Search

This work presents a first evaluation of the performance of the Integrated Multisatellite Retrievals for GPM (IMERG) precipitation product over the upper Blue Nile basin of Ethiopia. One of the unique features of this study is the availability of hourly rainfall measurements from an experimental rain gauge network in the area. Both the uncalibrated and calibrated versions of IMERG are evaluated, and their performance is contrasted against another high-resolution satellite product, which is the Kalman filter (KF)-based Climate Prediction Center (CPC) morphing technique (CMORPH). The analysis is performed for hourly and daily time scales and at spatial scales that correspond to the nominal resolution of satellite products, which is 0.1° spatial resolution. The period analyzed is focused on a single wet season (May-October 2014). Evaluation is performed using several statistical and categorical error metrics, as well as spatial correlation analysis to assess the ability of satellite products to represent spatial variability of precipitation in the area. Results show that both IMERG products have a better bias ratio and correlation coefficient on both time scales as compared to CMORPH. Comparison statistics show a slight improvement in the skill of detecting rainfall events in IMERG products compared to CMORPH. Results also show a decreasing trend in the detection ability of satellite products for increasing threshold values, highlighting the need to further improve detection during heavy precipitation. [ABSTRACT FROM AUTHOR]

• A machine learning outage prediction model (OPM) capable of predicting power outages at a circuit (the operational units of the power distribution network) resolution is developed to understand the impact of enhanced vegetation management standards on outages in the electrical distribution grid during storm events. • The authors find good matching between the increase in enhanced vegetation management (ETT) and reduction in trouble spots over the years from 2005 to 2019. • Most of the vegetation management on the electric grid was completed over the last five years of the study (2015–2019), and during those five years the OPM estimates annual reductions of trouble spots between 25.7 and 42.5 percent due to the tree trimming performed. • A novel feature of the OPM developed in this study is the ability to dynamically update it with different vegetation management scenarios at the circuit level and predict the effect of each scenario on trouble spots in the electric grid for storm events. • The authors also find increased prediction accuracy (a reduction in mean absolute percentage error of 4.1% and 5.2% for the two models trained) when including vegetation management data as an input variable to predict trouble spots in the electric grid. This paper develops a machine learning outage prediction model (OPM) to serve as a simulation framework capable of quantifying the reduction in damages to the distribution electric grid due to vegetation management for storm events. The model covers the Eversource Energy distribution grid territory in Connecticut and uses a random forest model with input variables for vegetation, vegetation management, land cover, drought, elevation, weather and electrical infrastructure to predict outages for each circuit (the operational units of the power distribution network). The model is trained on 165 storms from the years 2005 to 2019. The results show that over the last five years of the study (2015–2019) the annual reduction in trouble spots in the electric grid due to enhanced tree trimming is between 25.7 and 42.5% and there is good matching between increased trouble spot reduction and increased vegetation management. Further, we demonstrate improved outage predictions when including vegetation management data as an input variable, with a 4.1% reduction in mean absolute percentage error in leave-one-storm-out cross-validation. This framework could be used to examine varying vegetation management scenarios and the results should be useful for decision makers such as utilities, municipalities and regulators in optimizing vegetation management and broader grid resilience enhancement plans. [ABSTRACT FROM AUTHOR]

Abstract: The ambient sound field in the ocean is a combination of natural and manmade sounds. Consequently, the interpretation of the ambient sound field can be used to quantify these processes. In the frequency range from 1 to 50kHz, the general character of ocean ambient sound is a slowly changing background that is closely associated with local wind speed, interspersed with shorter time scale events such as rain storms, ships and animal calls. At lower frequencies the underwater ambient sound budget includes geologically generated sound activities including underwater volcanic eruptions, seismic and seepage faults that generate bubbles, etc. that can also potentially be classified and quantified. Acoustic data are collected on hydrophones. Hydrophones are simple, robust sensors that can be deployed on most ocean instrumentation systems including surface or sub-surface moorings, bottom mounted systems, drifters, ARGO floats or autonomous underwater platforms. A dedicated oceanic underwater recorder called a passive acoustic listener (PAL) has been developed. A principal issue is to accurately distinguish different sound sources so that they can be quantified as part of a sound budget, and then quantified if appropriate. Based on ongoing data collected from the Poseidon II network the retrieval potential of multi-parameters from underwater sound, including meteorological (i.e., precipitation and winds) and in general geophysical, anthropogenetic (i.e., ships, submarines, etc.) and biological (whales, etc.) sources is presented. [ABSTRACT FROM AUTHOR]

• The NCAR ensemble is able to represent the spatial and temporal characteristics of soil state across the CONUS. • Soil state estimates from the NCAR ensemble exhibited better performance compared to GFS over west and southeast CONUS. • The ensemble spread characteristics suggest that future limited-area continuously cycling EnKF systems could successfully assimilate soil state observations. Accurate characterization of soil state is imperative for weather and climate research, hydrological modeling, agriculture, water resources management, and drought monitoring. Limited-area data assimilation (DA) systems can provide high-resolution real-time soil state analyses. However, soil state representations from regional continuously cycling DA systems have not been examined. This study evaluates soil moisture and temperature estimates and spread characteristics from a real-time, limited-area, continuously-cycling ensemble Kalman filter (EnKF) data assimilation (DA) system run at the National Center for Atmospheric Research (NCAR) for 2.7 years over the conterminous United States (CONUS). Soil moisture and temperature data from both the EnKF and operational Global Forecast System (GFS) analysis are compared against the North American Land Data Assimilation System (NLDAS-2) soil state estimates. Results demonstrate the continuously-cycling EnKF was generally able to represent the spatial and temporal characteristics of soil moisture and temperature across the CONUS, demonstrating the feasibility of using limited-area continuously cycling EnKFs as a method to improve simulations of soil moisture and temperature. Soil state estimates from the NCAR ensemble exhibited lower Bias and RMSE compared to GFS analyses over west and southeast CONUS. Additionally, the ensemble spread characteristics suggest that future limited-area continuously cycling EnKF systems could successfully assimilate both in situ and remotely-sensed soil moisture and temperature observations to further improve soil state analyses. [ABSTRACT FROM AUTHOR]

Considering the reduction of steep power ramps caused by renewable energy penetration, the present study evaluates the potential of utilizing existing water supply infrastructure as small-scale pumped-hydroelectric storage (PHS) units. A novel methodology is developed that estimates the total storage capacity via the available space in five water supply systems in Connecticut (United States), by solely employing water level measurements and considering a water release schedule that favors power-demand peak reduction. The study culminates in the estimation of optimal areal requirements for hypothesized solar farms neighboring the studied systems, based on deficit minimization between storage capacity and energy generation over the study period. Analysis reveals that, in most cases, the seasonal pattern of the storage capacity is in accordance with the behavior of the corresponding hydraulic head timeseries, showing higher values in winter/autumn. Given certain assumptions regarding the infrastructure, a yearly average of approximately 95 MWh/day can be stored as an intersystem aggregate, illustrating that the amalgamation of existing water supply reservoirs could provide significant energy storage capacity to complement substation variability. Finally, the area covered by PV arrays increases proportionally with the storage capacity, while the calculated solar energy output demonstrates significant variability relative to the PHS storage potential. • Water supply infrastructure is evaluated as potential small-scale pumped-hydroelectric storage (PHS) units. • The storage capacity is estimated considering a water release pattern that favors power demand peak reduction. • Estimation of PHS potential is conducted via a novel methodology that depends solely on water level measurements. • The relation between storage potential and solar energy reveals the importance of complementary renewable energy sources. [ABSTRACT FROM AUTHOR]

• The Metastatistical Extreme Value Distribution (MEVD) framework applied to four short-record high-resolution quasi-global sub-daily precipitation products.. • MEVD parametric uncertainty and its propagation effect in the quantile is investigated. • Impact of spatial scale mismatch on the frequency analysis of extremes is quantified. • We provide an assessment of satellite-based precipitation products in different hydroclimatic regions. Precipitation extremes and associated hydrological hazards pose a significant global risk to society and economy. To be effective, mitigation strategies require the best possible estimation of the intensity and frequency of precipitation extremes. Traditional approaches to precipitation frequency analysis rely on long-term records from in-situ observations, which are limited in terms of global coverage. Satellite-based precipitation products provide global coverage, but errors in these estimates may lead to large biases in the quantification of extremes. Previous studies have demonstrated the ability of the novel Metastatistical Extreme Value Distribution (MEVD) framework to provide robust estimates of high quantiles in the presence of short-term data records and the uncertainties typical of remote sensing precipitation products. Here, we evaluate MEVD-based precipitation frequency analyses for four widely used quasi-global precipitation products (IMERG-v6, GSMaP-v6, CMORPH-v1.0, and MSWEP-v2) over high-density gauge networks in five hydroclimatic regions (Austria, Italy, Florida, Texas, and Arizona). We show dependence of MEVD-based estimation error on the characteristics of each dataset and the hydroclimatic region. Additionally, we evaluate the sub-grid variability of extreme precipitation and demonstrate the impact of spatial scale mismatch (that is, single in-situ gauge versus satellite pixel) on the frequency analysis of extremes. This work provides an assessment of the use of MEVD for estimating precipitation extremes from globally available datasets and an understanding of the variability of sub-daily precipitation extremes in different hydroclimatic regions of the world. [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]

This article compares two nonparametric tree-based models, quantile regression forests (QRF) and Bayesian additive regression trees (BART), for predicting storm outages on an electric distribution network in Connecticut, USA. We evaluated point estimates and prediction intervals of outage predictions for both models using high-resolution weather, infrastructure, and land use data for 89 storm events (including hurricanes, blizzards, and thunderstorms). We found that spatially BART predicted more accurate point estimates than QRF. However, QRF produced better prediction intervals for high spatial resolutions (2-km grid cells and towns), while BART predictions aggregated to coarser resolutions (divisions and service territory) more effectively. We also found that the predictive accuracy was dependent on the season (e.g., tree-leaf condition, storm characteristics), and that the predictions were most accurate for winter storms. Given the merits of each individual model, we suggest that BART and QRF be implemented together to show the complete picture of a storm's potential impact on the electric distribution network, which would allow for a utility to make better decisions about allocating prestorm resources., (© 2016 Society for Risk Analysis.)