The Hydrological Cycle of Central Asia under Climate and Land Use Changes assessed by a re-gional climate model with novel schemes for irrigation and inland lakes
Physical Geography
Final Report Abstract
Through the HYDRASIA project, an attempt has been made to better understand the complex hydrological cycle in Central Asia (CA) where not many tools are yet available to quantify and assess its sensitivity towards changing land use and future climate scenarios. The development of a climate model approach along with its evaluation and application for the water cycle in CA was the main goal of the HYDRASIA. More specifically, further development and use of a Regional Climate Model (RCM; REMO for the current project) to represent the atmospheric components of hydrological cycle at very high resolution, implementing a sophisticated irrigation scheme and FLAKE module to represent the inland lakes in REMO. The three-dimensional and hydrostatic version of RCM REMO2015 was used to carry out different simulations. The dynamical core of the model is based on Global Climate Model (GCM) ECHAM4, although numerous improvements have been made since. One of the important achievements from HYDRASIA project is the implementation of interactive mosaic-based vegetation (iMOVE) version coupled to REMO over CA for the first time. Earlier, the vegetation in REMO was static and changing on intra-annual scale instead of inter-annual scale. But with the new coupled version, the vegetation in the model can react directly to the simulated climate and weather conditions instead of being static. Before our study over CA, the coupled version was used only over Europe to study the changing land use and land cover and for the comparison of different RCMs. A period of 16-year staring from 2000 to 2016 was simulated with both standalone REMO and its coupled version iMOVE at 50 km and 10 km to analyze whether increased horizontal resolution add value in capturing the climatic process over the region. Various statistical metrices were used along mean bias patterns to evaluate the models. Results showed the better representation of climatic features by both models at high resolution especially the vegetation parameters such as Leaf Area Index (LAI) and temperature by the coupled version REMO-iMOVE. This better simulation is caused by the implementation of more realistic and interactive vegetation and related atmospheric processes which consequently add value to the regional climate model. Furthermore, a novel irrigation scheme was developed and implemented in the state-of-the art RCM REMO to study the effect of irrigation which can modulate the regional climate along with the hydroclimatic extremes such as rainfall duration, frequency etc. over CA. The results after consideration of irrigation in the models show the higher precipitation over the irrigated regions of the domain. The higher precipitation was mostly observed during the summer season, but the biases get reduced when multilayer soil scheme is used which is developed within our working group. The cooling effect because of irrigation was also observed especially in the areas with higher irrigation activity in comparison to the areas with little/no irrigation. The cooling effect occurs because of partitioning of energy fluxes more towards latent heat fluxes instead of sensible heat flux. These features were captured well with our newly developed irrigation scheme. Results from the already implemented are quite promising but further tuning needs to be done. Further development of the irrigation scheme specially to implement it in coupled version REMO-iMOVE having an interactive land surface model with interactive vegetation has been planned for another project over South Asia region.
Publications
-
Performance of a regional climate model with interactive vegetation (REMO-iMOVE) over Central Asia. Theoretical and Applied Climatology, 150(3-4), 1385-1405.
Rai, P.; Ziegler, K.; Abel, D.; Pollinger, F. & Paeth, H.
