Search

• Precipitation annual maxima estimated from NEXRAD are evaluated over the CONUS. • Accuracy of radar-based estimates varies with magnitude and duration. • Uncertainty in radar-based IDF curves exhibits climate dependence. The lack of knowledge on precipitation frequency over ungauged areas introduces a significant source of uncertainty in relevant engineering designs and risk estimation procedures. Radar-based observations offer precipitation information over ungauged areas and thus have gained increasing attention as a potential solution to this problem. However, due to their relative short data records and inherent uncertainty sources, their ability to provide accurate estimates on the frequency of precipitation extremes requires evaluation. This study involves the evaluation of at-site precipitation frequency estimates from NEXRAD Stage IV radar precipitation dataset. We derive precipitation annual maxima series from the 16yrs record (2002–2017) of NEXRAD and we compare against 539 long-term (50yrs) hourly gauge records. In addition, Intensity-Duration-Frequency (IDF) curves are estimated from both radar and gauge dataset and compared. IDF estimation is based on fitting the Generalize Extreme Value distribution to annual precipitation maxima. Evaluation is carried out over the contiguous United States and results are grouped and presented for five dominant climate classes and for a range of return period and precipitation durations. NEXRAD was shown to overestimate intensities at shorter durations (1- and 3-h) and low quantiles, while it tends to underestimate higher quantiles at longer durations (24 h). In addition, evaluation of the IDF curves estimated from NEXRAD revealed a distinct geographic dependence with certain regions exhibiting a tendency to overestimation (e.g. east of the Rocky Mountains) or underestimation (Midwest). Overall, this analysis suggests that, while significant discrepancies may exist, there are several cases where NEXRAD provide estimates within the uncertainty bounds of the reference rain gauge dataset. The climate/geographic region and the temporal duration are important aspects to consider. Findings provided in this work on these aspects will hopefully serve as a general guideline for those interested in using NEXRAD estimates for further research or applications on precipitation extremes. [ABSTRACT FROM AUTHOR]

An extensive evaluation of nine global-scale high-resolution satellite-based rainfall (SBR) products is performed using a minimum of 6 years (within the period of 2000-13) of reference rainfall data derived from rain gauge networks in nine mountainous regions across the globe. The SBR products are compared to a recently released global reanalysis dataset from the European Centre for Medium-Range Weather Forecasts (ECMWF). The study areas include the eastern Italian Alps, the Swiss Alps, the western Black Sea of Turkey, the French Cévennes, the Peruvian Andes, the Colombian Andes, the Himalayas over Nepal, the Blue Nile in East Africa, Taiwan, and the U.S. Rocky Mountains. Evaluation is performed at annual, monthly, and daily time scales and 0.25° spatial resolution. The SBR datasets are based on the following retrieval algorithms: Tropical Rainfall Measuring Mission Multisatellite Precipitation Analysis (TMPA), the NOAA/Climate Prediction Center morphing technique (CMORPH), Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN), and Global Satellite Mapping of Precipitation (GSMaP). SBR products are categorized into those that include gauge adjustment versus unadjusted. Results show that performance of SBR is highly dependent on the rainfall variability. Many SBR products usually underestimate wet season and overestimate dry season precipitation. The performance of gauge adjustment to the SBR products varies by region and depends greatly on the representativeness of the rain gauge network. [ABSTRACT FROM AUTHOR]

The current statistical methods applied in flood frequency analysis require long data records to obtain reliable estimates, particularly for long return periods. Moreover, the choice of the statistical model and the parameter estimation procedure may introduce uncertainty in the estimates. In this work, we investigate the sensitivity of flood frequency analysis to various sample sizes, statistical models, and parameter estimation methods over six major hydrological regions in the contiguous United States. Results show that flood frequency estimates based on annual maximum series approach convergence to the reference values (estimates derived from 70 years record) in terms of median for 35‐year or longer records. However, the uncertainty remains significant and a record of 35 years (20 years) is associated with ~50% (100%) larger uncertainty on the estimated 100‐year flood. The generalised extreme value distribution combined with maximum likelihood estimation method is associated with the largest uncertainty, while the log‐Pearson type III exhibits comparable bias and smaller uncertainty. Application of the partial duration series approach to 20‐year records shows no significant advantage. Our findings suggest that the hydroclimatic characteristics of the catchments exhibit limited impact on the uncertainty. [ABSTRACT FROM AUTHOR]

• An analytical multifractal (MF) approximation for IDF curve estimation is tested. • The MF model is compared to annual maxima and peaks-over-threshold approaches. • The MF IDF estimates are more accurate, especially for short rainfall records. • The MF parameters resemble the local topography and rainfall climatology. • The MF parameter setting is particularly appealing for ungauged locations. The present study uses hourly rainfall measurements from 36 raingauge stations in the northeastern United States with more than 55 years of recordings, to assess the accuracy and robustness of three conceptually different parametric approaches to produce intensity-duration-frequency (IDF) curves from rainfall records: (a) the well celebrated parametric IDF model based on annual maxima (AM) series, initially introduced by Rossi and Villani (1994) and later extended by Koutsoyiannis et al. (1998), (b) a peaks-over-threshold (PoT) model based on proper fitting of a reparametrized generalized Pareto (GP) distribution to the scaled excesses of the average rainfall intensity over various durations d , and (c) the analytical approximation introduced by Langousis et al. (2009) for estimation of multifractal (MF) rainfall maxima. Since its development, the third approach has never been tested in real world applications, and/or compared to routinely used approaches for IDF curve estimation. The obtained results show that the MF approach may constitute an important tool for hydrological applications, as it produces more accurate and robust IDF estimates than the other two approaches, irrespective of the duration d and return period T of interest, especially in cases where rainfall records exhibit significant length limitations. In addition, the parameters of the MF analytical approximation are found to be descriptive of the local topography and rainfall climatology of the region, providing a physically meaningful modeling setting that is particularly suited for regional frequency analysis and IDF estimation at ungauged locations. [ABSTRACT FROM AUTHOR]

• Weibull distribution models well sub-daily precipitation extremes over CONUS. • Significant relation between storm characteristics and statistics of extremes is found. • Tail heaviness depends on climate type and storm characteristics. In this paper we use a statistical framework to analyze the relation between storm properties and the statistics of extreme precipitation. We identify storm events using a 24-hour dry hiatus separation. We investigate the statistics of the hourly maximum intensity for the heaviest storm events at durations of 1 and 24 h. A two-parameter Weibull distribution is used to represent precipitation frequencies at several stations from a quality-controlled hourly precipitation dataset over the contiguous United States, encompassing seven climate zones. The Spearman correlation between the distribution parameters and a selection of storm properties (duration, intensity, decorrelation time, convective-like ratio) and station properties (elevation and latitude) is used to measure the relation of these properties with the statistics of extreme precipitation. Our results indicate that observed annual maximum hourly precipitation over the entire study area are likely samples from the used distribution, implying that a two-parameter Weibull distribution is suitable for modeling hourly and 24-hour precipitation extremes over the contiguous United States. The spatial variability of shape parameters obtained for hourly events showed lighter tails in the west coast when compared to the rest of CONUS, while the central-north displays heavier tails. We identify statistically significant links (at the 95% confidence level) between storm characteristics connected with the underlying processes (e.g., typical storm duration, typical temporal autocorrelation, proportion of convective-like storms) and the parameters of the distribution. Notably, characteristics typical of convective precipitation, e.g., sharp decorrelation time and high hourly intensity, tend to be related to distributions with heavier tails. These results provide a first step towards linking the characteristics of storms with the local statistics of extremes. [ABSTRACT FROM AUTHOR]