2009 STAR Seminars

The Hyperion on NASA's EO-1 satellite launched in 2000 is among the few Earth-observing hyperspectral instruments in the 400- 2500nm spectral range with a 30 meter spatial resolution. In addition to its wide ranging applications in mining, geology, forestry, agriculture, and environmental management, Hyperion measurements are especially useful for characterizing vicarious calibration sites to resolve spectral related calibration issues. This seminar will introduce the calibration of the Hyperion instrument using both onboard devices and lunar observations, and the instrument performance since launch. Its applications to the vicarious site characterization such as the Dome C and Desert sites, as well as the benefits to NOAA operational instruments will be discussed.

A suite of products has been developed and evaluated to assess hazards presented by convective downbursts to aircraft in flight derived from the current generation of Geostationary Operational Environmental Satellite (GOES) (11-P). The existing suite of GOES microburst products employs the GOES sounder to calculate risk based on conceptual models of favorable environmental profiles for convective downburst generation. Recent testing and validation have found that the GOES microburst products are effective in the assessment and short-term forecasting of downburst potential and associated wind gust magnitude. Two products, the Microburst Windspeed Potential Index (MWPI) and a multispectral GOES imager product, have demonstrated capability in downburst potential assessment. Both the GOES sounder MWPI and imager microburst risk products are predictive linear models that consist of a set of predictor variables that generates output of expected microburst risk. This presentation compares and contrasts the sounder and imager microburst products and outlines the advantages of each product in the nowcasting process. An updated assessment of the sounder MWPI and imager microburst products, case studies demonstrating effective operational use of the microburst products, and validation results is presented.

Climate monitoring and research require development of thematic long-term satellite data products as well as comprehensive reanalysis data products. However, calibration and data consistency have been a major issue in producing reliable satellite and reanalysis climate products. Many long-term satellite climate products suffer from spurious climate jumps induced by satellite transition and calibration-related instrument changes. To reconcile the problem, intercalibration and reprocessing are required for intersatellite biases removal before satellite data are used for climate analysis and reanalysis data assimilation. For these purposes, NOAA/NESDIS is recalibrating MSU/AMSU/SSU observations from NOAA, NASA, and MetOp orbiting satellite series. Climate quality atmospheric temperature climate data records (CDRs) are generated from these recalibration/reprocessing effort.

This talk will review the current status on the reprocessing of 30-year MSU/AMSU/SSU data using simultaneous nadir overpass methods. We introduce a well- intercalibrated MSU/AMSU atmospheric temperature CDR for climate change monitoring. We discuss and present our proposed solutions for bias correction issues in the MSU/AMSU/SSU CDR development that includes elimination of the warm target contamination, limb adjustment, diurnal drift adjustment, short overlap problem, channel frequency differences, residual bias removal, and CO₂ leaking problem in the SSU cell pressure, etc. Updated 30-year atmospheric temperature trends derived from the MSU/AMSU CDRs will be presented. We also propose some methods/principles for testing reliability of the MSU/AMSU-derived climate trends. Finally, a science team on the MSU/AMSU/SSU CDRs is established under NOAA Scientific Data Stewardship Program and we discuss plans for the team work.

Accurate representation of the regional characteristics of anisotropic light scattering by land surfaces under a wide range of sky conditions is required (1) for modeling atmospheric shortwave radiative fluxes; (2) for modeling the energy exchange between the earth and atmosphere; and (3) for determining the lower boundary conditions for atmospheric radiative transfer models. However, uncertainties arise when satellite retrievals of surface bidirectional reflectance distribution function (BRDF) are directly compared against in-situ observations. In particular, the spatial variability of ground-based estimates of the BRDF introduces errors within the footprint of satellite sensor retrievals that are very difficult to quantify and are oftentimes ignored. Empirical quality of BRDF data is rarely certain and knowledge of their uncertainties is essential to understand its effect on higher-level surface biophysical products (e.g. vegetation indexes, surface albedo, LAI/FPAR, burned area, land cover, and land cover change). This would enable robust accuracy assessments that include evaluations of measurement, scaling, and analytical (or model-driven) errors. Linking airborne angular reflectance measurements for a given surface location yields the underlying reflectance anisotropy (or BRDF shape) of that location.

This talk will outline an algorithm suitable for such a task using airborne angular reflectance measurements available from NASA's Cloud Absorption Radiometer (CAR); a 14-channel airborne scanning radiometer with a spectral range from 0.331-2.345µm. This information was used to quantify the differences in the directional reflectance data, and related measures of vegetation structure, at multiple spatial scales. A new set of gridding functions were also created to exploit the geometric efficiency of CAR observations. The routines allocate the airborne angular reflectance measurements acquired by the CAR into the most frequently sampled spatial intervals obtained for a given flight path. Under well-planned flight scenarios, this technique can be used to derive a combination of one-of-a-kind maps of the underlying reflectance anisotropy that are optimized to a specific spatial scale. This enables "datamatchups" with ancillary data sources (e.g., land use/cover maps), thereby improving the utility of the CAR retrievals in regional mapping and characterization of terrestrial ecosystems.

NOAA leadership's recent policy guidance gives effective science communication a very high priority. The purpose of this talk is to discuss strategies and techniques for getting news about STAR science and accomplishments out to a wider audience across all the available media channels. We will discuss how to get STAR news content to NOAA and NESDIS, as well as STAR, communication decision makers, and to empower effective participation of STAR scientists in this process.

Sea ice is an important indicator of climate change, and a key component of the polar climate system. Ongoing loss of Arctic sea ice has serious implications for climate change, ocean circulation, the Arctic ecosystem, and economic development in the region. A real shrinkage of Arctic sea ice has been observed over the last three decades, and its decline is now proceeding faster than forecasted. A record minimum ice extent was reached in September 2007. The latest satellite observations of sea ice freeboard also reveal a decline in ice thickness, in line with the observed changes in ice extent and the loss of multiyear ice.

An extensive monitoring of Arctic-wide sea ice thinning using satellite altimeters is now necessary to determine whether such observations are part of a sustained negative trend in Arctic ice thickness or a reflection of the natural, interannual variability. It is key to first validate satellite altimeter data over sea ice. We achieve this by making comparisons with "ground-truth" observations gathered from low altitude aircraft under-flights and in-situ measurements collected on the sea ice itself. We will discuss recent validation experiments which we have conducted in the Arctic, with particular emphasis on the Canada Basin Sea Ice Thickness (CBSIT) experiment completed earlier this year.

This seminar will include a review of the land-related data and products available from the USGS Center for Earth Resources Observation and Science (EROS) facility and the USGS/NASA Land Processes Distributed Active Archive Center (LP DAAC) that may be applicable for STAR research activities. Data sets and products reviewed will primarily include those available from the Landsat, MODIS and ASTER sensors. These sensors have spatial resolutions that range from 15m to 1000m and temporal resolutions from 1 to 16 days. An update on the activities of the CEOS Land Surface Imaging Constellation will also be discussed.

Over the past few years advances in technology, and applied mathematics have enabled essentially new approaches to scientific questions. Improvements in sensor technology, cheap digital storage, fast networking, an growing international collaborations have provided scientists with access to unprecedented quantities of high quality data. Not only have technological advancements have improved access to vast amounts of data, but advances in cluster, cloud and supercomputing have increased access to previously unimagined computational power. To exploit this new computational power and data wealth, it is becoming clear that statistical tools which can cluster, classify and perform probabilistic inference on large high dimensional data sets are become crucial scientific tools.

For NOAA applications the approaches suggested by statistical data analysis complement and support physical analysis and models. Helping search for and quantify, complex and subtle relationships in high dimensional data, they can serve as a starting point for forecasters, atmospheric and marine scientists for further investigation, as well as providing functionally predictive models. Our group at CCNY has built expertise collaborating with NOAA scientists to apply statistical pattern analysis toward compression of multi-spectral imager and sounder data.

In this talk we will show how we are applying similar methods, in collaboration and at the suggestion of NOAA scientists, to several important tasks such as quantitative restoration of multi- spectral imager from damaged sensors, detection of algae blooms, and quantitatively estimating missing channels. We will also show a new tool we have developed, called Graphyte, which allows scientists globally to collaborate, access data, perform computation on high performance computers, interface extensive libraries written in C/C++, Fortran, Java, and Python, as well as create interactive visualizations

In July 2008, NOAA Coral Reef Watch launched a new seasonal prediction tool for coral bleaching conditions to augment its real-time satellite monitoring. A model of thermal stress from two weeks to three months in the future was developed through collaboration with the Physical Sciences Division of the NOAA Earth System Research Laboratory, to forecast the risk of coral bleaching well in advance of such events. Such forecasting tools provide critical and timely decision support for coral reef managers and scientists worldwide.

The work that will be presented has been largely supported by NOAA's Sectoral Applications Research Program (SARP).

In spite of the progress in space-borne aerosol sensing capability over the last decade, the quantification of the net effect of aerosols on the radiative transfer balance of the earth- atmosphere system remains a challenge. The development of the capability to detect aerosol absorption from space using near-UV observations was one the most important breakthroughs of the last decade in aerosol remote sensing from space. The technique, developed from analysis of observations by the TOMS instrument has been extensively used for the global mapping of absorbing aerosols. Aerosol absorption can be measured from space in the near UV by taking advantage of the interaction between Rayleigh scattering and particle absorption. The OMI near-UV retrieval algorithm draws its heritage from a similar inversion procedure. The OMI near-UV aerosol algorithm retrieves information on column aerosol properties making use of the backscattered radiances at 342.5, 388 and 471 nm. Details of this retrieval and its properties as well as directions for new product development will be covered in this talk.

Recent evidence supports the idea that sub-surface ocean thermal structure plays an important role in modulating air-sea fluxes during hurricane passage which in turn affects intensity change. Given the sparsely of in situ data, it has been difficult to provide region to basin-wide estimates of isotherm depths and upper ocean heat content variability. Satellite-derived sea surface height anomalies (SHA) from multiple platforms carrying radar altimeters are updated each day with the latest track information. These multiple-platform data are objectively analyzed and compared to other standard products (e.g., AVISO) to ensure consistency with the surface height fields. During Katrina and Rita (2005), for example, biases between these products were 2 to 5 cm with regression slopes of O(1). The analyzed SHA field is combined with a hurricane season climatology (June through November), to estimate isotherm depths and oceanic heat content (OHC) using a reduced gravity model at 0.25° intervals in the Atlantic and Eastern Pacific Oceans where rapid intensity change is often observed.

Thermal structure measurements from aircraft experiments, long- term moorings and volunteer observing ships are used to carefully evaluate this satellite-derived upper ocean variability. Regression statistics reveal small biases with slopes of O(0.9) between the subsurface measurements compared with isotherm depths (20 and 26°C) and OHC fields. Root-mean-square differences in OHC range between 10 to 15 kJ cm^-2 or roughly 10 to 15% of the mean signals. Similar values are found for isotherm depth differences between in situ and satellite-derived values. Updated, daily OHC estimates are input into the Statistical Hurricane Intensity Prediction Scheme (SHIPS) for intensity forecasts at the National Hurricane Center. The OHC values have been shown to reduce error in the SHIPS intensity forecasts between 5 to 22% in the western part of the Atlantic Ocean basin (e.g., Ivan).

In the remote retrieval of the ocean (and inland) water near- surface properties, it is important to accurately remove the atmospheric and water surface effects from satellite sensor- measured signals. This process, which corrects more than 90% of satellite sensor-measured signals, is termed as atmospheric correction. The NASA standard atmospheric correction algorithm for Sea-viewing Wide Field-of-view Sensor (SeaWiFS) and Moderate Resolution Imaging Spectroradiometer (MODIS) uses two near- infrared (NIR) bands for retrieval of aerosol properties with assumption of the black water at the NIR wavelengths. SeaWiFS (1997-present) and MODIS-Aqua (2002-present) have been producing high quality ocean color products over global open oceans. For the turbid waters in the ocean coastal regions (and inland lakes), however, water could have significant contributions in the NIR bands, leading to significant errors in the satellite-derived water property products. Recently, an atmospheric correction algorithm using the shortwave infrared (SWIR) bands has been developed for producing improved water optical and biological properties over turbid waters. In this presentation, I provide overview of the SeaWiFS/MODIS atmospheric correction algorithm that is currently used for deriving the ocean color products. The new approach using the SWIR bands for atmospheric correction is then described. I will demonstrate advantages of the new approach by comparing water optical and biological property results derived from the SWIR atmospheric correction algorithm and from the standard (NIR) algorithm. Some specific applications for deriving ocean color products along the China east coastal regions, as well as for monitoring and assessment of Lake Taihu blue-green algae bloom during the spring of 2007, will be presented and discussed.

This talk will provide an overview of some current assimilation projects being accomplished in the Atmospheric and Oceanic Physics Department of the Applied Research Laboratory and the Meteorology Department of The Pennsylvania State University. In addition to using various standard assimilation techniques, including Newtonian relaxation, Extended Kalman Filter, Ensemble Kalman Filter, and 4DVAR, the group has developed a new GAVAR method that sets up an optimization problem and solves it using the robust Genetic Algorithm. Most of the research has been accomplished in the context of atmospheric transport and dispersion, emphasizing obtaining wind field variables given observations of pollutant concentrations. In spite of the one-way coupling, we have been able to consistently infer winds from concentration observation. Applications of this technique include back-calculating unknown source parameters and meteorological parameters. A current project is expanding the technique to larger scales by characterizing volcano emissions using satellite observations, modeling, and assimilation techniques. The team is also using assimilation techniques for downscaling. In addition, field studies are characterizing the variability in smoke plumes.

For over a decade the Marine Optical Buoy (MOBY) has been the primary vicarious calibration facility for satellite ocean color observations. Approximately 5% of the ocean color signal that is measured by a satellite (Lt) originates from the sea surface. Thus, we must resolve small variations in a large signal to derive any meaningful information from ocean color satellite imagery. In order to measure ocean color with the accuracies necessary to meet NOAA's mission goal requirements, vicarious calibration using highly calibrated and well characterized instrumentation is required. MOBY has provided this level of high quality measurements since the launch of SeaWiFS in 1997. MOBY is located in coastal Hawaiian waters near the island of Lanai, and has collected near continuous upwelled submarine light measurements that are used to calculate the water-leaving radiances that are measured by satellite. MOBY calibrations are NIST traceable and provide a vital climate quality data link between SeaWiFS, MODIS and foreign sensors and will continue that connection into the VIIRS NPP/NPOESS era. This talk will present the history and need for MOBY, provide details into the operations and calibrations of MOBY, and will give plans for a technology refresh in the near future.

Coral Reef Watch (CRW) has been widely praised for its coral bleaching product suite. It has been extensively used by US and international reef managers and lawmakers to predict and understand the onset and severity of mass coral bleaching. The current suite of algorithms is based solely on satellite sea surface temperature (SST) retrievals. While they accurately predict the onset of coral bleaching and give a good indication of the severity of the event, they do not accurately predict mortality and have no ability to distinguish differential responses among various coral species. What we know as thermal coral bleaching is caused by accumulated light stress, and the sensitivity of corals to light is modulated by temperature. As SST is a function of incoming solar radiation, the current CRW SST-based product suite indirectly includes light. Our knowledge of coral physiology has come a long way in the last decade and most of the processes causing coral bleaching are now much better understood. The international World Bank/GEF funded Coral Reef Targeted Research programme provided CRW with the opportunity to team up with the world's foremost experts in coral physiology of coral bleaching and begin the development of a satellite product that combines light and temperature. At the same time, work at STAR has made satellite measures of surface light over the oceans possible. It is hoped that this product will improve our ability to predict the severity and mortality of coral bleaching and will also provide information on the levels of stress needed to bleach various species. This seminar will describe the algorithm, which is soon to be implemented as an experimental satellite product.

Climate models suggest that the meridional overturning circulation (MOC) in the Atlantic, and the accompanying oceanic heat flux, vary considerably on interannual time scales. In addition to abrupt climate change scenarios in which the MOC can virtually shut off (Manabe and Stouffer, 1993; Vellinga and Wood, 2002), the "normal" interdecadal variation may range from 20% to 30% of its long-term mean value, according to some models (e.g., Hakkinen, 1999). However, until recently no direct measurement system had been put in place that could provide regular estimates of the meridional overturning circulation to determine its natural variability or to assess these model predictions. Such a system is now deployed along 26.5°N in the Atlantic as part of the joint U.K./U.S. RAPID-MOCHA program, which has been continuously observing the MOC since March 2004. This presentation will describe this program and the scientific results achieved so far.

Uncertainty associated with vertical gas exchange at ocean surface is a major contributor of uncertainty in global carbon budget assessment. The estimate of ocean carbon uptake varies from 1.1 PgC/yr (Liss and Merlivat (1986) to 3.3 PgC/yr (Wanninkhof and McGillis, 1999) as a result of difference in air-sea gas exchange estimates.

High resolution lidar measurements of ocean surface roughness may lead to significant reduction in global air-sea gas exchange uncertainty. Air-sea exchange is linearly proportional to wave slopes at all wave scales (wave number ranging from 50 to 800 rad/m), especially the smaller scale waves such as capillary waves (Frew et al., 2003). The air-sea gas exchange is currently parameterized to wind information associated with microwave measurements (such as QuikScat and AMSR-E). Microwave measurement of ocean surface roughness is directly related to lower frequency surface waves (<50 rad/m). The link between microwave measurement and higher frequency waves is nonlinear. At shorter wavelengths (1 micron), lidar measures wave slope variance of all waves more directly. Thus it provides direct and accurate gas exchange information. High resolution near surface wind speed can also be derived from the lidar ocean surface roughness measurements (Hu et al. 2008).

One of the shortcomings of satellite based lidar measurement (such as CALIPSO) is its limited spatial coverage (nadir or near nadir only). It is thus highly desirable to study global gas exchange and near surface wind with combine lidar/SAR measurements since SAR provides high spatial ocean surface backscatter at a wider swath. This talk intends to introduce the lidar ocean surface roughness measurements from CALIPSO, and to initiate discussions on potential collaborations between NOAA and NASA in the field of high resolution near surface wind and gas exchange studies with combined lidar/SAR measurements.