Search

Thousands of lakes on the Tibetan Plateau (TP) play a critical role in the regional water cycle, weather, and climate. In recent years, the areas of TP lakes underwent drastic changes and have become a research hotspot. However, the characteristics of the lake-atmosphere interaction over the high-altitude lakes are still unclear, which inhibits model development and the accurate simulation of lake climate effects. The source region of the Yellow River (SRYR) has the largest outflow lake and freshwater lake on the TP and is one of the most densely distributed lakes on the TP. Since 2011, three observation sites have been set up in the Ngoring Lake basin in the SRYR to monitor the lake-atmosphere interaction and the differences among water-heat exchanges over the land and lake surfaces. This study presents an eight-year (2012–19), half-hourly, observation-based dataset related to lake–atmosphere interactions composed of three sites. The three sites represent the lake surface, the lakeside, and the land. The observations contain the basic meteorological elements, surface radiation, eddy covariance system, soil temperature, and moisture (for land). Information related to the sites and instruments, the continuity and completeness of data, and the differences among the observational results at different sites are described in this study. These data have been used in the previous study to reveal a few energy and water exchange characteristics of TP lakes and to validate and improve the lake and land surface model. The dataset is available at National Cryosphere Desert Data Center and Science Data Bank.

Integrated modeling is a critical tool to evaluate the behavior of coupled human–freshwater systems. However, models that do not consider both fast and slow processes may not accurately reflect the feedbacks that define complex systems. We evaluated current coupled human–freshwater system modeling approaches in the literature with a focus on categorizing feedback loops as including economic and/or socio-cultural processes and identifying the simulation of fast and slow processes in human and biophysical systems. Fast human and fast biophysical processes are well represented in the literature, but very few studies incorporate slow human and slow biophysical system processes. Challenges in simulating coupled human–freshwater systems can be overcome by quantifying various monetary and non-monetary ecosystem values and by using data aggregation techniques. Studies that incorporate both fast and slow processes have the potential to improve complex system understanding and inform more sustainable decision-making that targets effective leverage points for system change. Electronic supplementary material The online version of this article (10.1007/s13280-018-1136-6) contains supplementary material, which is available to authorized users.

Environmental and water issues are essentially complex interdisciplinary problems. Multiple models from different disciplines are usually integrated to solve those problems. Integrated modeling environment is an effective technical approach to model integration. Although a number of modeling environments worldwide are available, they cannot meet current challenges faced. Their old-fashion designs and original development purposes constrain their possible applications to the domain of hydrologic or land surface modeling. One of the challenges is that we intend to link knowledge database or ontology system to the modeling environment in order to make the modeling support more intelligent and powerful. In this paper, we designed and implemented an integrated modeling environment (HIME) for hydrological and land surface modeling purpose in a much extendable, efficient and easy use manner. With such design, a physical process was implemented as a module, or component. A new model can be generated in an intuitive way by linking module icons together and establishing their relationships. Following an introduction to the overall architecture, the designs for module linkage and data transfer between modules are described in details. Using XML based meta-information, modules in either source codes or binary form can be utilized by the environment. As a demonstration, with the help of HIME, we replaced the evaporation module of TOPMODEL with the evapotranspiration module from the Noah land surface model which explicitly accounts for vegetation transpiration. This example showed the effectiveness and efficiency of the modeling environment on model integration.