The influence of atmospheric drivers, environmental factors, and urban land use on extreme hourly precipitation trends over the CONtiguous United States for 40 years at 4-km resolution (CONUS404)

Abstract

High-resolution datasets provide unique insights into extreme precipitation dynamics, capturing atmospheric, environmental, and anthropogenic influences missed by coarser data. Here, we use the 4 km CONUS404 dataset (1980–2021) to analyze trends in extreme hourly precipitation across the contiguous USA and adjacent regions. Using the 42 highest hourly precipitation values (HP42) from the 42-year dataset, we estimate regression slopes for their annual occurrence and intensity. ANOVA analysis examines the effects of elevation and land use on HP42 trends, while Multiple Linear Regression (MLR) assesses the effects of atmospheric drivers (dew point temperature, El Niño, La Niña, and North Atlantic Oscillation). Positive frequency slopes dominate central and northeastern regions, while decreases occur in the West and Southwest. Magnitude slopes are less spatially consistent but near zero in the high-elevation, arid regions of the West. Dew point temperature (TD) drives magnitude trends, while frequency trends are influenced by TD, La Niña, and the positive North Atlantic Oscillation index. Elevation significantly shapes frequency trends, with higher trends at lower and medium elevations (200–1000 m) and weaker trends above 1500 m. Land use impacts vary with elevation; Urban areas show decreasing frequency across several elevations, while natural land uses such as forests and wetlands often exhibit an increase or stabilization in precipitation trends at various elevations. Aggregating variables to coarser resolutions improves MLR model performance, unveiling significant factors by reducing noise. Hotspot analysis reveals that larger cities (e.g., New York, Los Angeles) have concentrated precipitation hotspots, while smaller cities (e.g., Memphis and Nashville) exhibit scattered trends. The overlap between frequency and magnitude clusters highlights shared drivers, suggesting increased vulnerability in peri-urban areas. These findings underscore the need for adaptive strategies addressing the complex interplay of urbanization, elevation, and climate factors.