Comparison and validation of MODIS and VEGETATION global LAI products over four BigFoot sites in North America

A new set of recently developed leaf area index (LAI) algorithms has been employed for producing a global LAI dataset at 1 km resolution and in time-steps of 10 days, using data from the Satellite pour l'observation de la terre (SPOT) VEGETATION (VGT) sensor. In this paper, this new LAI product is compared with the global MODIS Collection 4 LAI product over four validation sites in North America. The accuracy of both LAI products is assessed against seven high resolution ETM+ LAI maps derived from field measurements in 2000, 2001, and 2003. Both products were closely matched outside growing season. The MODIS product tended to be more variable than the VGT product during the summer period when the LAI was maximum. VGT and ETM+ LAI maps agreed well at three out of the four sites. The median relative absolute error of the VGT LAI product varied from 24% to 75% at 1 km scale and it ranged from 34% to 88% for the MODIS LAI product. The importance of correcting field measurements for the clumping effect is illustrated at the deciduous broadleaf forest site (HARV). Inclusion of the sub-pixel land cover information improved the quality of LAI estimates for the prairie grassland KONZ site. Further improvement of the global VGT LAI product is suggested by production and inclusion of pixel-specific global foliage clumping index and forest background reflectance maps that would serve as an input into the VGT LAI algorithms.     

A hybrid inversion method for mapping leaf area index from MODIS data: experiments and application to broadleaf and needleleaf canopies

Leaf area index (LAI) is an important variable needed by various land surface process models. It has been produced operationally from the Moderate Resolution Imaging Spectroradiometer (MODIS) data using a look-up table (LUT) method, but the inversion accuracy still needs significant improvements. We propose an alternative method in this study that integrates both the radiative transfer (RT) simulation and nonparametric regression methods. Two nonparametric regression methods (i.e., the neural network [NN] and the projection pursuit regression [PPR]) were examined. An integrated database was constructed from radiative transfer simulations tuned for two broad biome categories (broadleaf and needleleaf vegetations). A new soil reflectance index (SRI) and analytically simulated leaf optical properties were used in the parameterization process. This algorithm was tested in two sites, one at Maryland, USA, a middle latitude temperate agricultural area, and the other at Canada, a boreal forest site, and LAI was accurately estimated. The derived LAI maps were also compared with those from MODIS science team and ETM+ data. The MODIS standard LAI products were found consistent with our results for broadleaf crops, needleleaf forest, and other cover types, but overestimated broadleaf forest by 2.0-3.0 due to the complex biome types.     

Comparison of burned area estimates derived from SPOT-VEGETATION and Landsat ETM+ data in Africa: Influence of spatial pattern and vegetation type

An algorithm for burned area mapping in Africa based on classification trees was developed using SPOT-VEGETATION (VGT) imagery. The derived 1 km spatial resolution burned area maps were compared with 30 m spatial resolution maps obtained with 13 Landsat ETM+ scenes, through linear regression analysis. The procedure quantifies the bias in burned area estimation present in the low spatial resolution burned area map. Good correspondence was observed for seven sites, with values of the coefficient of determination (R²) ranging from 0.787 to 0.983. Poorer agreement was observed in four sites (R² values between 0.257 and 0.417), and intermediate values of R² (0.670 and 0.613) were obtained for two sites. The observed variation in the level of agreement between the Landsat and VGT estimates of area burned results from differences in the spatial pattern and size distribution of burns in the different fire regimes encompassed by our analysis. Small and fragmented burned areas result in large underestimation at 1 km spatial resolution. When large and compact burned areas dominate the landscape, VGT estimates of burned area are accurate, although in certain situations there is some overestimation. Accuracy of VGT burned area estimates also depends on vegetation type. Results showed that in forest ecosystems VGT maps underestimate substantially the amount of burned area. The most accurate estimates were obtained for woodlands and grasslands. An overall linear regression fitted with the data from the 13 comparison sites revealed that there is a strong relationship between VGT and Landsat estimates of burned area, with a value of R² of 0.754 and a slope of 0.803. Our findings indicate that burned area mapping based on 1 km spatial resolution VGT data provides adequate regional information.     

Using lidar and effective LAI data to evaluate IKONOS and Landsat 7 ETM+ vegetation cover estimates in a ponderosa pine forest

Structural and functional analyses of ecosystems benefit when high accuracy vegetation coverages can be derived over large areas. In this study, we utilize IKONOS, Landsat 7 ETM+, and airborne scanning light detection and ranging (lidar) to quantify coniferous forest and understory grass coverages in a ponderosa pine (Pinus ponderosa) dominated ecosystem in the Black Hills of South Dakota. Linear spectral mixture analyses of IKONOS and ETM+ data were used to isolate spectral endmembers (bare soil, understory grass, and tree/shade) and calculate their subpixel fractional coverages. We then compared these endmember cover estimates to similar cover estimates derived from lidar data and field measures. The IKONOS-derived tree/shade fraction was significantly correlated with the field-measured canopy effective leaf area index (LAI_e) (r²=0.55, p<0.001) and with the lidar-derived estimate of tree occurrence (r²=0.79, p<0.001). The enhanced vegetation index (EVI) calculated from IKONOS imagery showed a negative correlation with the field measured tree canopy effective LAI and lidar tree cover response (r²=0.30, r=−0.55 and r²=0.41, r=−0.64, respectively; p<0.001) and further analyses indicate a strong linear relationship between EVI and the IKONOS-derived grass fraction (r²=0.99, p<0.001). We also found that using EVI resulted in better agreement with the subpixel vegetation fractions in this ecosystem than using normalized difference of vegetation index (NDVI). Coarsening the IKONOS data to 30 m resolution imagery revealed a stronger relationship with lidar tree measures (r²=0.77, p<0.001) than at 4 m resolution (r²=0.58, p<0.001). Unmixed tree/shade fractions derived from 30 m resolution ETM+ imagery also showed a significant correlation with the lidar data (r²=0.66, p<0.001). These results demonstrate the power of using high resolution lidar data to validate spectral unmixing results of satellite imagery, and indicate that IKONOS data and Landsat 7 ETM+ data both can serve to make the important distinction between tree/shade coverage and exposed understory grass coverage during peak summertime greenness in a ponderosa pine forest ecosystem.     

Ganges and Indus river basin land use/land cover (LULC) and irrigated area mapping using continuous streams of MODIS data

The overarching goal of this study was to map irrigated areas in the Ganges and Indus river basins using near-continuous time-series (8-day), 500-m resolution, 7-band MODIS land data for 2001-2002. A multitemporal analysis was conducted, based on a mega file of 294 wavebands, made from 42 MODIS images each of 7 bands. Complementary field data were gathered from 196 locations. The study began with the development of two cloud removal algorithms (CRAs) for MODIS 7-band reflectivity data, named: (a) blue-band minimum reflectivity threshold and (b) visible-band minimum reflectivity threshold.A series of innovative methods and approaches were introduced to analyze time-series MODIS data and consisted of: (a) brightness-greenness-wetness (BGW) RED-NIR 2-dimensional feature space (2-d FS) plots for each of the 42 dates, (b) end-member (spectral angle) analysis using RED-NIR single date (RN-SD) plots, (c) combining several RN-SDs in a single plot to develop RED-NIR multidate (RN-MDs) plots in order to help track changes in magnitude and direction of spectral classes in 2-d FS, (d) introduction of a unique concept of space-time spiral curves (ST-SCs) to continuously track class dynamics over time and space and to determine class separability at various time periods within and across seasons, and (e) to establish unique class signatures based on NDVI (CS-NDVI) and/or multiband reflectivity (CS-MBR), for each class, and demonstrate their intra- and inter-seasonal and intra- and inter-year characteristics. The results from these techniques and methods enabled us to gather precise information on onset-peak-senescence-duration of each irrigated and rainfed classes.The resulting 29 land use/land cover (LULC) map consisted of 6 unique irrigated area classes in the total study area of 133,021,156 ha within the Ganges and Indus basins. Of this, the net irrigated area was estimated as 33.08 million hectares—26.6% by canals and 73.4z5 by groundwater. Of the 33.08 Mha, 98.4% of the area was irrigated during khariff (Southwest monsoonal rainy season during June-October), 92.5% irrigated during Rabi (Northeast monsoonal rainy season during November-February), and only 3.5% continuously through the year.Quantitative Fuzzy Classification Accuracy Assessment (QFCAA) showed that the accuracies of the 29 classes varied from 56% to 100%—with 17 classes above 80% accurate and 23 classes above 70% accurate.The MODIS band 5 centered at 1240 nm provided the best separability in mapping irrigated area classes, followed by bands 2 (centered at 859 nm), 7 (2130 nm) and 6 (1640 nm).     

Spatially and temporally continuous LAI data sets based on an integrated filtering method: Examples from North America

Leaf Area Index (LAI) is an important biophysical variable for characterizing the land surface vegetation. Global LAI product has been routinely produced from the MODerate resolution Imaging Spectroradiometer (MODIS) aboard the Terra and Aqua satellite platforms. However, the MODIS standard LAI product is not continuous both spatially and temporally. To fill the gaps and improve the quality, we have developed a data filtering algorithm. This filter, called the temporal spatial filter (TSF), integrates both spatial and temporal characteristics for different plant functional types. The spatial gaps are first filled with the multi-year averages of the same day. If the values are missing over all years, the pixel is filled with a new estimate using the vegetation continuous field-ecosystem curve fitting method. The TSF integrates both the multi-seasonal average trend (background) and the seasonal observation. We implement this algorithm using the MODIS Collection 4 LAI product over North America. Comparison of the TSF results with the Savitzky-Golay filter indicates that the TSF performs much better in restoring the spatial and temporal distribution of seasonal LAI trends. The new LAI product has been validated by comparing with field measurements and the derived LAI maps from ETM+ data at a broadleaf forest site and an agricultural site. The validation results indicate that the new LAI product agrees better with both the field measurements and LAI values obtained from the ETM+ than does the MODIS LAI standard product, which usually shows higher LAI values.     

A scalable approach to mapping annual land cover at 250 m using MODIS time series data: A case study in the Dry Chaco ecoregion of South America

Land use and land cover (LULC) maps from remote sensing are vital for monitoring, understanding and predicting the effects of complex human-nature interactions that span local, regional and global scales. We present a method to map annual LULC at a regional spatial scale with source data and processing techniques that permit scaling to broader spatial and temporal scales, while maintaining a consistent classification scheme and accuracy. Using the Dry Chaco ecoregion in Argentina, Bolivia and Paraguay as a test site, we derived a suite of predictor variables from 2001 to 2007 from the MODIS 250 m vegetation index product (MOD13Q1). These variables included: annual statistics of red, near infrared, and enhanced vegetation index (EVI), phenological metrics derived from EVI time series data, and slope and elevation. For reference data, we visually interpreted percent cover of eight classes at locations with high-resolution QuickBird imagery in Google Earth. An adjustable majority cover threshold was used to assign samples to a dominant class. When compared to field data, we found this imagery to have georeferencing error < 5% the length of a MODIS pixel, while most class interpretation error was related to confusion between agriculture and herbaceous vegetation. We used the Random Forests classifier to identify the best sets of predictor variables and percent cover thresholds for discriminating our LULC classes. The best variable set included all predictor variables and a cover threshold of 80%. This optimal Random Forests was used to map LULC for each year between 2001 and 2007, followed by a per-pixel, 3-year temporal filter to remove disallowed LULC transitions. Our sequence of maps had an overall accuracy of 79.3%, producer accuracy from 51.4% (plantation) to 95.8% (woody vegetation), and user accuracy from 58.9% (herbaceous vegetation) to 100.0% (water). We attributed map class confusion to limited spectral information, sub-pixel spectral mixing, georeferencing error and human error in interpreting reference samples. We used our maps to assess woody vegetation change in the Dry Chaco from 2002 to 2006, which was characterized by rapid deforestation related to soybean and planted pasture expansion. This method can be easily applied to other regions or continents to produce spatially and temporally consistent information on annual LULC.     

HJ-1A/B CCD与Landsat TM/ETM+植被指数的交互比较

利用5对同日过境的HJ-1A/B CCD和Landsat TM/ETM+影像对,研究了二者植被指数(NDVI,SAVI,EVI)之间的定量关系。选用其中的3对影像对作为实验影像,通过对均匀同质实验区对应的植被指数进行回归分析求出二者之间的转换方程,用未参与实验的2对影像对来验证所求转换方程的有效性,并对二者植被指数之间的差异进行了分析。结果表明：两种传感器对应的植被指数之间存在极显著的线性正相关关系,所求的转换方程具有较高的精度,可以利用转换方程将两种传感器的植被指数进行互为转换,有利于二者植被监测结果的互为补充,而两种传感器在光谱响应函数上的不同造成了二者植被指数间存在差异。     

Scaling of land surface temperature using satellite data: A case examination on ASTER and MODIS products over a heterogeneous terrain area

Land surface temperature (LST) is a key parameter in numerous environmental studies. Surface heterogeneity induces uncertainty in pixel-wise LST. Spatial scaling may account for the uncertainty, however, different approaches lead to differences in scaled values. Satellite-retrieved LST may be representative of the pixel-wise LST and useful for scaling analysis, but the limited accuracy of retrieved values adds uncertainty into the scaled values. Based on the Stefan-Boltzmann (S-B) law, this study proposed scaling approaches for LST over flat and relief areas to explore the combined uncertainties in scaling using satellite-retrieved data. To take advantage of simultaneous, multi-resolution observations at coincident nadirs by the Advanced Spaceborne Thermal Emission Reflection Radiometer (ASTER) and the MODerate-resolution Imaging Spectroradiometer (MODIS), LST products from these two sensors were examined for part of the Loess Plateau in China. 90-m ASTER LST data were scaled up to 1 km using the proposed approaches, and variation in the LST was generally reduced after scaling. Amongst the sources of uncertainties, surface heterogeneity (emissivity) and different scaling approaches resulted in very minor differences, with a maximum difference of 0.2 K for the upscaled LST. Terrain features, taken as an areal weighting factor, had negligible effects on the upscaled value. Limited accuracy of the retrieved LST was the major uncertainty. The overall LST increased 0.6 K on average with correction for terrain-induced angular effect and 0.4 K for both angular and adjacency effects over the study area. Accounting for terrain correction in scaling is necessary for rugged areas. With terrain correction, the upscaled ASTER LST achieved an agreement of − 0.1 ± 1.87 K and a root mean square error (RMSE) of 1.87 K overall with the 1-km MODIS LST rectified by Wan et al.'s approach [Wan, Z., Zhang, Y., Zhang Q., Li, Z.-L. (2002b), Validation of the land-surface temperature products retrieved from Terra Moderate Resolution Imaging Spectroradiometer data. Remote Sensing of Environment, 83, 163-180]. Refining the rectification approach resulted in a better agreement of − 0.2 ± 1.57 K and a RMSE of 1.58 K.