Alternative ways to set the blue water footprint cap at sub-catchment level : a case study for the Yellow River basin in China

The growing water demand of the world’s population burden the renewable freshwater or blue water resources globally. The supply of blue water reduces locally and becomes more erratic due to climate change. This combination threatens and has already affected ecosystems in river basins that depend on blue water resources. The consumption of blue water resources is referred to as the blue water footprint (BWF). The ecosystems can be protected by setting an upper limit of blue water consumption which ensures the sustainability of the BWF. The upper limit of sustainable consumption is defined as the natural runoff minus a certain amount of blue water that is reserved for the environment defined as the environmental flow requirements (EFR). The upper limit of blue water consumption can be implemented in a river basin as a policy tool in the form of a BWF cap. Until recently BWF caps were calculated monthly at river basin level. However, this spatial resolution does not capture the spatial variability of consumption or generation of blue water. For the practical implication that captures this variability, the BWF cap needs to be defined at a smaller spatial scale. How to spatially define the BWF cap remains a point of interest. The objective of this study is to examine the effect on the blue water scarcity of defining alternative monthly BWF caps at sub-catchment level. Four alternative allocation principles using two EFR methods for BWF cap setting at sub-catchment level were investigated. The effect of the allocation principles and EFR methods were evaluated by computing monthly blue water scarcity at sub-catchment level in a case study for the Yellow River basin (YRB) from 2010 to 2014. The natural runoff is simulated using the Soil and Water Assessment Tool (SWAT). The SWAT model used physical characteristics and meteorological data of the basin to determine the natural runoff. The blue water availability (BWA) for consumption is calculated for every sub-catchment by subtracting EFRs from the natural runoff. The BWA was allocated over the sub-catchments according to four scenarios for the calculation of the BWF caps: 1) natural conditions and not accounting for other sub-catchments (default scenario), 2) presence of reservoirs and not accounting for other sub-catchments (reservoir scenario), 3) presence of reservoirs and accounting for other sub-catchments based on relative population (population-based scenario) and 4) presence of reservoirs and accounting for other sub-catchments based on relative past demand of blue water in the form of BWF (demand-based scenario). All these four scenarios have been assessed for two different methods to compute EFR: the presumptive standard approach (PRE) and the variable monthly flow method (VMF). Blue water scarcity was most prominently present during spring and summer for every scenario and more so for PRE than VMF. The default scenario showed the largest number of sub-catchments with water scarcity in the highest category due to mismatch in timing between BWA and BWF. The reservoir scenario decreased this number by changing the timing of BWA and its spatial distribution over sub-catchments. The population-based scenario decreased the water scarcity and it was more equally distributed over the sub-catchments. Finally, the demand-based scenario decreased water scarcity to the lowest level over the sub-catchments with smaller regional differences between sub-catchments. The timing of BWA is influenced by its natural temporal and spatial variability, and the operation of reservoirs. An allocation based on the use-what-is-there principle leads to large regional differences in BWF caps and water scarcity, increasing from upstream to downstream. An allocation principle that considers other sub-catchments mitigates regional differences between upstream and downstream BWF caps, and water scarcity. The allocation considering other sub-catchments and based on past demand shows the most equal distribution of water scarcity if the pattern of past demand and actual demand correspond.