Banghua Yan

A microwave land emissivity model

Satellite observations using microwave radiometers operating near the window regions are strongly affected by surface emissivity. Presently, the measurements obtained over land are not directly utilized in numerical weather prediction models because of uncertainties in estimating the emissivity. This study develops a new model to quantify the land emissivity over various surface conditions. For surfaces such as snow, deserts, and vegetation, volumetric scattering was calculated using a two-stream radiative transfer approximation. The reflection and transmission at the surface-air interface and lower boundary were derived by modifying the Fresnel equations to account for crosspolarization and surface roughness effects. Several techniques were utilized to compute the optical parameters for the medium, which is used in the radiative transfer solution. In the case of vegetation, geometrical optics is used because the leaf size is typically larger than the wavelength. For snow and deserts, a dense medium theory was adopted to take into account the coherent scattering of closely spaced particles. The emissivity spectra at frequencies between 4.9 and 94 GHz was simulated and compared with the ground-based radiometer measurements for bare soil, grass land, and snow conditions. It is shown that the features including the spectra, variability, and polarization agree well with the measurements. The simulated global distribution of land surface emissivity is also compared with the satellite retrievals from the Advanced Microwave Sounding Unit (AMSU). It is found that the largest discrepancies primarily occur over high latitudes where the snow properties are complex and least understood.

MiRS: An All-Weather 1DVAR Satellite Data Assimilation and Retrieval System

A 1-D variational system has been developed to process spaceborne measurements. It is an iterative physical inversion system that finds a consistent geophysical solution to fit all radiometric measurements simultaneously. One of the particularities of the system is its applicability in cloudy and precipitating conditions. Although valid, in principle, for all sensors for which the radiative transfer model applies, it has only been tested for passive microwave sensors to date. The Microwave Integrated Retrieval System (MiRS) inverts the radiative transfer equation by finding radiometrically appropriate profiles of temperature, moisture, liquid cloud, and hydrometeors, as well as the surface emissivity spectrum and skin temperature. The inclusion of the emissivity spectrum in the state vector makes the system applicable globally, with the only differences between land, ocean, sea ice, and snow backgrounds residing in the covariance matrix chosen to spectrally constrain the emissivity. Similarly, the inclusion of the cloud and hydrometeor parameters within the inverted state vector makes the assimilation/inversion of cloudy and rainy radiances possible, and therefore, it provides an all-weather capability to the system. Furthermore, MiRS is highly flexible, and it could be used as a retrieval tool (independent of numerical weather prediction) or as an assimilation system when combined with a forecast field used as a first guess and/or background. In the MiRS, the fundamental products are inverted first and then are interpreted into secondary or derived products such as sea ice concentration, snow water equivalent (based on the retrieved emissivity) rainfall rate, total precipitable water, integrated cloud liquid amount, and ice water path (based on the retrieved atmospheric and hydrometeor products). The MiRS system was implemented operationally at the U.S. National Oceanic and Atmospheric Administration (NOAA) in 2007 for the NOAA-18 satellite. Since then, it has been extended to run for NOAA-19, Metop-A, and DMSP-F16 and F18 SSMI/S. This paper gives an overview of the system and presents brief results of the assessment effort for all fundamental and derived products.

Intercalibration Between Special Sensor Microwave Imager/Sounder and Special Sensor Microwave Imager

The F16 satellite was successfully launched on October 18, 2003, carrying the first special sensor microwave imager/sounder (SSMIS) onboard. In this paper, the SSMIS imaging channels 12-18 are intercalibated against the F15 special sensor microwave/imager (SSM/I) instrument using simultaneous conical overpassing (SCO) observations from both satellites. Results show that the SSMIS antenna temperatures have a mean bias as large as 1-2 K with a maximum of 3 K at 22.235 GHz with respect to F15. It appears that the mean biases at frequencies from 19.35 to 37 GHz do not strongly depend on the region and season, although the biases at the 91.655-GHz channels are slightly variable. The intercalibration analysis also shows that the nonlinearity may be one of the major sources resulting in differences between F15 SSM/I and F16 SSMIS measurements. For improved calibration and for the future SSM/I and SSMIS reprocessing, the SCO data are further utilized to resolve the SSMIS and SSM/I nonlinearity terms using a newly developed calibration algorithm. With the derived nonlinearity correction, the mean biases of the antenna temperatures between F15 and F16 are significantly reduced. To intercalibrate SSMIS to the same reference as SSM/I, SSMIS imaging channels can also be linearly mapped to the same and similar F15 SSM/I channels using the SCO matchup data. After the linear mapping, SSMIS snow-free land, snow, and sea ice surface emissivities are consistent with those derived from SSM/I.

Satellite Data Assimilation in Numerical Weather Prediction Models. Part II: Uses of Rain-Affected Radiances from Microwave Observations for Hurricane Vortex Analysis

Abstract A hybrid variational scheme (HVAR) is developed to produce the vortex analysis associated with tropical storms. This scheme allows for direct assimilation of rain-affected radiances from satellite microwave instruments. In the HVAR, the atmospheric temperature and surface parameters in the storms are derived from a one-dimension variational data assimilation (1DVAR) scheme, which minimizes the cost function of both background information and satellite measurements. In the minimization process, a radiative transfer model including scattering and emission is used for radiance simulation (see Part I of this study). Through the use of 4DVAR, atmospheric temperatures from the Advanced Microwave Sounding Unit (AMSU) and surface parameters from the Advanced Microwave Scanning Radiometer (AMSR-E) are assimilated into global forecast model outputs to produce an improved analysis. This new scheme is generally applicable for variable stages of storms. In the 2005 hurricane season, the HVAR was applied for t...

Special Sensor Microwave Imager (SSM/I) Intersensor Calibration Using a Simultaneous Conical Overpass Technique

Abstract A new intersensor calibration scheme is developed for the Defense Meteorological Satellite Program Special Sensor Microwave Imager (SSM/I) to correct its scan-angle-dependent bias, the radar calibration beacon interference on the F-15 satellite, and other intersensor biases. The intersensor bias is characterized by the simultaneous overpass measurements with the F-13 SSM/I as a reference. This sensor data record (SDR) intersensor calibration procedure is routinely running at the National Oceanic and Atmospheric Administration and is now used for reprocessing all SSM/I environmental data records (EDR), including total precipitable water (TPW) and surface precipitation. Results show that this scheme improves the consistency of the monthly SDR’s time series from different SSM/I sensors. Relative to the matched rain products from the Tropical Rainfall Measuring Mission, the bias of SSM/I monthly precipitation is reduced by 12% after intersensor calibration. TPW biases between sensors are reduced by 7...

Use of a One-Dimensional Variational Retrieval to Diagnose Estimates of Infrared and Microwave Surface Emissivity Over Land for ATOVS Sounding Instruments

A 1-D variational retrieval of surface emissivity is developed and applied for the Advanced Microwave Sounding Unit modules A and B, along with the High-resolution Infrared Radiation Sounder. This algorithm offers simultaneous retrieval of infrared and microwave emissivity and increases the separation of the emissivity and land surface temperature signals. The initial estimate of the emissivity for the surface-sensitive channels is made by a combination of physical and empirical microwave emissivity models and, in the infrared, by indexing laboratory measurements to vegetation databases. It is found that the initial estimates of emissivity for snow-free vegetated land areas are within 1% of the retrieved infrared values and, in the microwave, within 4% for all snow-free points and within 2% for the vast majority. The emissivity for snow-covered and sea-ice areas remains problematic, and further investigation is required. It is also shown that the average zenith angle dependence of the emissivity is less than 0.0024 for infrared wavelengths and less than 0.0055 for microwave frequencies if the viewing zenith angles greater than 40 are neglected.

A New Sea-Ice Concentration Algorithm Based on Microwave Surface Emissivities—Application to AMSU Measurements

Passive microwave sea-ice retrieval algorithms are typically tuned to brightness temperature measurements with simple treatments of weather effects. The new technique presented is a two-step algorithm that variationally retrieves surface emissivities from microwave remote sensing observations, followed by the retrieval of sea-ice concentration from surface emissivities. Surface emissivity spectra are interpreted for determining sea-ice fraction by comparison with a catalog of sea-ice emissivities to find the closest match. This catalog was computed off-line from known ocean, first-year, and multiyear sea-ice reference emissivities for a range of fractions. The technique was adjusted for application to the Advanced Microwave Sounding Unit (AMSU)/Microwave Humidity Sensor observations, and its performance was compared to the National Oceanic and Atmospheric Administration (NOAA)s AMSU heritage sea-ice algorithm and to NOAAs operational Interactive Multi-sensor Snow and Ice Mapping System taken as ground truth. Assessment results indicate a performance that is superior to the heritage algorithm particularly over multiyear ice and during the warm season.

Effects of Microwave Desert Surface Emissivity on AMSU-A Data Assimilation

A microwave land emissivity library has been developed from the Advanced Microwave Sounding Unit (AMSU) data for improving satellite data assimilation. Over the desert, surface emissivity is classified according to soil type into several spectra. For sand, loamy sand, and sandy loam, which contain some large mineral particles, the emissivity spectra generally decrease with frequency. For other desert types whose compositions are dominated by mineral particles smaller than a few hundred micrometers, the emissivity values are almost constant or slightly increasing with frequency. These emissivity features are consistent with those from the land emissivity data set developed at Météo-France. Moreover, both the emissivity library and the Météo-France data set are applied to the assimilation of the AMSU-A data in the National Centers for Environmental Prediction Global Forecast System (GFS). In comparison with the microwave land emissivity model previously developed by Weng , both the emissivity library and the Météo-France data set improve the utilization of the AMSU-A data in the GFS. The increased use of the AMSU-A data through the emissivity library or the data set results in positive impacts on the global medium-range forecasts over either the Southern or Northern Hemispheres.

Assessments of F16 Special Sensor Microwave Imager and Sounder Antenna Temperatures at Lower Atmospheric Sounding Channels

The main reflector of the Special Sensor Microwave Imager/Sounder (SSMIS) aboard the Defense Meteorological Satellite Program (DMSP) F-16 satellite emits variable radiation, and the SSMIS warm calibration load is intruded by direct and indirect solar radiation. These contamination sources produce antenna brightness temperature anomalies of around 2 K at SSMIS sounding channels which are obviously inappropriate for assimilation into numerical weather prediction models and remote sensing retrievals of atmospheric and surface parameters. In this study, antenna brightness temperature anomalies at several lower atmospheric sounding (LAS) channels are assessed, and the algorithm is developed for corrections of these antenna temperature anomalies. When compared against radiative transfer model simulations and simultaneous observations from AMSU-A aboard NOAA-16, the SSMIS antenna temperatures at 52.8, 53.6, 54.4, 55.5, 57.3, and 59.4 GHz after the anomaly correction exhibit small residual errors (<0.5 K). After such SSMIS antenna temperatures are applied to the National Center for Environmental Prediction Numerical Weather Prediction (NWP) model, more satellite data is used and the analysis field of the geopotential height is significantly improved throughout troposphere and lower stratosphere. Therefore, the SSMIS antenna temperatures after the anomaly correction have demonstrated their potentials in NWP models.

Ten-year (1993-2002) time-series of microwave land emissivity

A new algorithm is developed to retrieve the land surface emissivity at four SSM/I frequencies. The coefficients used in the algorithm are directly derived from a training data set including land emissivity and SSM/I measured brightness temperatures. The training data set is also derived from SSM/I brightness temperatures by removing the effects of atmospheric emission and surface temperature. In doing so, the upwelling and downwelling brightness temperature as well as atmospheric transmittance are computed directly from the global data assimilation system (GDAS) temperature and moisture profiles, and surface temperature. Global monthly mean emissivity maps are generated from the daily products. The retrievals under rain regions are not used in the averaging. The rainy pixels were screened out using SSM/I precipitation algorithm.