Mohammad Yunus Naseri
Virginia Polytechnic Institute and State University (Virginia Tech)
| Subject Areas: | civil and environmental engineering, Water resources systems |
Recent Activity
ABSTRACT:
This resource includes city-level aggregated residential water consumption data from single-family households and analysis code accompanying the manuscript "Patterns and Predictors of Residential Indoor Water Use Across Major US Cities". The dataset comprises daily water consumption patterns aggregated from 26,441 single-family households across 39 major US metropolitan statistical areas in the conterminous US. While the original data was collected at 5-second intervals using Flume's smart water monitoring sensors at individual households, this public dataset provides city-level daily aggregations to protect privacy. The data captures both total indoor water use and specific end uses (e.g., shower and toilet), along with aggregated household characteristics (e.g., house size and value), appliance presence (e.g., humidifiers and reverse osmosis systems), and daily climate variables (temperature, precipitation), the latter obtained from the Parameter-elevation Regression on Independent Slopes Model (PRISM). Two Jupyter Notebooks are included: one implementing functional data analysis to identify distinct usage patterns across city clusters, and another executing mixed-effects random forest analysis to investigate the influence of household features, appliances, and weather on water consumption patterns.
Contact
| (Log in to send email) |
| All | 0 |
| Collection | 0 |
| Resource | 0 |
| App Connector | 0 |
Created: Jan. 8, 2025, 2:06 p.m.
Authors: Naseri, Mohammad Yunus · Grant Bernosky · Peter Mayer · Marston, Landon
ABSTRACT:
This resource includes city-level aggregated residential water consumption data from single-family households and analysis code accompanying the manuscript "Patterns and Predictors of Residential Indoor Water Use Across Major US Cities". The dataset comprises daily water consumption patterns aggregated from 26,441 single-family households across 39 major US metropolitan statistical areas in the conterminous US. While the original data was collected at 5-second intervals using Flume's smart water monitoring sensors at individual households, this public dataset provides city-level daily aggregations to protect privacy. The data captures both total indoor water use and specific end uses (e.g., shower and toilet), along with aggregated household characteristics (e.g., house size and value), appliance presence (e.g., humidifiers and reverse osmosis systems), and daily climate variables (temperature, precipitation), the latter obtained from the Parameter-elevation Regression on Independent Slopes Model (PRISM). Two Jupyter Notebooks are included: one implementing functional data analysis to identify distinct usage patterns across city clusters, and another executing mixed-effects random forest analysis to investigate the influence of household features, appliances, and weather on water consumption patterns.