Spatial modeling of water consumption on construction sites

Understanding aspects related to water consumption on construction sites is essential for directing actions aimed at water conservation in the urban environment. This study aims to analyze the variables that impact water use on construction sites in Joinville and how these variables manifest in a spatial analysis. Data from 63 construction sites are analyzed using descriptive statistics, correlational (Spearman correlation), and spatial (Moran Index and geographically weighted regression - GWR) techniques. The total water consumption of the sample ranged from 143 m³ to 6865 m³ and showed a positive correlation with the built area, number of floors, number of apartments, and construction time. GWR revealed that the relationship between water consumption and independent variables varied according to location, which the linear regression model did not capture. The research suggests that considering local factors is crucial for the efficient management of water consumption on construction sites since GWR indicated that variables affect consumption differently in each region.