Showing papers by "Robert Meneghini published in 2017"
Goddard Space Flight Center1, Marshall Space Flight Center2, Colorado State University3, University of Maryland, College Park4, NASA Headquarters5, National Institute of Information and Communications Technology6, National Oceanic and Atmospheric Administration7, Japan Aerospace Exploration Agency8, University of Maryland, Baltimore County9, University of Tokyo10, Texas A&M University11
TL;DR: The GPM mission collects essential rain and snow data for scientific studies and societal benefit and aims to provide real-time information about rainfall and snowfall to improve understanding of climate change.
Abstract: The GPM mission collects essential rain and snow data for scientific studies and societal benefit.
525 citations
TL;DR: Comparisons of the standard deviations for the fixed and variable-averaged grids are given as a function of incidence angle and surface type using a three-month set of data.
Abstract: For an airborne or spaceborne radar, the precipitation-induced path attenuation can be estimated from the measurements of the normalized surface cross section, $\sigma ^{\mathbf {0}}$ , in the presence and absence of precipitation. In one implementation, the mean rain-free estimate and its variability are found from a lookup table (LUT) derived from previously measured data. For the dual-frequency precipitation radar aboard the global precipitation measurement satellite, the nominal table consists of the statistics of the rain-free $\sigma ^{0}$ over a $0.5 {^{\circ }} \times 0.5 {^{\circ }}$ latitude–longitude grid using a three-month set of input data. However, a problem with the LUT is an insufficient number of samples in many cells. An alternative table is constructed by a stepwise procedure that begins with the statistics over a $0.25 {^{\circ }} \times 0.25 {^{\circ }}$ grid. If the number of samples at a cell is too few, the area is expanded, cell by cell, choosing at each step that cell that minimizes the variance of the data. The question arises, however, as to whether the selected region corresponds to the smallest variance. To address this question, a second type of variable-averaging grid is constructed using all possible spatial configurations and computing the variance of the data within each region. Comparisons of the standard deviations for the fixed and variable-averaged grids are given as a function of incidence angle and surface type using a three-month set of data. The advantage of variable spatial averaging is that the average standard deviation can be reduced relative to the fixed grid while satisfying the minimum sample requirement.
4 citations