Abstract
The investigation of regional vulnerability to extreme hydroclimatic events (e.g., floods and hurricanes) is quite challenging due to its dependence on reliable precipitation estimates. Better understanding of past precipitation trends is crucial to examine changing precipitation extremes, optimize future water demands, stormwater infrastructure, extreme event measures, irrigation management, etc., especially if combined with future climate and population projections. The objective of the study is to investigate the spatial-temporal variability of average and extreme precipitation at a sub-regional scale, specifically in the Southern Mid-Atlantic United States, a region characterized by diverse topography and is among the fastest-growing areas in North America. Particularly, this work investigates past precipitation trends and patterns using the North American Land Data Assimilation System, Version 2 (NLDAS-2, 12 km/1 h resolution) reanalysis dataset during 1980–2018. Both parametric (linear regression) and non-parametric (e.g., Theil-Sen) robust statistical tools are employed in the study to analyze trend magnitudes, which are tested for statistical significance using the Mann-Kendall test. Standard precipitation indices from ETCCDI are also used to characterize trends in the relative contribution of extreme events to precipitation in the area. In the region an increasing trend (4.3 mm/year) is identified in annual average precipitation with ~34% of the domain showing a significant increase (at the 0.1 significance level) of +3 to +5 mm/year. Seasonal and sub-regional trends are also investigated, with the most pronounced increasing trends identified during summers along the Virginia and Maryland border. The study also finds a statistically significant positive trend (at a 0.05 significance level) in the annual maximum precipitation. Furthermore, the number of daily extremes (daily total precipitation higher than the 95th and 99th percentiles) also depicts statistically significant increases, indicating the increased frequency of extreme precipitation events. Investigations into the proportion of annual precipitation occurring on wet days and extremely wet days (95th and 99th percentile) also indicate a significant increase in their relative contribution. The findings of this study have the potential to improve local-scale decision-making in terms of river basin management, flood control, irrigation scheme scheduling, and stormwater infrastructure planning to address urban resilience to hydrometeorological hazards.
Highlights
Precipitation is an essential component of Earth’s hydrologic cycle, one that influences complex interactions among elements of the biosphere, such as soil moisture (Sehler et al, 2019), vegetation growth (Chen et al, 2020), and streamflow (Goovaerts, 2000; Lobligeois et al, 2014)
As region-wide annual precipitation trends are not statistically significant, the non-parametric MK test is applied across the study area at the pixel scale to detect trends in annual average precipitation at three different significance levels (i.e., 0.01, 0.05, and 0.10)
This study examines sub-regional changes in precipitation both spatially and temporally across the southern mid-Atlantic region using the gridded NLDAS-2 reanalysis precipitation product
Summary
Precipitation is an essential component of Earth’s hydrologic cycle, one that influences complex interactions among elements of the biosphere, such as soil moisture (Sehler et al, 2019), vegetation growth (Chen et al, 2020), and streamflow (Goovaerts, 2000; Lobligeois et al, 2014). Extreme daily precipitation patterns in the contiguous United States (U.S.) have shown variability (Hoerling et al, 2016). During the period of 1910–1996, Karl and Knight (1998) demonstrate that the intensity of extreme precipitation events (top 10%) increased across the contiguous U.S along with a 53% increase in annual total precipitation. Many studies examine variability at the hourly scale. One such effort by Prein et al (2017) suggests that short duration (hourly) events are projected to increase along with temperatures in the contiguous U.S, especially in areas with abundant moisture. Another study by Brown et al (2020) indicates a significant increase in the 90th percentile hourly accumulation at 36% of recording stations across the southeastern U.S from 1960 to 2017. Continued rainfall increases in already wet regions and continuous drying of dry areas are expected to occur in a warming climate (Chou and Neelin, 2004)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
