Spatiotemporal Modeling of Net Primary Productivity of Eastern Mediterranean Biomes under Different Regional Climate Change Scenarios

The present study modeled how future terrestrial net primary productivity (NPP) changes spatiotemporally for the eastern Mediterranean biomes of Turkey using Carnegie Ames Stanford Approach (CASA) model, Moderate Resolution Imaging Spectroradiometer (MODIS) data, the four regional climate change scenarios (RCP 2.6, RCP 4.5, RCP 6.0 and RCP 8.5), and such ancillary data as percent tree cover, land use/cover map, soil texture, and normalized difference vegetation index (NDVI). A mean decrease occurred by 3.2% in the present (2000-2010) mean annual NPP of the most productive biome-deciduous broadleaf forest-in response to the average increases by 5oC in maximum temperature, by 2oC in minimum temperature, by 276 mm in maximum rainfall, and the average decrease by 69 mm in minimum rainfall among the RCPs for the future (2070-2100) period. The maximum annual NPP increases occurred by 4.4% for evergreen needleleaf forest, by 3.9% for grassland, 3.4% for cropland, 2.2% for mixed forest, and 1.1% for shrubland in response to RCP 8.5. Deciduous broadleaf forest NPP appeared to be more vulnerable than the other biomes to the decreased rainfall and the increased air temperature of the water-limited growing season projected by the RCPs. Increasing winter and spring temperatures appeared to benefit the earlier spring green-up of grasses, evergreen needleleaf trees, crops, and shrubs. Our results suggest that a shift from deciduous broadleaf forest toward conifer forest may become more widespread, in particular, in the southern, low-altitude areas of the study region.