Search Results

You are looking at 1 - 10 of 19 items for

  • Author or Editor: Larry Stowe x
  • Refine by Access: All Content x
Clear All Modify Search
Alexander Ignatov and Larry Stowe

Abstract

The present second-generation aerosol retrieval algorithm over oceans used at NOAA/National Environmental Satellite, Data, and Information Service (NESDIS) separately retrieves two values of aerosol optical depth, τ1 and τ2, from Advanced Very High Resolution Radiometer (AVHRR) channels 1 and 2 centered at λ1 = 0.63 (operational) and λ2 = 0.83 μm (experimental), respectively. From these, an effective Ångström exponent α, related to particle size, can be derived as α = −ln(τ1/τ2)/ln(λ1/λ2). The single-channel lookup tables, relating reflectance to optical depth in the retrievals, have been precalculated with the Dave (1973) scalar radiative transfer (RT) model. This first part of a two-part paper describes the retrieval algorithm, with emphasis on its RT modeling related elements, and documents the transition to the Second Simulation of the Satellite Signal in the Solar Spectrum (6S; 1997) RT model. The new 6S RT model has the capability to account for reflection from wind-roughened sea surface, offers a wide choice of flexible aerosol and gaseous absorption models, and allows easy convolution with the sensor's spectral response. The value of these new features for aerosol remote sensing from AVHRR is discussed in detail. The transition effect is quantified by directly applying the Dave- and 6S-based algorithms to four large datasets of NOAA-14 AVHRR measurements, collected between February 1998 and May 1999 over the latitudinal belt of 5°–25°S. Statistics of the differences (δτ = τDaveτ6S and δα = αDaveα6S) are as follows: averages − 〈δτ1〉 < 1 × 10−3, 〈δτ2〉 ≈ −4 × 10−3, and 〈δα〉 ≈ +8 × 10−2; and standard deviations are στ1 ∼ 6 × 10−3, στ2 ∼ 4 × 10−3, and σα ≈ 9 × 10−2. These are found to be well within a few percent of typical values of τ and α and their respective ranges of variability, thus ensuring a smooth transition and continuity in the operational aerosol retrieval. On the other hand, the 6S model provides a much more flexible RT modeling tool compared to the previously used Dave code.

Full access
Alexander Ignatov and Larry Stowe

Abstract

This second part of a two-part study evaluates retrievals of aerosol optical depths, τ 1 and τ 2, in Advanced Very High Resolution Radiometer (AVHRR) channels 1 and 2 centered at λ 1 = 0.63 and λ 2 = 0.83 μm, and an effective Ångström exponent, α, derived therefrom as α = −ln(τ 1/τ 2)/ln(λ 1/λ 2). The retrievals are made with the Second Simulation of the Satellite Signal in the Solar Spectrum (6S) radiative transfer model from four NOAA-14 AVHRR datasets, collected between February 1998 and May 1999 in the latitudinal belt of 5°–25°S. A series of quality control (QC) checks applied to the retrievals to identify outliers are described. These remove a total of ∼1% of points, which presumably originate from channel misregistration, residual cloud in AVHRR cloud-screened pixels, and substantial deviations from the assumptions used in the retrieval model (e.g., bright coastal and high altitude inland waters). First, from examining histograms of the derived parameters it is found that τ and α are accurately fit by lognormal and normal probability distribution functions (PDFs), respectively. Second, the scattergrams τ 1 versus τ 2 are analyzed to see if they form a coherent pattern. They do indeed converge at the origin, as expected, but frequently are outside of the expected domain in τ 1τ 2 space, defined by two straight lines corresponding to α = 0 and α = 2. This results in a low bias in α, which tends to fill in an interval of α ∈ [−1, 1] rather than α ∈ [0, 2]. Third, scattergrams of α versus τ are used to empirically confirm a previously drawn theoretical conclusion that errors in α are inversely proportional to τ. More in-depth quantitative analyses suggest that the AVHRR-derived Ångström exponent becomes progressively more meaningful when τ > 0.2. Geographical trends are studied to demonstrate that the selected ocean area is reasonably uniform to justify application of consistency checks to reveal angular trends in the retrievals. These checks show that in most cases, the artifacts in the retrieved τ and α are statistically insignificant. On average, the analyses suggest that the retrieved τ 1, τ 2, and α show a high degree of self- and interconsistency, with the exception of a troublesome May 1999 dataset. The most prominent problem noticed so far is the inconsistency between τ 1 and τ 2, persistent from one dataset to another, which calls for fine-tuning some (not aerosol-model related) elements of the retrieval algorithm. These adjustments will be discussed elsewhere.

Full access
Larry L. Stowe Jr.

Abstract

The effects of particulate matter on the radiance of the emerging terrestrial infrared radiation for six wave-numbers in the wings and centers of the 6.3-µm water vapor and 15-µm carbon dioxide absorption bands have been theoretically evaluated using the mathematical formulation developed by Sekera and Stowe. Atmospheric models based on measured aerosol particle concentration and radiosonde data have been used in the computations, characterizing atmospheric turbidity conditions in low, middle and high latitudes. Empirical formulas by Golubitskiy and Moskalenko were used in the calculation of the molecular absorption parameters, while the scattering and absorption data for the particulates were taken from tables by Deirmendjian.

The changes in radiance of the upward radiation emerging from models of clear atmospheres due to the presence of particulates have been computed for nadir angles 0° ≤ ζ ≤ 70° and found to be rather small, reducing the radiance from a clear atmosphere by less than 1% in most cases. However, certain quantities used in the determination of temperature and humidity vertical profiles from terrestrial radiation measurements by satellite (inversion problems) are subject to larger changes if the effects of particulate matter are disregarded.

Full access
Larry L. Stowe
Full access
Alexander Ignatov and Larry Stowe

Abstract

This paper outlines the processing stream for aerosol retrievals over oceans from the visible and infrared scanner [VIRS; a five-channel radiometer similar to the National Oceanic and Atmospheric Administration’s Advanced Very High Resolution Radiometer (AVHRR)] aboard the Tropical Rainfall Measuring Mission (TRMM) satellite, launched in November 1997. Emphasis is on 1) the applying the previously developed AVHRR second-generation aerosol retrieval algorithm to VIRS data to derive an aerosol parameter, indicative of particle size; 2) removing the unwanted “thermal leak” signal in the 1.61-μm channel; 3) giving examples of the first aerosol retrievals from space at this wavelength; and 4) assessing the accuracy of the retrievals with theoretical error analyses and empirical self- and interconsistency checks. Aerosol optical depths τA1 and τA2 are retrieved from reflected solar radiances in VIRS channels 1 and 2 centered at wavelengths λ 1 = 0.63 and λ 2 = 1.61 μm, using two independent lookup tables. When τA1 and τA2 exceed a certain threshold τAmin an effective Ångström exponent α related to particle size is derived as α = −ln(τA1/τA2)/ln(λ 1/λ 2). Channel 2 is contaminated by a thermal leak, originating from a secondary spectral response peak centered at ∼5.2 μm. If uncorrected, it leads to errors in τA2 of 100% or more. To minimize this error, nighttime VIRS “dark” radiances in channel 2 have been related empirically to radiances in channels 4 and 5 (10.8 and 12 μm, respectively), and view angle through regression analyses. The reflected component in channel-2 daytime measurements is estimated by subtracting the empirically derived thermal component from the total signal and is used in the retrieval of τA2. Theoretical error analysis is used to identify the limitations of the VIRS retrieval algorithm, whereas actual retrievals are preliminarily evaluated using a set of specially developed empirical checks. The checks show, on average, a high degree of self- and interconsistency but also identify problems with the retrievals, the most noteworthy being trends in retrieved optical depths with viewing and illumination angles. These problems will be tackled in the next-generation aerosol retrieval algorithm.

Full access
Kenneth R. Knapp and Larry L. Stowe

Abstract

In spite of numerous studies on the remote sensing of aerosols from satellites, the magnitude of aerosol climate forcing remains uncertain. However, data from the Advanced Very High Resolution Radiometer (AVHRR) Pathfinder-Atmosphere (PATMOS) dataset—a statistical reduction of more than 19 yr of AVHRR data (1981–2000)—could provide nearly 20 yr of aerosol history. PATMOS data have a daily 110 × 110 km2 equal-area grid that contains means and standard deviations of AVHRR observations within each grid cell. This research is a first step toward understanding aerosols over land with PATMOS data. Herein, the aerosol optical depth is retrieved over land at numerous Aerosol Robotic Network (AERONET) sites around the globe using PATMOS cloud-free reflectances. First, the surface bidirectional reflectance distribution function (BRDF) is retrieved using a lookup table created with a radiative transfer model and the Rahman BRDF. Aerosol optical depths are then retrieved using the retrieved BRDF parameters and the PATMOS reflectances assuming a globally constant aerosol model. This method is applied to locations with ground truth measurements, where comparisons show that the best retrievals are made by estimating the surface reflectance using observations grouped by month. Random errors (i.e., correlation coefficients and standard error of estimate) in this case are lower than those where the surface BRDF is allowed year-to-year variations. By grouping the comparison results by land cover type, it was found that less noise is expected over forested regions, with a significant potential for retrieval for 80% of all land surfaces. These results and analyses suggest that the PATMOS data can provide valuable information on aerosols over land.

Full access
Richard Hucek, Larry Stowe, and Robert Joyce

Abstract

Accuracy estimates for the broadband CERES-I (Clouds and Earth's Radiant Energy System Instrument) measurements of daily average radiant exitance are presented. This is a continuation of the authors’ earlier CERES sampling studies published as Part I and II. Daily averaging errors result from not sampling the entire 24-h period with a system of polar satellites. Instantaneous errors, the subject of the previous studies, are also included. Separate estimates for daily average emitted longwave (LW) and reflected shortwave (SW) radiant fluxes are given.

The earth SW and LW reference radiation fields are derived from 3-h Geosynchronous Operational Environmental Satellite data, time interpolated between image times, and partitioned into upwelling radiances using scene-dependent angular dependence models (ADMs). Perturbations in these ADMs are introduced to cause instantaneous angular sampling errors (also referred to as ADM errors). These ADM errors, along with spatial sampling errors, are propagated through the time integration process for a more realistic estimate of the daily average error. Three satellite observing configurations are considered. They represent individually, and in combination, a proposed European Polar Orbiting Platform and National Aeronautics and Space Administration Earth Observing System-A sun-synchronous polar-orbiting satellite system. The Earth Radiation Budget Experiment single and multiple satellite time and space averaging algorithms are used for the satellite retrieval. One-satellite spatial root-mean-square (rms) daily averaged SW flux errors of 11–17 W m−2 are obtained for 2.5° latitude-longitude regions over the area studied (15°S–45°N, 50°–120°W). The two-satellite system has errors that are some 40%–60% less, having values between 5 and 9 W m−2. Only the two-satellite system can meet the 10 W m−2 user accuracy requirement for regional daily averaged SW fluxes. Longwave flux errors of 5–6 W m−2 and 3–4 W m−2, respectively, are found for the one- and two-satellite configurations.

The largest component of CERES 2.5° daily averaged target area error is due to sparse temporal sampling. The ADM error propagated into the daily average becomes more important as the temporal sampling error is reduced with the two-satellite system. For this system, the ADM error component (of the daily averaged error) for SW radiation reaches a magnitude that can be as large as 8 W m−2 at high solar zenith angles (SZA), where scene anisotropy is usually greatest. Over the study domain, up to 15% of the total rms error is due to ADM errors. Moreover, CERES 2.5° zonal mean daily averaged errors exhibit a latitudinal dependence of some 7 W m−2 for a 60° change in latitude in the presence of 30% systematic errors in the ADMs. This is largely attributable to the SZA dependence of instantaneous ADM error. Without ADM errors, zonal mean daily averaged target area biases range up to 3–4 W m−2 with an irregular latitudinal variation.

Full access
Larry Stowe, Richard Hucek, Philip Ardanuy, and Robert Joyce

Abstract

Much of the new record of broadband earth radiation budget satellite measurements to be obtained during the late 1990s and early twenty-first century will come from the dual-radiometer Clouds and Earth's Radiant Energy System Instrument (CERES-1) flown aboard sun-synchronous polar orbiters. Simulation studies conducted in this work for an early afternoon satellite orbit indicate that spatial rms sampling errors of instantaneous CERES-I shortwave flux estimates will range from about 8.5 to 14.0 W m−2 on a 2.5° latitude and longitude grid resolution. Root-mean-square errors in longwave flux estimates are only about 20% as large and range from 1.5 to 3.5 W m−2. These results are based on an optimal cross-track scanner design that includes 50% footprint overlap to eliminate gaps in the top-of-the-atmosphere coverage, and a “smallest” footprint size to increase the ratio in the number of observations lying within to the number of observations lying on grid area boundaries.

Total instantaneous measurement error depends additionally on the variability of anisotropic reflectance and emission patterns and on the retrieval methods used to generate target area fluxes. Three retrieval procedures are investigated, all relying on a maximum-likelihood estimation technique for scene identification. Observations from both CERES-1 scanners (cross-track and rotating azimuth plane) are used. One method is the baseline Earth Radiation Budget Experiment (ERBE) procedure, which assumes that errors due to the use of mean angular dependence models (ADMs) in the radiance-to-flux inversion process nearly cancel when averaged over grid areas. In a second (estimation of N) method, instantaneous ADMs are estimated from the multiangular, collocated observations of the two scanners. These observed models replace the mean models in the computation of the satellite flux estimates. In the third (scene flux) approach, separate target-area retrievals are conducted for each ERBE scene category and their results are combined using area weighting by scene type. The ERBE retrieval performs best when the simulated radiance field departs from the ERBE mean models by less than 10%. For larger perturbations, both the scene flux and collocation methods produce less error than the ERBE retrieval. The scene flux technique is preferable, however, because it involves fewer restrictive assumptions.

Full access
Larry L. Stowe, Paul A. Davis, and E. Paul McClain

Abstract

An algorithm for the remote sensing of global cloud cover using multispectral radiance measurements from the Advanced Very High Resolution Radiometer (AVHRR) on board National Oceanic and Atmospheric Administration (NOAA) polar-orbiting satellites has been developed. The CLAVR-1 (Clouds from AVHRR-Phase I) algorithm classifies 2 × 2 pixel arrays from the Global Area Coverage (GAC) 4-km-resolution archived database into CLEAR, MIXED, and CLOUDY categories. The algorithm uses a sequence of multispectral contrast, spectral, and spatial signature threshold tests to perform the classification. The various tests and the derivation of their thresholds are presented. CLAVR-1 has evolved through experience in applying it to real-time NOAA-11 data, and retrospectively through the NOAA AVHRR Pathfinder Atmosphere project, where 16 years of data have been reprocessed into cloud, radiation budget, and aerosol climatologies. The classifications are evaluated regionally with image analysis, and it is concluded that the algorithm does well at classifying perfectly clear pixel arrays, except at high latitudes in their winter seasons. It also has difficulties with classifications over some desert and mountainous regions and when viewing regions of ocean specular reflection. Generally, the CLAVR-1 fractional cloud amounts, when computed using a statistically equivalent spatial coherence method, agree to within about 0.05–0.10 of image/analyst estimates on average. There is a tendency for CLAVR-1 to underestimate cloud amount when it is large and to overestimate it when small.

Full access
Larry Stowe, Philip Ardanuy, Richard Hucek, Peter Abel, and Herbert Jacobowitz

Abstract

A set of system simulations has been performed to evaluate candidate scanner designs for an Earth Radiation Budget Instrument (ERBI) for the Earth Observing System (EOS) of the late 1990s. Five different instruments are considered: 1) the Active Cavity Array (ACA), 2) the Clouds and Earth's Radiant Energy System-Instrument (CERES-1), 3) the Conically Scanning Radiometer (CSR), (4) the Earth Radiation Budget Experiment Cross-Track Scanner (ERBE), and 5) the Nimbus-7 Biaxial Scanner (N7). Errors in instantaneous, top-of-the-atmosphere (TOA) satellite flux estimates are assumed to arise from two measurement problems: the sampling of space over a given geographic domain, and sampling in angle about a given spatial location. In the limit where angular sampling errors vanish [due to the application of correct angular dependence models (ADMs) during inversion], the accuracy of each scanner design is determined by the instrument's ability to map the TOA radiance field in a uniform manner. In this regard, the instruments containing a cross-track scanning component (CERES-1 and ERBE) do best. As errors in ADMs are encountered, cross-track instruments incur angular sampling errors more rapidly than biaxial instruments (N7, ACA, and CSR) and eventually overtake the biaxial designs in their total error amounts. A latitude bias (north-south error gradient) in the ADM error of cross-track instruments also exists. This would be objectionable when ADM errors are systematic over large areas of the globe. For instantaneous errors, however, cross-track scanners outperform biaxial or conical scanners for 2.5° latitude × 2.5° longitude target areas. providing that the ADM error is less than or equal to 30%.

A key issue is the amount of systematic ADM error (departures from the mean models) that is present at the 2.5° resolution of the ERBE target areas. If this error is less than 30%, then the CERES-I, ERBE, and CSR, in order of increasing error, provide the most accurate instantaneous flux estimates, within 2–3 W m−2 of each other in reflected shortwave flux. The magnitude of this error is near the 10 W m−2 accuracy requirement of the user community. Longwave flux errors have been found to have the same space and time characteristics as errors in shortwave radiation, but only about 25% as large.

Full access