Search Results

You are looking at 31 - 40 of 59 items for :

  • Author or Editor: Eric A. Smith x
  • Refine by Access: All Content x
Clear All Modify Search
Peter Bauer
,
Paul Amayenc
,
Christian D. Kummerow
, and
Eric A. Smith

Abstract

The objective of this paper is to establish a computationally efficient algorithm making use of the combination of Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) and precipitation radar (PR) observations. To set up the TMI algorithm, the retrieval databases developed in Part I served as input for different inversion techniques: multistage regressions and neural networks as well as Bayesian estimators. It was found that both Bayesian and neural network techniques performed equally well against PR estimates if all TMI channels were used. However, not using the 85.5-GHz channels produced consistently better results. This confirms the conclusions from Part I. Generally, regressions performed worse; thus they seem less suited for general application due to the insufficient representation of the nonlinearities of the TB–rain rate relation. It is concluded that the databases represent the most sensitive part of rainfall algorithm development.

Sensor combination was carried out by gridding PR estimates of rain liquid water content to 27 km × 44 km horizontal resolution at the center of gravity of the TMI 10.65-GHz channel weighting function. A liquid water dependent database collects common samples over the narrow swath covered by both TMI and PR. Average calibration functions are calculated, dynamically updated along the satellite track, and applied to the full TMI swath. The behavior of the calibration function was relatively stable. The TMI estimates showed a slight underestimation of rainfall at low rain liquid water contents (<0.1 g m−3) as well as at very high rainfall intensities (>0.8 g m−3) and excellent agreement in between. The biases were found to not depend on beam filling with a strong correlation to rain liquid water for stratiform clouds that may point to melting layer effects.

The remaining standard deviations between instantaneous TMI and PR estimates after calibration may be treated as a total retrieval error, assuming the PR estimates are unbiased. The error characteristics showed a rather constant absolute error of <0.05 g m−3 for rain liquid water contents <0.1 g m−3. Above, the error increases to 0.6 g m−3 for amounts up to 1 g m−3. In terms of relative errors, this corresponds to a sharp decrease from >100% to 35% between 0.05 and 0.5 g m−3. The database ambiguity, that is, the standard deviation of near-surface rain liquid water contents with the same radiometric signature, provides a means to estimate the contribution from the simulations to this error. In the range where brightness temperatures respond most sensitively to rainwater contents, almost the entire error originates from the ambiguity of signatures. At very low and very high rain rates (<0.05 and >0.7 g m−3) at least half of the total error is explained by the inversion process.

Full access
Song Yang
,
Kwo-Sen Kuo
, and
Eric A. Smith

Abstract

This investigation seeks a better understanding of the assorted mechanisms controlling the global distribution of diurnal precipitation variability based on the use of the Tropical Rainfall Measuring Mission (TRMM) microwave radiometer and radar data. The horizontal distributions of precipitation’s diurnal cycle are derived from 8 yr of TRMM Microwave Imager (TMI) and Precipitation Radar (PR) measurements involving three TRMM standard rain retrieval algorithms; the resultant distributions are analyzed at various spatiotemporal scales. The results reveal both the prominent and expected late-evening (LE) to early-morning (EM) precipitation maxima over oceans and the counterpart prominent and expected mid- to late-afternoon (MLA) maxima over continents. Moreover, and not generally recognized, the results reveal a widespread distribution of secondary maxima, which generally mirror their counterpart regime’s behavior, occurring over both oceans and continents. That is, many ocean regions exhibit clear-cut secondary MLA precipitation maxima, while many continental regions exhibit just as evident secondary LE–EM maxima. This investigation is the first comprehensive study of these globally prevalent secondary maxima and their widespread nature, a type of study only made possible when the analysis procedure is applied to a high-quality global-scale precipitation dataset.

The characteristics of the secondary maxima are mapped and described on global grids using an innovative clock-face format, while a current study that is to be published at a later date provides physically based explanations of the seasonal regional distributions of the secondary maxima. In addition to a primary “explicit” maxima identification scheme, a secondary “Fourier decomposition” maxima identification scheme is used as a cross-check to examine the amplitude and phase properties of the multimodal maxima. Accordingly, the advantages and ambiguities resulting from the use of a Fourier harmonic analysis are investigated.

Full access
Sandra E. Yuter
,
Robert A. Houze Jr.
,
Eric A. Smith
,
Thomas T. Wilheit
, and
Edward Zipser

Abstract

The Tropical Rainfall Measuring Mission (TRMM) Kwajalein Experiment (KWAJEX) was designed to obtain an empirical physical characterization of precipitating convective clouds over the tropical ocean. Coordinated datasets were collected by three aircraft, one ship, five upper-air sounding sites, and a variety of continuously recording remote and in situ surface-based sensors, including scanning Doppler radars, profilers, disdrometers, and rain gauges. This paper describes the physical characterization of the Kwajalein cloud population that has emerged from analyses of datasets that were obtained during KWAJEX and combined with long-term TRMM ground validation site observations encompassing three rainy seasons. The spatial and temporal dimensions of the precipitation entities exhibit a lognormal probability distribution, as has been observed over other parts of the tropical ocean. The diurnal cycle of the convection is also generally similar to that seen over other tropical oceans. The largest precipitating cloud elements—those with rain areas exceeding 14 000 km2—have the most pronounced diurnal cycle, with a maximum frequency of occurrence before dawn; the smallest rain areas are most frequent in the afternoon. The large systems exhibited stratiform rain areas juxtaposed with convective regions. Frequency distributions of dual-Doppler radar data showed narrow versus broad spectra of divergence in the stratiform and convective regions, respectively, as expected because strong up- and downdrafts are absent in the stratiform regions. The dual-Doppler profiles consistently showed low-level convergence and upper-level divergence in convective regions and midlevel convergence sandwiched between lower- and upper-level divergence in stratiform regions. However, the magnitudes of divergence are sensitive to assumptions made in classifying the radar echoes as convective or stratiform. This sensitivity implies that heating profiles derived from satellite radar data will be sensitive to the details of the scheme used to separate convective and stratiform rain areas. Comparison of airborne passive microwave data with ground-based radar data indicates that the pattern of scattering of 85-GHz radiance by ice particles in the upper portions of KWAJEX precipitating clouds is poorly correlated with the precipitation pattern at lower levels while the emission channels (10 and 19 GHz) have brightness temperature patterns that closely correspond to the lower-level precipitation structure. In situ ice particle imagery obtained by aircraft at upper levels (∼11 km) shows that the concentrations of ice particles of all densities are greater in the upper portions of active convective rain regions and lower in the upper portions of stratiform regions, probably because the active updrafts convey the particles to upper levels, whereas in the stratiform regions sedimentation removes the larger ice particles over time. Low-level aircraft flying in the rain layer show similar total drop concentrations in and out of convective cells, but they also show a sudden jump in the concentration of larger raindrops at the boundaries of the cells, indicating a discontinuity in growth processes such as coalescence at the cell boundary.

Full access
William E. Lewis
,
Eastwood Im
,
Simone Tanelli
,
Ziad Haddad
,
Gregory J. Tripoli
, and
Eric A. Smith

Abstract

The potential usefulness of spaceborne Doppler radar as a tropical cyclone observing tool is assessed by conducting a high-resolution simulation of an intense hurricane and generating synthetic observations of reflectivity and radial velocity. The ground-based velocity track display (GBVTD) technique is used to process the radial velocity observations and generate retrievals of meteorologically relevant metrics such as the maximum wind (MW), radius of maximum wind (RMW), and radius of 64-kt wind (R64). Results indicate that the performance of the retrieved metrics compares favorably with the current state-of-the-art satellite methods for intensity estimation and somewhat better than current methods for structure (i.e., wind radii).

Full access
Ralph R. Ferraro
,
Eric A. Smith
,
Wesley Berg
, and
George J. Huffman

Abstract

The success of any passive microwave precipitation retrieval algorithm relies on the proper identification of rain areas and the elimination of surface areas that produce a signature similar to that of precipitation. A discussion on the impact of and on methods that identify areas of rain, snow cover, deserts, and semiarid conditions over land, and rain, sea ice, strong surface winds, and clear, calm conditions over ocean, are presented. Additional artifacts caused by coastlines and Special Sensor Microwave/Imager data errors are also discussed, and methods to alleviate their impact are presented. The strengths and weaknesses of the “screening” techniques are examined through application on various case studies used in the WetNet PIP-2. Finally, a methodology to develop a set of screens for use as a common rainfall indicator for the intercomparison of the wide variety of algorithms submitted to PIP-2 is described.

Full access
Kwo-Sen Kuo
,
Eric A. Smith
,
Ziad Haddad
,
Eastwood Im
,
Toshio Iguchi
, and
Alberto Mugnai

Abstract

In developing the upcoming Global Precipitation Measurement (GPM) mission, a dual-frequency Ku–Ka-band radar system will be used to measure rainfall in such a fashion that the reflectivity ratio intrinsic to the measurement will be sensitive to underlying variations in the drop size distribution (DSD) of rain. This will enable improved techniques for retrieving rain rates, which are dependent upon several key properties of the DSD. This study examines this problem by considering a three-parameter set defined by liquid water content (W), DSD effective radius (r e ), and DSD effective variance (υ e ). Using radiative transfer simulations, this parameter set is shown to be related to a radar reflectivity factor and specific attenuation in such a fashion that details of the DSDs are immaterial under constant W, and thus effectively represent important variations in DSD that affect rain rate but with a minimal number of parameters. The analysis also examines the effectiveness of including some measure of mean Doppler fall velocity of raindrops ( υ ), given that the fundamental properties of a given precipitation situation are uniquely defined by a combination of a drop mass spectrum and drop vertical velocity spectrum. The results of this study have bearing on how future dual-frequency precipitation retrieval algorithms could be formulated to optimize the sensitivity to underlying DSD variability, a problem that has greatly upheld past progress in radar rain retrieval.

Full access
Ziad S. Haddad
,
Jonathan P. Meagher
,
Stephen L. Durden
,
Eric A. Smith
, and
Eastwood Im

Abstract

The threat of flooding from landfalling tropical cyclones is a function of the local variation in rain rate and rain accumulation. To date, these have been inferred from single-frequency radar reflectivity measurements. However, the Tropical Rainfall Measuring Mission experience has confirmed that one of the main difficulties in retrieving rain profiles using a single-frequency radar is the unknown raindrop size distribution (DSD). A dual-frequency radar such as the one planned for the upcoming Global Precipitation Measurement (GPM) core satellite is expected to help sort out at least part of this DSD-induced ambiguity. However, the signature of precipitation at 14 GHz does not differ greatly from its signature at 35 GHz (the GPM radar frequencies). To determine the extent of the vertical variability of the DSD in tropical systems and to quantify the effectiveness of a dual-frequency radar in resolving this ambiguity, several different models of DSD shape are considered and used to estimate the rain-rate and mean-diameter profiles from the measurements made by Jet Propulsion Laboratory’s (JPL’s) airborne second generation precipitation radar (PR-2) over Hurricanes Gabrielle and Humberto during the Fourth Convection and Moisture Experiment (CAMEX-4) in September 2001. It turns out that the vertical structures of the rain profiles retrieved from the same measurements under different DSD assumptions are similar, but the profiles themselves are quantitatively significantly different.

Full access
Eric A. Smith
,
Harry J. Cooper
,
Xuwu Xiang
,
Alberto Mugnai
, and
Gregory J. Tripoli

Abstract

A cloud-radiation model is used to investigate the relationship between emerging microwave brightness temperatures (T B 's) and vertically distributed mixtures of liquid and frozen hydrometeors as a means to establish the framework for a hybrid statistical-physical rainfall retrieval algorithm. The focus in this study is on the microwave characteristics of an intense hailstorm in which cold-rain microphysics dominate the precipitation process. The T B calculations exhibit a high degree of intercorrelation across a wide frequency range (15–128 GHz) due to the pervasive influence of large ice particles on attenuation of upwelling radiation emerging from the rain layers. When the radiative emission source is near blackbody, fluctuations of the mixing ratios of ice particles are wholly responsible for the T B variations, as opposed to fluctuations in the cloud-or raindrop mixing ratios. Supercooled cloud drops, suspended in the graupel layers, can exert influence on the T B 's but only at the higher frequencies. Emission by the large ice particles themselves becomes an important radiative source to the emerging T B 's at the top of the atmosphere as the graupel-mixing ratios increase and effectively block the radiative sources from within the liquid layers.

Strong relationships are found between the emerging T B 's and various rain parameters, but these correlations are misleading in that the T B 's are largely controlled by fluctuations in ice-particle mixing ratios, which in turn are highly correlated to fluctuations in liquid-particle mixing ratios. This does not negate the use of empirically based brightness-temperature-rain-rate (T B -RR) algorithms as useful tools for estimating precipitation (i.e., graupel particles ultimately fall out as rain), but it does point to a basic problem that remote-sensing methodology must address. More specifically, the hydrometeor profiles used for T B -RR algorithms must not be specified in an ad hoc fashion. It is argued that a cloud model can overcome the ad hoc assumptions.

It is shown that the lowest SSM/I frequency (19 GHz) is actually a better estimator of columnar ice water content than the highest frequency (85 GHz). This is because both cloud-water emission and multiple scattering by ice particles are more prevalent at 85 GHz than at 19 GHz (which tends to be mostly influenced by single scattering). As a consequence, 85 GHz is much more sensitive to the configurational details of the vertical distribution of large ice particles and to the presence of supercooled cloud drops within the lower ice layers.

Full access
Simone Tanelli
,
Eastwood Im
,
Stephen L. Durden
,
Luca Facheris
,
Dino Giuli
, and
Eric A. Smith

Abstract

For vertical Doppler velocity measurements of a homogeneous rain field, the standard spectral moment estimation techniques commonly used by ground-based and airborne Doppler rain radars can be readily extended for spaceborne application, provided that the radar antenna size is chosen to adequately reduce the satellite motion-induced Doppler spectral broadening. When encountering an inhomogeneous rain field, on the other hand, the nonuniform beam filling (NUBF) causes additional biases on Doppler velocity estimates, which (i) often reach several meters per second, (ii) cannot be corrected with standard spectral moment techniques, and (iii) are strongly dependent on the along-track reflectivity profile within the radar footprint. One approach to overcome this difficulty is to further increase the antenna size such that the radar's horizontal resolution would be sufficiently small to resolve the inhomogeneity in rain cells. Unfortunately, this approach is very challenging in terms of antenna technology and spacecraft resources and accommodation.

In this paper, an alternate data processing approach is presented to overcome the NUBF difficulty. This combined frequency–time (CFT) processing technique is used to process a series of Doppler spectra collected over measurement volumes that are partially overlapping in the along-track direction. Its expected performance is evaluated through a spaceborne simulation study using three case studies from high-resolution 3D rainfall datasets acquired by the NASA JPL airborne rain mapping radar. In each of these cases, each representing a different rain regime with a different degree of spatial variability, the CFT technique can effectively remove the NUBF-induced bias such that the mean Doppler velocity estimates achieve the desired accuracy of 1 m s−1 for a signal-to-noise ratio greater than 10 dB.

Full access
Byung-Ju Sohn
,
Eui-Seok Chung
,
Johannes Schmetz
, and
Eric A. Smith

Abstract

A method and a passive microwave retrieval algorithm have been developed to retrieve upper-tropospheric water vapor (UTW) from Special Sensor Microwave Water Vapor Profiler (SSM/T-2) measurements taken at three discrete frequencies near the 183-GHz water vapor line. The algorithm is based on physical relaxation utilizing statistical covariance information to provide initial-guess profiles and to constrain the updating step in the relaxation process. The scheme incorporates a method to remove SSM/T-2 brightness temperature bias in comparison with collocated simulated brightness temperatures. Correction functions are designed for the three SSM/T-2 183-GHz channels. The algorithm is validated against radiosonde observations and collocated SSM/T-2 brightness temperatures. Under clear-sky and nonprecipitating-cloud conditions, the UTW retrievals exhibit an rms error of 0.68 kg m−2 with integrated water vapor biases below 5% for the upper-tropospheric layers of 700–500 and 500–200 hPa. The retrieval provides an independent source of satellite-derived water vapor information in the upper troposphere, distinct from upper-tropospheric humidity information retrieved from thermal infrared (IR) measurements around the 6.3-μm water vapor absorption band. The microwave retrievals can then be used to cross-check IR retrievals and/or to augment IR retrievals, dependent upon the problem at hand.

Full access