Geospatial analysis and quality control of monsoon season precipitation data from citizen reporters near Tucson, Arizona
|Title||Geospatial analysis and quality control of monsoon season precipitation data from citizen reporters near Tucson, Arizona|
|Publication Type||Theses and Dissertations|
|Year of Publication||2019|
|University||University of Arizona|
Half of annual precipitation in Southern Arizona falls in convective thunderstorms associated with the North American monsoon season (June 15 to September 30). Monsoon precipitation varies widely over distances (several km to less than one km) equal to or smaller than the spacing of government rain gauges and the resolution of precipitation raster products (roughly 4 km). A detailed spatial characterization of monsoon precipitation is desirable for emergency responders and flood planners. Denser point precipitation data are available in Arizona through the Rainlog citizen rain gauge network. A geospatial comparison was conducted between daily monsoon precipitation data for the Tucson area from Rainlog stations, the NWS ALERT gauge network, and PRISM rasters to determine the fitness of Rainlog data for direct comparison with institutional data. Data harvesting, quality control and storage in an open-source geodatabase were scripted, and SQL was used within the database to perform spatial queries and regressions. A rule-based algorithm was developed using PRISM raster values at gauge locations to address report timing and other errors in the Rainlog data. The cleaned Rainlog data were subsequently compared with data from nearest neighbor stations and neighborhoods of nearby stations, from both gauge networks, to allow for additional quality control testing. Bias and root mean squared error between data from neighboring gauges were found to be similar for the Rainlog and ALERT gauge networks, provided an adequately long history of reporting by the Rainlog stations. Rainlog data represent a valuable supplement to institutional data in characterizing monsoon precipitation.