Impacts of landuse changes on water availability, evapotranspiration, cloud and precipitation formation

To investigate the effect of meso-gamma-scale simple or concurrent landuse changes on evapotranspiration, cloud and precipitation formation on the short time-scale (1 d) a non-hydrostatic meteorological model was enlarged by hydro-meteorological modules (e.g., parameterizations of cloud microphysics, soil-vegetation-processes, subgrid-scale heterogeneity, surface and river runoff, etc.). It is shown exemplarily that the enlarged model is able to simulate the typical effects of various landuse types. Applying this model for weather situations during which local recycling of water plays an important role, simulations were performed with different landuse for various regions of North- and East-Germany. The results substantiate that landuse changes hardly affect the domain-averages of temperature and humidity, but that they can appreciably to significantly modify the domain-averages of surface temperature as well as the temporal course and the spatial distributions of evapotranspiration, cloudiness, and precipitation (Mölders 1998). Even small-scale landuse changes can affect the atmospheric boundary layer in their lee as well as, by advection of momentum, water vapor, and cloudy air, even on long-distance. The modified heating and moisture-convergence yield different priorities of the paths of cloud-microphysical processes. Due to the interaction microphysics-dynamics the wind field can also be affected. The modified insolation results in altered distributions of the variables of states, water- and energy-fluxes, in which positive as well as negative feedback can be established between evapotranspiration, clouds and precipitation. The magnitude of the impact of simple or concurrent landuse changes on evapotranspiration, cloud and precipitation depends substantially on the differences in the previous and preceding thermal, hydrologic, and dynamical properties, the size of the individual patches that underwent landuse changes, the resulting heterogeneity, the dominating landuse type in the environs of the landuse changes, as well as on the patterns of the landscape with respect to the direction of the near-surface wind, the direction and velocity of the geostrophic wind. Landuse changes will provide the largest differences if they concurrently modify the dynamics of their environs (Mölders 1999). The effects of concurrent landuse changes have not necessarily to be stronger than those of simple landuse changes. A measure was defined to point out where an enhancement or a reduction occurs. According to this measure wet areas seem to be more sensitive to concurrent landuse changes than dry areas. The results of simulations performed with different parameterizations (and parameters) show that the simulated temporal development and the spatial distributions of evapotranspiration, cloudiness and precipitation can be very sensitive to the kind of parameterizations and the parameters applied. The results of these studies suggest that a detailed knowledge of the soils, their horizons, the actual distributions of soil moisture, surface temperatures, landuse, LAI, albedo, and emissivity (of the plants as well as of the soils) are required as input data as well as a detailed knowledge of the ice-crystal types, and the distributions of the water classes is necessary to evaluate mesoscale models and/or to improve their parameterizations. As small-scale landuse changes affect only slightly the 1. moments, but appreciably to significantly the 2. moments, fluxes should be measured to detect the effects of landuse changes (Mölders 1998).

Acknowledgement: This project was funded by the DFG under contracts Mo770/1-1, Mo770/1-2 and Mo770/2-1.

Future investigation will focus on the impact of landuse changes on the water cycle at longer time scales and on the sensitivity of subarctic weather and climate to landuse changes in the Subarctic itself and outside regions.