Detected Changes in Precipitation Extremes at Their Native Scales Derived from In Situ Measurements
A specialized spatial extreme value analysis was used to characterize trends in the climatology of extreme precipitation over the contiguous United States. The essence of the method was to first estimate the climatology of extreme precipitation based on station data and then use a data-driven statistical approach to interpolate trend estimates to their native scales, quantify the resulting uncertainty, and robustly determine their statistical significance.
The paper presented trends in extreme precipitation derived exclusively from in situ measurements and used a spatial statistical analysis as a data-driven method for interpolating estimates at weather stations to a high-resolution (0.25 degree) grid over the contiguous United States. The authors furthermore detected statistically significant changes in the climatology of extreme precipitation, finding the most significant changes in the fall and few significant changes otherwise. The high-resolution estimates of trends presented in the paper are an important contribution to the literature in that (1) they are not based on a gridded daily product, and (2) in spite of being based on (sparse) weather station measurements, they resolve the changes to their native scales (instead of simply presenting trends for large spatial averages). Presenting the trends at a high resolution provides important local information that is relevant for impacts.
The gridding of daily accumulated precipitation—especially extremes—from ground-based station observations is problematic due to the fractal nature of precipitation, and therefore estimates of long period return values and their changes based on such gridded daily data sets are generally underestimated. Here, the authors characterized high-resolution changes in observed extreme precipitation from 1950 to 2017 for the contiguous United States (CONUS) based on in situ measurements only. The analysis utilized spatial statistical methods that allowed derivation of gridded estimates that do not smooth extreme daily measurements and are consistent with statistics from the original station data while increasing the resulting signal to noise ratio. Furthermore, the study used a robust statistical technique to identify significant pointwise changes in the climatology of extreme precipitation while carefully controlling the rate of false positives. The paper presented and discussed seasonal changes in the statistics of extreme precipitation: the largest and most spatially-coherent pointwise changes are in fall, with approximately 33% of CONUS exhibiting significant changes. Other seasons displayed very few meaningful pointwise changes, illustrating the difficulty in detecting pointwise changes in extreme precipitation based on in situ measurements.