Among the many ecosystem services that influence human wellbeing, water yield is of great importance as many agricultural, industrial, and domestic activities depend on it [1
]. On one hand, the total amount of water yield influences or restricts the way in which people use water resources [4
]. Additionally, the spatiotemporal variation in water yield is also important, and often leads to the challenge of how to allocate water resources between different seasons, and between upstream and downstream areas [2
]. For arid and semi-arid regions, especially where the climate is highly seasonal, the baseflow that is slowly released by upstream areas due to the interception of vegetation or soil during the rainy season is highly valuable for downstream residents [6
]. However, the quick flow that is soon released after heavy rainfalls also needs to be quantified since it directly contributes to flood and has substantial impact on soil erosion. Therefore, understanding the spatiotemporal variation in water yield including the baseflow, local recharge, and quick flow as well as their driving factors is critical for developing appropriate water resources management strategies [3
Many studies have simulated and quantified water yield services as well as their influencing factors over different temporal and spatial scales using different models and methods. Zomlot et al. [10
] assessed long-term average recharge and baseflow using a spatially-distributed water balance model WetSpass, and then analyzed the controlling factors using correlation analysis method. Ahiablame et al. [9
] analyzed the influence of climate variability and agricultural land use change on annual baseflow, which was estimated through the Web-based Hydrograph Analysis Tool (WHAT). Jiang et al. [11
] assessed changes in multiple ecosystem services in the Three-River Headwaters Region using the annual water yield model in InVEST (Integrated Valuation of Environmental Services and Tradeoffs), showing that the increase in precipitation significantly enhanced the water yield and soil erosion. Wang et al. [12
] assessed the impact of land use and land cover change on streamflow based on the SWAT (Soil and Water Assessment Tool), and found that streamflow decreased on agricultural land, but increased in forest areas.
Although there are already many well developed hydrology models such as SWAT and VIC (Variable Infiltration Capacity) [13
], there is still a huge need for hydrology models that can provide quick and spatially explicit assessment on seasonal water yield to better guide land and water resources management. As one of the most widely used ecosystem service assessment tools, InVEST first released the seasonal water yield model (SWYM) as a module in its 3.3.1 version in 2016 [6
]. Unlike its previous annual water yield model, SWYM computes spatial indices that quantify the relative contribution of a parcel of land to the generation of both baseflow and quick flow [6
]. It can also be easily used to explore the effects of climate and/or land use change on water yield [15
]. With its simple interface, reduced data needs, and multiple-scenario setting function, SWYM has huge potential in understanding water resources and informing land management practices. However, so far, it has only been used in a handful of studies [15
]. SWYM also lacks empirical validation and sensitivity analysis compared to the annual water yield model, which has been widely used and sufficiently validated in many areas [17
Climate change and human activities are the two main driving factors influencing ecosystems, and their effects are usually intertwined [20
]. The Tibetan Plateau is one of the most sensitive areas to global climate change, with a significant temperature rise of 0.3 °C per decade over 50 years [21
], and the precipitation fluctuates greatly in some areas [23
]. Studies have shown that climate change has significantly affected the grassland quality [20
], which is the main vegetation type in the Tibetan Plateau. Located at the south-eastern part of the Tibetan Plateau, the Lhasa River Basin is the social and economic center of the Tibet Autonomous Region. How water yield changes spatiotemporally and how to better manage water resources under the combined influence of climate change and rapid socioeconomic developments have become an increasingly urgent challenge to the Lhasa River Basin and many other areas in the world.
In this paper, we used the SWYM to assess the spatiotemporal water yield changes of China’s Lhasa River Basin between 1990 and 2015, and analyzed its influencing factors. Specifically, we aimed to: (1) analyze the sensitivity of SWYM inputs and validate the model with observed data; (2) assess the spatiotemporal variation in the baseflow, local recharge, and quick flow from 1990 to 2015; and (3) quantify the contributions of three factors, namely precipitation, land cover, and NDVI (normalized difference vegetation index) on water yield changes, then discuss the underlying driving forces. We highlight the need for decision making to consider the combination of human and climatic drivers when targeting water resources management.
The SWYM has been proven to be an efficient tool for revealing the effects of climate, land cover, and NDVI change on water yield by delivering the spatial results of baseflow, local recharge, and quick flow, which together depicts the seasonal flow characteristics. According to relative contribution analysis, baseflow and local recharge were mainly affected by precipitation and NDVI change in the Lhasa River Basin over the study period, while the quick flow was mainly affected by precipitation. Additionally, land cover change began to exert greater influence after 2010. While climate change and land cover change are widely studied and recognized as two of the main driving forces on ecosystem services, this study showed that the vegetation change, which is usually driven by both climate change and human activities, is also important in terms of water yield. For the Lhasa River Basin, the monitoring and management of hydrological processes should be strengthened, and management strategies that explicitly take into account rapidly changing climate and human activities should be developed.