Open Access

Jianhui Wei

• Land-atmosphere interaction analysis still lacks adequate methods for explicitly answering one of the center hydrological question of "what is the fate of evapotranspired water", and likewise, "where, when, and to what extent evapotranspiring water of one region returns as precipitation in the same or another region", and "what are water residence times across the atmosphere". In this thesis, a regional climate model (RCM) based evapotranspiration tagging (ET-Tagging) algorithm has been applied for the first time to the subtropical monsoon and mountainous region of Southeast China. 15 month simulations (October 2004 to December 2005) were performed to investigate where and to which extent the tagged evapotranspired water from a predefined source region around the Poyang Lake (the largest freshwater lake in China) returning to the land surface as precipitation. To assess the relative importance of the individual evapotranspiration components for the ET-driven precipitation, the modelLand-atmosphere interaction analysis still lacks adequate methods for explicitly answering one of the center hydrological question of "what is the fate of evapotranspired water", and likewise, "where, when, and to what extent evapotranspiring water of one region returns as precipitation in the same or another region", and "what are water residence times across the atmosphere". In this thesis, a regional climate model (RCM) based evapotranspiration tagging (ET-Tagging) algorithm has been applied for the first time to the subtropical monsoon and mountainous region of Southeast China. 15 month simulations (October 2004 to December 2005) were performed to investigate where and to which extent the tagged evapotranspired water from a predefined source region around the Poyang Lake (the largest freshwater lake in China) returning to the land surface as precipitation. To assess the relative importance of the individual evapotranspiration components for the ET-driven precipitation, the model has been extended to partition the total evapotranspiration into direct evaporation (consisting of evaporation from inland water bodies, evaporation from the top soil layer, and evaporation of precipitation intercepted by the canopy) and transpiration and to track the two partitions separately. Using this extended ET-Tagging partitioning algorithm, the contributions of direct evaporation and transpiration to precipitation were estimated separately. Since, the atmospheric water residence time is a fundamental descriptor that provides information on the timescales of evapotranspiration (direct evaporation and transpiration) and precipitation, the algorithm has been further extended with an age tracer approach. This newly developed age-weighted ET-Tagging partitioning algorithm allows to calculate the atmospheric residence times of evapotranspired water (of direct evaporated water and of transpired water, separately), and to assess the celerity of the atmospheric branch of the hydrological cycle. The most important results are: In 2005 the contribution of moisture originating from the Poyang Lake region to the annual precipitation in Southeast China reaches a value of up to 1.2%, with a maximum of 6% near the Poyang Lake region in August. A budget analysis indicates that in 2005 the tagged precipitation falling in the source area, in the Poyang Lake basin, and in Southeast China equate to 0.8%, 2.7%, and 10.7% of the evapotranspired water in the Poyang Lake region, respectively. Regarding the ET-Tagging partitioning, in 2005 direct evaporated water accounts for 64% and transpired water for 36% of the total tagged moisture with a mean age of around 36 hours for both. The condensed tagged moisture consists of a large proportion (69.2%) of the young (7 hours) direct evaporated water and of a small amount of the older (11 hours) transpired water. Similarly, 69% of total tagged precipitation originates from direct evaporation of water with a mean atmospheric residence time of 6.6 hours, whereas 31% comes from vegetation transpiration with a longer residence time of 10.7 hours. To explore the source-target relations under consideration of the respective precipitation regime (wet or dry weather conditions), source-specific precipitation efficiencies are introduced. For the period under investigation, the source-specific precipitation efficiency for direct evaporation generally dominates. Only during the comparatively dry August (with short residence times) and in the winter months (with small absolute transpiration values), the source-specific precipitation efficiency for transpiration is larger. This thesis shows large spatial and temporal variations of the magnitude of tagged precipitation. There is a pronounced difference where and how long direct evaporation and transpiration are able to contribute to precipitation. Impacted by the East Asian monsoon, the patterns of the atmospheric water residence times change over the months. The results depict the seasonal variations of the prevailing meteorological conditions and the varying interactions between land surface and atmosphere. The interactions control the annual cycle of the individual contributions to precipitation and the corresponding atmospheric water residence times, emphasizing the important impacts of vegetation cover and land use on the regional hydrological cycle. In conclusion, with the age-weighted ET-Tagging partitioning algorithm developed in this work, the different fate of transpired and direct evaporated water in the atmospheric branch of the hydrological cycle is depicted. This algorithm has the potential to be used for addressing how the hydrological cycle changes and potentially accelerates with climate change.…