Column atmospheric water vapor and vegetation liquid water retrievals from Airborne Imaging Spectrometer data

Gao, Bo‐Cai; Goetz, Alexander

doi:10.1029/jd095id04p03549

Cited by 376 publications

(183 citation statements)

References 41 publications

(20 reference statements)

Supporting

Mentioning

177

Contrasting

Unclassified

Order By: Relevance

“…For an efficient and fast correction of image data acquired with multi-or hyperspectral sensors this code has been modified and extended, i.e., by integrating modules to account for varying terrain height and sensor altitude (Hill & Sturm, 1991), to estimate the aerosol optical thickness directly from dark objects in the image (Hill, 1993) and to correct for illumination effects (Hill et al, 1995). Recently, the software package has been refined to cope with specific requirements of airborne hyperspectral image data by adding a module that, taking up concepts originally developed by Gao and Goetz (1990), produces spatially distributed maps of atmospheric water vapour to be included into the atmospheric correction (Hill & Mehl, 2003).…”

Section: Image Data and Processingmentioning

confidence: 99%

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Schlerf¹,

Atzberger²,

Hill³

2005

Remote Sensing of Environment

277

171

View full text Add to dashboard Cite

This study systematically evaluated linear predictive models between vegetation indices (VI) derived from radiometrically corrected airborne imaging spectrometer (HyMap) data and field measurements of biophysical forest stand variables (n=40). Ratio-based and soil-linerelated broadband VI were calculated after HyMap reflectance had been spectrally resampled to Landsat TM channels. Hyperspectral VI involved all possible types of two-band combinations of ratio VI (RVI) and perpendicular VI (PVI) and the red edge inflection point (REIP) computed from two techniques, inverted Gaussian Model and Lagrange Interpolation. Cross-validation procedure was used to assess the prediction power of the regression models. Analyses were performed on the entire data set or on subsets stratified according to stand age. A PVI based on wavebands at 1088 nm and 1148 nm was linearly related to leaf area index (LAI) (R 2 =0.67, RMSE=0.69 m 2 m À2 (21% of the mean); after removal of one forest stand subjected to clearing measures: R 2 =0.77, RMSE=0.54 m 2 m À2 (17% of the mean). A PVI based on wavebands at 885 nm and 948 nm was linearly related to the crown volume (VOL) (R 2 =0.79, RMSE=0.52). VOL was derived from measured biophysical variables through factor analysis (varimax rotation). The study demonstrates that for hyperspectral image data, linear regression models can be applied to quantify LAI and VOL with good accuracy. For broadband multispectral data, the accuracy was generally lower. It can be stated that the hyperspectral data set contains more information relevant to the estimation of the forest stand variables LAI and VOL than multispectral data. When the pooled data set was analysed, soil-line-related VI performed better than ratio-based VI. When age classes were analysed separately, hyperspectral VI performed considerably better than broadband VI. Best hyperspectral VI in relation with LAI were typically based on wavebands related to prominent water absorption features. Such VI are related to the total amount of canopy water; as the leaf water content is considered to be relatively constant in the study area, variations of LAI are retrieved. D

show abstract

Section: Image Data and Processingmentioning

confidence: 99%

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Schlerf¹,

Atzberger²,

Hill³

2005

Remote Sensing of Environment

277

171

View full text Add to dashboard Cite

show abstract

“…Further corrections for SWIR2 reflectance were made using a large (dark) lake and (bright) cumulus clouds. Assuming both targets to be spectrally flat in the SWIR2 (Gao and Goetz [1990] and our field spectra from lake), the ATREM-corrected reflectance values were used to create a gain and offset for each SWIR2 band, which were then applied to complete the reflectance calibration. This final step was necessary to remove apparent errors in the ATREM correction for methane, which strongly absorbs radiation beginning at ---2200 nm.…”

Section: Aviris Datamentioning

confidence: 99%

Subpixel canopy cover estimation of coniferous forests in Oregon using SWIR imaging spectrometry

Lobell¹,

Asner²,

Law³

et al. 2001

J. Geophys. Res.

View full text Add to dashboard Cite

Abstract. The percent cover of vegetation canopies is an important variable for many land-surface biophysical and biogeochemical models and serves as a useful measure of land cover change. Remote sensing methods to estimate the subpixel fraction of vegetation canopies with spectral mixture analysis (SMA) require knowledge of the reflectance properties of major land cover units, called endmembers. However, variability in endmember reflectance across space and time has 14mited the interpretation and general applicability of SMA approaches. In this study, a subpixel vegetation cover of coniferous forests in Oregon, United States, was successfully estimated by employing shortwave infrared reflectance measurements (SWIR2 region, 2080-2280 nm) collected by the NASA Airborne Visible Infrared Imaging Spectrometer (AVIRIS). The approach presented here, referred to as AutoSWIR [Asner and Lobell, 2000], was originally developed for semiarid and arid environments and exploits the low SWIR2 variability of materials found in most ecosystems. SWIR2 field spectra from Oregon were compared with spectra from an arid systems database, revealing significant differences only for soil reflectance. However, SWIR2 variability remained low, as indicated by field spectra and principal component analysis, and AutoSWIR was then modified to use coniferous forest spectra collected in Oregon. Subsequent high spatial resolution estimates of forest canopy cover agreed well with estimates from low-altitude air photos (rms = 3%), demonstrating the successful extension of AutoSWIR to a coniferous forest ecosystem. The generality of AutoSWIR facilitates accurate estimates of vegetation cover that can be automatically retrieved from SWIR2 spectral measurements collected by forthcoming spaceborne imaging spectrometers such as NASA's New Millenium Program EO-1 Hyperion. These estimates can then be used to characterize landscape heterogeneity important for landsurface, atmospheric, and biogeochemical research. . However, unresolved heterogeneity in vegetation properties in these ecosystems can be significant; Kimball et al. [1999], for instance, found that subpixel-scale land cover complexity in boreal forests could lead to annual net primary production (NPP) errors of more than 14% using an ecosystem process model. In another study, Bonan et al. [1993] showed that landscape heterogeneity in coniferous forests had a highly nonlinear effect on sensible heat and evapotranspiration calculations, with errors as high as 46% and 15%, respectively. The ability to accurately estimate vegetation cover in coniferous forests, and thereby constrain models used to understand land-surface, atmospheric, and biogeochemical processes, is of great importance in these regions. Remote sensing is a relatively cheap and fast method for estimating showed that the reflectance variability of major land cover types (e.g., green vegetation, senescent vegetation, and soil) was dominated by variation in overall albedo of the shortwave region from 400 to 2500 nm. They also showed ...

show abstract

“…The correction is derived from the ratios of the 0.935-and 0.905-m reflectances obtained from the MODIS on Aqua for a sample of the regions observed with the AVHRRs. The reflectance ratio R 0.935 /R 0.905 is a measure of the column water vapor burden (Gao and Goetz 1990). The correction used here is derived from the R 0.935 /R 0.905 ratios using results presented by Heidinger et al (2002, their Fig.…”

Section: Noaa-16 and Noaa-17 064-and 084-m Reflectancesmentioning

confidence: 99%

Using Sun Glint to Check the Relative Calibration of Reflected Spectral Radiances

Luderer

Coakley

Tahnk

2005

Journal of Atmospheric and Oceanic Technology

View full text Add to dashboard Cite

Observations of sunlight reflected from regions of sun glint are used to check the relative calibration of spectral radiances obtained with imaging radiometers. Reflectances at different wavelengths for sun-glint regions are linearly related. Provided that the atmosphere is reasonably transparent at the wavelengths, the aerosol burden is reasonably light, 0.64-m optical depth less than 0.2; the particles constituting the aerosol are reasonably large, as is the case for marine aerosols; and the solar zenith angle is less than about 35°, the linear relationships between reflectances at different wavelengths are rather insensitive to the factors that govern the reflectances themselves. The relationships are remarkably insensitive to atmospheric composition, surface wind speed and direction, illumination, and viewing geometry. The slopes and offsets of the linear relationships are used to assess the relative accuracies of the calibrations of the different channels. Such assessments would appear to be attractive for checks on the in-flight performance of aircraft-borne imaging radiometers. Here, observations of reflectances at 0.64, 0.84, 1.6, and 2.1 m for regions of sun glint obtained with the Terra and Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) instruments are shown to be consistent with each other. Observations of the 0.64-and 1.6-m reflectances for the Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS) instrument are shown to be inconsistent with the MODIS observations, the VIRS 1.6-m gain appearing to be too low by 9%. The 0.64-, 0.84-, and 1.6-m reflectances obtained with the NOAA-16 and NOAA-17 Advanced Very High Resolution Radiometers (AVHRRs) for December 2002 are shown to be inconsistent with each other and inconsistent with the MODIS observations. Based on observations of the extensive ice sheets of Antarctica, the NOAA-16 0.64-m gain is found to be too low by 5% and that for the 0.84-m reflectance is too low by 12%; the NOAA-17 0.64-m gain is found to be accurate (within 2%), but the 0.84-m gain is too low by 15%. With the gains adjusted, the 0.64-and 0.84-m reflectances obtained for regions of sun glint with the AVHRRs are consistent with each other and consistent with the Terra and Aqua MODIS observations. These results suggest that the gain for the NOAA-16 AVHRR 1.6-m reflectance is accurate (within 1%) and that for the NOAA-17 AVHRR is too low by 5%. All of the observations were made with the AVHRR in the low-reflectance (high gain) mode. The accuracy of these assessments is expected to be about 5%.

show abstract

Column atmospheric water vapor and vegetation liquid water retrievals from Airborne Imaging Spectrometer data

Cited by 376 publications

References 41 publications

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Remote sensing of forest biophysical variables using HyMap imaging spectrometer data

Subpixel canopy cover estimation of coniferous forests in Oregon using SWIR imaging spectrometry

Using Sun Glint to Check the Relative Calibration of Reflected Spectral Radiances

Contact Info

Product

Resources

About