The upper and middle reaches of the Heihe Watershed were discre-tized into a grid network of 9,790 cells at 4 km2 resolution. Multiple databases of DEM (at 100 m resolution), land use/cover for the year 2000, meteorological and hydrological databases for 1978 to 2000 were provided by The Chinese Academy of Sciences (CAS) Cold and Arid Regions Environmental and Engineering Research Institute (CAREERI). These databases were used to derive relevant input vari-ables for the DLBRM using the AVDLBRM interface for each grid cell (Croley and He 2005; He and Croley 2007a). Since the soil database of 1999 (1:250,000) from the Gansu Province only contained soil types, we compiled both soil survey data collected by Xiao and his group (2006) and SPAW (Soil - Plant - Atmosphere – Water) Field & Pond Hydrology model (by the U.S. Department of Agriculture Agricultural Service and Natural Resources Conservation Service) to determine relevant soil attributes for each of the soil types. Such attributes in-clude soil texture, depth of USZ and LSZ, water holding capacity (%) and permeability (cm/hr). Manning’s coefficients were assigned to each cell by the hydrological response units (HRUs), which was determined according to the combination of land use, soil texture, and slope (He and Croley 2007b). Average daily river flow rates (in m3/s) were converted into daily outflow volumes and used to con-duct a systematic search of the parameter space to minimize the root mean square errors (RMSEs) between actual and simulated daily outflow volumes at the watershed outlet (Croley et al. 2005; Croley and He 2006).
Model calibration and verification
The DLBRM was calibrated over the period of 1978-1987 for each of the 9,790 cells (4-km2) at daily intervals. The calibration shows a 0.69 correlation between simulated and observed watershed out-flows and a 0.072 mm/d root mean square error. The ratio of model to actual mean flow was 1.011; and the ratio of model to actual flow standard deviation was 0.68. Over a separate verification period (1990-2000), the model demonstrated a 0.71 correlation between simulated and observed watershed outflows and a 0.006 mm/d Figure 2. Tank cascade schematic of Distributed Large Basin
Runoff Model.
RMSE; the ratio of model to actual mean flow was 1.409; and the ratio of model to actual flow standard deviation was 0.72. The simulated annual water budget (averages of the 1990-2000) shows that annual surface net supply from both rainfall and snow melt was about 8.92 billion (109) m3 (Figure3), which mainly came from Qilian Mountain in the upper reach area. The USZ stored about 391 billion m3 of wa-ter, the largest storage among all the four storage tanks (USZ, LSZ, groundwater zone, and surface storage). Surface runoff from the USZ averaged about 0.54 billion m3, while a much larger portion of water (8.37 billion m3) percolated down to the LSZ. A majority (94%) of the percolated water evaporated to the atmosphere from the LSZ and the rest flowed to the stream in the form of interflow. There was hardly any deep percolation to the groundwater since the LSZ is up to 200 m deep in much of the middle reach area (Pan and Qian 2001). The average annual outflow at the outlet (Zhengyixia) of the middle reach was about 1.05 billion m3 to the downstream (Figure3)
Discussion
The Qilian Mountain (up to 5,500 m above sea level) makes up the upper reach of the Heihe Watershed. Due to the high altitude and steep slope of the mountain area, much of the snow melt and rainfall becomes surface runoff. Once reaching the mountain outlet (Yin-gluoxia Station), the water quickly percolates to the deep, coarse sandy and loamy soils in the alluvial fan (up to 200 m deep) which is the main agricultural oasis in the middle reach (between the mountain outlet at Yingluoxia Station and the middle reach outlet at Zhengyixia Station) (Cheng et al. 1999). As annual precipitation in the oasis is less than 200 mm, the majority of the river flow is used to irrigate crops like spring wheat, corn and rice in the oasis, depleting river flow downstream of Zhangye City (Figure1). As the LSZ is up to a few hundred meters deep, there is hardly any deep percolation to the groundwater. Instead, a portion of the water in the LSZ flows to the river channels though interflow. This simulated phenomenon is similar to the findings by other researchers such as Cheng et al.
(1999), Pan and Qian (2001), and Jia et al. (2005). Groundwater
re-charge is only observable in the middle reach area with groundwater level less than 5 m deep and daily precipitation more than 10 mm.
Irrigation return flows first percolate to the groundwater zone and then flow to the river channel in certain middle reach areas.
The simulated average annual flow for 1990-2000 was about 1.05 x 109 m3 at the middle reach outlet (at Zhengyixia Station) under a normal precipitation year (P=50%). But the annual river flow was simulated to change from 0.80 x 109 m3 in 1991 (a dry year, P=75%) to 1.27 x 109 m3 in 1998 (a wet year, P=20%). It appears that under normal climatic years, the amount of flow passing the middle reach outlet is slightly over 1 x 109 m3, satisfying the requirement of deliv-ering 0.95 x 109 m3 downstream annually by the State Council. This amount of flow, however, only has an exceedence probability of 50 percent, meaning that the annual river flow is less than 1 x 109 m3 at Zhengyixia Station 50 percent of the time. In addition, this study excluded the impacts of irrigation in the middle reach to river flow in the simulations. In reality, increased withdrawals for agricultural irri-gation and urban supplies in the middle reach oasis, particularly after the 1980s, have significantly depleted the river flow downstream most of the time each year, shrinking the area of oasis downstream, damaging the aquatic ecosystem, drying up the West Juyan Lake, and causing the expansion of desert in the lower reach. It seems likely that these irrigation withdrawals are contributing to the high ratio of modeled to actual mean annual flow volumes of 1.4 achieved in the model calibration for the 1990-2000 model verification period.
Implementation of the State Council’s water allocation plan requires taking about 0.58 x 109 m3 water out of irrigation each year in the middle reach in order to deliver 0.95 x 109 m3 of water at Zhenyixia Station for rehabilitating the downstream ecosystem (Pan and Qian 2001; The City of Zhangye 2004). This goal seems achievable during normal climatic conditions and governmental entities (e.g. Cities of Zhangye and Jiuquan) in the middle reach, while coping with an annual economic loss of about $240 million, are taking a number of actions such as adjusting crop patterns, water pricing, and market transfer to deliver more water downstream. Between 2000 and 2008, for example, rice planting area was reduced by over 1,000 ha to cut
down irrigation withdrawals for delivering more water downstream (The City of Zhangye 2009). But under dry years, a significantly lesser amount of flow would be available at the Zhengyixia Station, making it much more problematic to deliver the targeted 0.95 x 109 m3 of water downstream.
Figure 3. Annual water budget (1990-2000 average in 109 m3) of the Upper-Middle Reaches of the Heihe Watershed
Conclusions
This paper simulated the hydrology of the Heihe Watershed, the 2nd largest terminal lake in arid Northwest China. The results show that Qilian Mountain in the upper reach area is the main runoff pro-duction area for the entire Heihe Watershed. On average, surface runoff and interflow contributed 51 and 49 percent of the river flow respectively for the period of 1990 to 2000. Annually the river was simulated to discharge slightly more than 1 x 109 m3 of water from the middle reach (at Zhengyixia station) downstream under nor-mal climatic conditions. While requiring a significant reduction in water withdrawals by water users in the middle reach, this amount seems to meet the mandate of delivering 0.95 x 109 m3 of water at Zhengyixia Station for rehabilitation of downstream ecosystems by the State Council. However, the amount of the flow at the middle reach outlet is much less under dry climatic conditions, making it much harder to deliver the required 0.95 x 109 m3 of water down-stream. In addition, climate change and rapid urban expansion will further intensify the water shortage problem in the Heihe Watershed.
Thus, how to develop a comprehensive water management plan to address the competing demands for water among agricultural irrigation, industrial development, urban supplies, and ecosystem protection remains a long term challenge between water users in the upper, middle, and lower reaches of the Heihe Watershed. Our future work will explore the impacts of climate change and human activities on watershed hydrology to support water resource deci-sion making in arid and semi-arid regions of China.
Acknowledgment
Partial support for this research is provided by the National Basic Research Program of Global Change (Grant No. 2010CB951002), the International Partnership Project of the Chinese Academy of Scienc-es, "The Basic Research for Water Issues of Inland River Basin in Arid Region" (CXTD-Z2005-2), and Scherer Endowment Fund of Western Michigan University Department of Geography. The data set used in the study is provided by “Environmental & Ecological Science Data Center for West China, National Natural Science Foundation of China”
(http://westdc.westgis.ac.cn). The assistance of Drs. Xin Li, Haiyang Xi and Luxin Zhai for data acquisition and processing is deeply ap-preciated.
References
Ů Chen, L. and Xiao, H., 2003. Heixi Mountain Soils and Their Ap-plications. Ocean Press, Beijing. P.132
Ů Cheng, G.D., Chang, B.Y., and Qu, Y.G., 1999. Assessment and Rational Use of Water Resources for Coordinated Economic Development and Ecologic and Environmental Protection in the Heihe Watershed. Lanzhou Institute of Glaciology and Geocry-ology. The Chinese Academy of Sciences, “95” National Key Science and Technology Project Report 96-912-03-03, Lanzhou, China, 276 p.
Ů Croley, T. E., II, and He, C., 2005. Distributed-parameter Large Basin Runoff Model I: model development. Journal of Hydrologi-cal Engineering 10(3):173-181.
Ů Croley, T. E., II, He, C. and Lee, D. H., 2005. Distributed-parameter Large Basin Runoff Model II: application. Journal of Hydrological Engineering 10(3):182-191.
Ů Croley, T.E. II. and He, C., 2006. Watershed Surface and Sub-surface Spatial Intraflows. Journal of Hydrological Engineering, 11(1):12-20.
Ů Feng, Q., Cheng, G. and Kunihiko, E., 2002. Towards sustainable development of the environmentally degraded river Heihe basin, China. Hydrological Science Journal 46(5): 647-658.
Ů Fu, G., Chen, S., Liu, C. and Shepard, D., 2004. Hydro-climatic trends of the Yellow River Basin for the last 50 years, Climatic Change, 65:149-178.
Ů He, C., J.F. Riggs and Y.T. Kang., 1993. Integration of geographic information systems and a computer model to evaluate impacts of agricultural runoff on water quality. Water Resources Bulletin, 29: 891-900.
Ů He, C., C. Shi, C. Yang, and B.P. Agosti., 2001. A Windows-based GIS-AGNPS interface. J. of American Water Resources Association.
37(2):395-406.
Ů He, C., 2003. Integration of GIS and simulation model for watershed management. Environmental Modeling and Software 18(8-9):809-813.
Ů He, C., Cheng, S. and Luo, Y., 2005. Desiccation of the Yellow River and the South Water Northward Diversion Project. Water International, 30(2):261-268.
Ů He, C. and Croley, T.E., 2007a. Application of a Distributed Large Basin Runoff Model in the Great Lakes Basin. Control Engineering Practice, 15(8): 1001-1011.
Ů He, C. and Croley, T.E., 2007b. Integration of GIS and Visualization for Distributed Large Basin Runoff Modeling of the Great Lakes Watersheds. In: Scarpati and Jones (eds). Environmental Change and Rational Water Use. Orientación Gráfica Editora S.R.L., Buenos Aires, Argentina, pp. 247-260.
Ů He, C., C. DeMarchi, T. E. Croley, Q. Feng, and T. Hunter, 2009.
Modeling the Hydrology of the Heihe Watershed in Northwest-ern China. Journal of Glaciology and Geocryology, 31 (3):410-421.
Ů Huang, Q. and Zhao, X., 2004. Factors affecting runoff change in the upper reaches of the Yellow River, Progress in Natural Science, 14(9):811-816.
Ů Jia, Y., Wang, H., Nin, G., yang, D., Wang, J., and Qin, Q., 2005. Prin-ciples and Applications of Distributed Watershed Hydrologicalal Models. Water Resources Pressing, Beijing, China. 283 p.
Ů Micklin, P.P., 1994. The Aral Sea Problem. Proc. of Instn. Civ. Engrs Civ. Engng. 102:114-121.
Ů Pan, Q.M. and Qian, S.L., 2001. Water Resources in the Heihe Water-shed. The Yellow River Press. Zhengzhou, P.R. China. 144 p.
Ů Reynolds, J.F., Smith, D.M.S., Lambin, E.F. et al., 2007. Global Desertification: Building a Science for Dryland Development. Sci-ence 316 (2007):847-851. 10.1126/sciSci-ence.1131634.
Ů The City of Zhangye, 2004. Research and Implementation of Wa-ter Saving Programs in the City of Zhangye - Technical Report.
Zhangye, Gansu Province, China.
Ů The City of Zhangye, 2009. Annual Statistics. Zhangye, Gansu Province, China.
Ů U.N. World Water Development Report, 2003. Water for People, Water for Life. UNESCO Publishing and Berghahn Books.
Ů Xiao, H., 2006. Analysis of types and mechanical composition of soil samples (unpublished data). The Chinese Academy of Sciences Cold and Arid Regions Environmental and Engineering Research Institute, Lanzhou, China.
Ů Yang, D., Li, C., Hu, H., Lei, Z., Yang, S., Kusuda , T., Koike, T., and Musiake, K., 2004. Analysis of water resources variability in the Yellow River of China during the last half century using historical data, Water Resources Research, 40, W06502.