Location: Southwest Watershed Research CenterTitle: Temporally downscaling a precipitation intensity factor for soil erosion modeling using the NOAA-ASOS weather station network
Submitted to: Catena
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/25/2020
Publication Date: 5/25/2020
Citation: Fullhart, A.T., Nearing, M.A., Mcgehee, R., Weltz, M.A. 2020. Temporally downscaling a precipitation intensity factor for soil erosion modeling using the NOAA-ASOS weather station network. Catena. https://doi.org/10.1016/j.catena.2020.104709.
Interpretive Summary: Climate inputs are an important part of soil erosion modeling, and the necessary data cannot always be found. Often we require data with high-frequency measurements, and these are not always easy to obtain. This makes commonly available precipitation records that are on an hourly or daily basis difficult to use for assessing hydrology and erosion. This work hypothesizes that these more commonly available, low-frequency measurement datasets may be used to estimate the intensity parameters that would be found by high-frequency measurements. The estimated intensity parameters are used in a weather generator called CLIGEN that creates climate records that may then be used to drive soil erosion models. The intensity parameters were determined by statistical methods and machine learning algorithms by relating high-frequency measurements to low-frequency measurements that come from the NOAA Automated Surface Observing Systems dataset with stations distributed across the United States. With these findings, erosion models such as the ARS-Rangeland Erosion and Hydrology model (RHEM) and the ARS-Water Erosion Prediction Project model (WEPP) may be driven with more commonly available precipitation datasets without sacrificing accuracy. In the analysis, the precipitation intensity parameter was accurately estimated within acceptable error bounds from both hourly and daily datasets, and for a variety of climate types. This work gives the potential for researchers who have limited climate data available to them to perform soil erosion modeling and therefore achieve greater geographical coverage of soil erosion models.
Technical Abstract: Precipitation intensity is an important meteorological input for water erosion and runoff applications. A commonly used intensity factor is maximum 30-min intensity (I30), which represents the sustained intensity of a storm. Determining I30 is challenging for two reasons. First, intensity can vary significantly over time, even within very short durations of 5 minutes or less. Second, the majority of precipitation data sets are limited by their fixed-interval nature, and I30 may not be constant within fixed measurement intervals. When intensity is simply averaged given the accumulation of a measurement interval, the temporal resolution of the precipitation data set biases the result. Therefore, in this study, bias adjustments were determined for a range of selected temporal resolutions and Köppen-Geiger climate regions in the United States. In this case, the intensity factor was monthly mean maximum 30-minute intensity (MX.5P), which is a parameter used to generate stochastic meteorological inputs for models that include the Rangeland Erosion and Hydrology model (RHEM) and the Water Erosion Prediction Project model (WEPP). The adjustment factors were obtained by using linear regression of reference MX.5P values derived from breakpoint data against MX.5P values aggregated from the breakpoint data to represent lower temporal resolutions. The resulting slope coefficients were used to determine bias adjustment factors. In addition, multivariate machine learning regression was used to obtain more complex correlations involving a host of predictor variables that may each be determined from daily precipitation statistics and the spatial location of each station. In total, 609 stations and 16 climate classifications were represented in the regressions. Linear regressions for climate classifications gave RMSE for values of MX.5P derived from hourly data ranging from 0.98-3.46 mm hr-1 with an average of 2.18 mm hr-1. For daily data, the error range was 2.83-8.44 mm hr-1 with an average of 5.61 mm hr-1. The multivariate regression using machine learning algorithms improved regressions for coarser resolutions, reducing error to the 3-4 mm hr-1 range for downscaled daily values.