Understanding, comparing, and accurately predicting water demand at different spatial scales is an important goal that will allow effective targeting of the appropriate operational and conservation efforts under an uncertain future. This study uses data relating to water consumption available at the household level, as well as postcode locations, household characteristics, and weather data in order to identify the relationships between spatial scale, influencing factors, and forecasting accuracy. For this purpose, a Gradient Boosting Machine (GBM) is used to predict water demand 1–7 days into the future. Results show an exponential decay in prediction accuracy from a Mean Absolute Percentage Error (MAPE) of 3.2% to 17%, for a reduction in group size from 600 to 5 households. Adding explanatory variables to the forecasting model reduces the MAPE up to 20% for the peak days and smaller household groups (20–56 households), whereas for larger aggregations of properties (100–804 households), the range of improvement is much smaller (up to 1.2%). Results also show that certain types of input variables (past consumption and household characteristics) become more important for smaller aggregations of properties, whereas others (weather data) become less important.

@en

Root zone storage capacity forms a crucial parameter in ecosystem functioning as it is the key parameter that determines the partitioning between runoff and transpiration. There is increasing evidence from several case studies for specific plants that vegetation adapts to the critical situation of droughts. For example, trees will, on the long term, try to improve their internal hydraulic conductivity after droughts, for example by allocating more biomass for roots. In spite of this understanding, the water storage capacity in the root zone is often treated as constant in hydrological models. In this study, it was hypothesized that root zone storage capacities are altered by deforestation and the regrowth of the ecosystem. Three deforested sub catchments as well as not affected, nearby control catchments of the experimental forests of HJ Andrews and Hubbard Brook were selected for this purpose. Root zone storage capacities were on the one hand estimated by a climate-based approach similar to Gao et al. (2014), making use of simple water balance considerations to determine the evaporative demand of the system. In this way, the maximum deficit between evaporative demand and precipitation allows a robust estimation of the root zone storage capacity. On the other hand, three conceptual hydrological models (FLEX, HYPE, HYMOD) were calibrated in a moving window approach for all catchments. The obtained model parameter values representing the root zone storage capacities of the individual catchments for each moving window period were then compared to the estimates derived from climate data for the same periods. Model- and climate-derived estimates of root zone storage capacities both showed a similar evolution. In the deforested catchments, considerable reductions of the root zone storage capacities, compared to the pretreatment situation and control catchments, were observed. In addition, the years after forest clearing were characterized by a gradual recovery of the root zone storage capacities, converging to new equilibrium conditions and linked to forest regrowth. Further trend analysis suggested a relatively quick hydrological recovery between 5 and 15 years in the study catchments. The results lend evidence to the role of both, climate and vegetation dynamics for the development of root zone systems and their controlling influence on hydrological response dynamics. @en