A study on urban household water consumption behavior under drought conditions

The increasing frequency of drought events has intensified the risk of water scarcity, posing significant challenges for urban domestic water supply. Reducing urban household water consumption is an important way to alleviate water stress during drought periods. However, due to various factors, it is difficult to determine a water-saving target that is within the residents' capacity. Here, taking Beijing, China as an example, we explored the socio-psychological factors behind urban household water use behaviors under drought conditions, and further quantified the compressible ratio of water quotas for flexible water use behaviors. Therefore, the present study was based on the theory of planned behavior (TPB) and extended TPB (ETPB) by adding drought risk perception as a variable to the theoretical framework. With the help of questionnaire method and structural equation modeling (SEM), the explanatory power of TPB and ETPB in predicting people's water saving intention and behavior was compared. Meanwhile, mathematical statistical analysis methods were employed to calculate the water quota for elastic water consumption behavior and the compressible proportion of urban residents' elastic water consumption under drought conditions. The results showed that drought risk perception has a significant positive correlation with subject norms and water reduction behavior under drought conditions. Furthermore, ETPB was more effective in analyzing water use intentions and behaviors. The predictive explanatory power of SEM for reducing water use increased from 44% to 50% after adding drought risk perception variable. In terms of quantification of elastic water use behavior, the average total water consumption in summer and winter were 71.3L/(p.d) and 52.9L/(p.d) under drought conditions, while it were 124.3 L/(p.d) and 108.9 L/(p.d) under normal conditions. And the compressible proportions of the total water quota for summer and winter elastic water use were 46.7% and 56.8%, respectively. The calculation results can provide a reference for the government to make emergency water supply decisions against drought.