sensing data
Context
Ecosystem goods and services (often abbreviated as "ecosystem services") refer to the benefits that people derive, directly or indirectly, from ecosystem functions (Costanza et al., 1997; MA, 2005). Mountain regions offer a vast range of ecosystem services to livelihood of both highland and lowland communities (Grêt-Regamey et al., 2012). Mountains cover approximately one-fifth of land surface but deliver services to half of humanity (Price et al., 2011). As landscape elements are intimately connected to ecosystem functions (Lovell and Johnston, 2009), global impacts of human-induced land conversions affect a wide range of goods and services that ecosystems provide to answer health, social, cultural and economic needs. The main target of WP4 is to link the provision of ecosystem services with forest cover dynamics. We have selected different services, including carbon storage, biodiversity, water provision and regulation, and natural hazard regulation, which involves here landslide and erosion regulation. We have thus analyzed the effect of forest cover change (both deforestation and afforestation trends) on these services.
Main results
1. Trade-offs between tree cover, carbon storage and biodiversity
This study explores the relationships between an increase in tree cover area (i.e., natural and planted-tree land covers) and changes in forest carbon storage and the potential of a landscape to provide habitat for native floristic biodiversity (Figure 1). Four areas experiencing an increase in tree cover were analyzed. We developed a metric estimating the potential to support native biodiversity based on tree cover type (plantation or natural forests) and the landscape pattern of natural and anthropogenic land covers (Figure 1). We used published estimates for forest and plantation carbon stocks for each region. Focus regions in northwestern Costa Rica, northern Vietnam, southern Chile and highland Ecuador all showed an increase in tree cover area of 390 %, 260 %, 123 % and 418 %, respectively (Figure 1).
Figure 1: Changes in carbon storage with tree cover increase. For each region, the 10 sub-landscapes with the largest increase in tree cover area are represented, along with the mean change vector (bold gray arrow). aTempisque watershed in Guanacaste, Costa Rica, b northern Vietnam, c Valdivia, southern Chile, d Andean intermontane basin of Ecuador. Gray arrows link values for the first and second time steps. For Ecuador,darker gray points represent transitions from páramo to pine plantations, and lighter gray points represent transitions from agriculture to pine plantations
Landscapes experiencing increases in natural secondary forest also experienced an increase in carbon stored above and below ground, and in the potential to support native floristic biodiversity. Study landscapes in Chile and Ecuador experiencing an expansion of exotic plantations saw their carbon stock decrease along with their potential to support native floristic biodiversity (Figure 1). This study shows that an increase in forest area does not necessarily imply an increased provision of ecosystem services when landscapes are reforesting with monoculture plantations of exotic tree species. Changes in the support of native biodiversity and the carbon stored in pulp rotation plantations, along with other ecosystem services, should be fully considered before implementing reforestation projects.
Additionnal information can be found in:
Hall J.M., Van Holt T., Daniels A.E., Balthazar V., Lambin E.F., 2012. Trade-offs between treecover, carbon storage and floristic biodiversity in reforesting landscapes. Landscape Ecology - Vol. 27, Pages 1135-1147.
2. Effects of forest cover changes on water provision and regulation
Rapid land use/-cover change has increasingly transformed the hydrological functioning of tropical Andean ecosystems (Figure 2). The hydrological response to forest cover change strongly depends on the initial state of the ecosystem. Relatively little is known about human-disturbed ecosystems where forest plantations have been established on highly degraded land. In this paper, we analyze the impact of forest change on water and sediment fluxes for a highly degraded Andean catchment (Figure 2). Different pathways of land cover change (1963–2007) are observed in the Jadan catchment, with deforestation taking place in remote uplands and recovery and reforestation in the middle and lower parts where agricultural and bare lands are prevalent.
Figure 2: Changes in hydrological functioning of a degraded Andean basin after forest cover change based on time series analyses of streamflow and rainfall. Reforestation in badlands is associated with a reduction in peakflows and sediment fluxes.
Time series analyses of streamflow and rainfall data (1979/1982–2005/2007) show significant shifts in the distribution of rainfall and flow data. Changes in discharge are not resulting from changes in precipitation, as the direction of change is opposite. The removal of native forest for rangeland or croplands (by −20 km2) is likely to have contributed to the increase in total annual water yield, through an increase in annual baseflow by 25 mm. The observed changes in peakflow are important as the 1st percentile highest flow rates were 54% lower, while the 1st percentile rainfall amounts increased by 52%. The observed decrease in peakflow cannot be explained by clearcut of native forest, but is likely to be related to reforestation of degraded lands as well as spontaneous recovery of vegetation on remaining grazing lands. Over the same time period, a major decrease in specific sediment yields and suspended sediment loads was observed. Although deforestation in the upper parts led to increased landslide activity, this change is not reflected in an increased sediment yield. Small upland rivers are often nearly completely blocked by landslide material, thereby reducing their potential to transport sediment. In contrast, the reduction in estimated erosion is likely to be caused by the reduction of the degraded areas in areal extent as well as to the (partial) recovery of the vegetation in these areas (Figure 2).
Additionnal information can be found in:
Molina A., Vanacker V., Balthazar V., Mora D., Govers G., 2012. Complex land cover change, water and sediment yield in a degraded Andean environment. Journal of Hydrology - Vol. 472–473, Pages 25-35.
3. Effects of forest cover changes on landslide activity
Tropical mountain regions are prone to landslide hazards (Figure 3). Given the current land pressure with increasing occupation of steep uplands, landslide hazards are expected to increase in the near future. Understanding the factors that control landslide hazards is therefore essential. Rare event logistic regression allows us to perform a robust detection of landslide controlling factors. This technique is here applied to the tropical Andes to evaluate the impact of dynamic land cover changes on landslide occurrences (Figure 3). Land cover change trajectories (i.e. dynamic evolution of land cover through time) were specifically included in the probabilistic landslide analysis. While natural physical processes such as slope undercutting by rivers and failure of oversteepened slopes are important in this tropical mountainous site, landslides are increasingly associated with human activities. The data show that land cover trajectories are associated with landslide patterns. In this humid mountainous site, forest degradation does not lead to a measurable increase in landslide occurrence. However, few years after forests are converted to pastures, a rapid decline of slope stability is observed. Land cover conversion from forest to pasture permanently reduces slope stability. It is assumed that major changes in soil properties and hydrology induced by the vegetation conversion play a role in accelerating landslide hazards (Figure 3).
Figure 3: A. Map of land cover trajectories and landslide occurences in Llavircay (n gives the number of landslides for each year) B. Position of the catchment in Ecuador C. Proportion of each trajectory in 2010.
Additionnal information can be found in:
Guns M. and Vanacker V. (2013) Forest Cover Change Trajectories and Their Impact on Landslide Occurrence in the Tropical Andes. Environmental Earth Sciences.
4. Human impact on sediment yield
A regional assessment of the spatial variability in sediment yields allows filling the gap between detailed, process-based understanding of erosion at field scale and empirical sediment flux models at global scale. In this paper, we focus on the intrabasin variability in sediment yield within the Andes and the Ethiopian Highlands as biophysical and anthropogenic factors are presumably acting together to accelerate soil erosion. The two mountain areas are characterized by an important spatial variability in sediment fluxes. Statistical analyses show that 41% of the observed variation in SSY can be explained by remote sensing proxy data of surface vegetation cover, rainfall intensity, mean annual temperature, and human impact (Figure 4). The comparison of a locally adapted regression model with global predictive sediment flux models indicates that global flux models such as the ART and BQART models are less suited to capture the spatial variability in area-specific sediment yields (SSY), but they are very efficient to predict absolute sediment yields (SY) (Figure 4). We developed a modified version of the BQART model that estimates the human influence on sediment yield based on a high resolution composite measure of local human impact (human footprint index) instead of countrywide estimates of GNP/capita. Our modified version of the BQART is able to explain 80% of the observed variation in SY for the Blue Nile and Atbara basins and thereby performs only slightly less than locally adapted regression models (Figure 4).
Figure 4: Calibration and validation results for the site-specific regression model (A and B) and the ART, BQART, and modified BQART model (C and D). Presented are the predicted and observed area-specific (SSY, t/km2/y) and absolute sediment yields (SY, t/y) for 50 catchments in the Blue Nile and Atbara River systems.
5. Trade-offs and synergies between forest cover changes and a set of key mountain ecosystem services
In progress