UMD Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/3
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
3 results
Search Results
Item QUANTIFICATION OF ERROR IN AVHRR NDVI DATA(2011) Nagol, Jyoteshwar Reddy; Prince, Stephen D; Vermote, Eric F; Geography; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Several influential Earth system science studies in the last three decades were based on Normalized Difference Vegetation Index (NDVI) data from Advanced Very High Resolution Radiometer (AVHRR) series of instruments. Although AVHRR NDVI data are known to have significant uncertainties resulting from incomplete atmospheric correction, orbital drift, sensor degradation, etc., none of these studies account for them. This is primarily because of unavailability of comprehensive and location-specific quantitative uncertainty estimates. The first part of this dissertation investigated the extent of uncertainty due to inadequate atmospheric correction in the widely used AVHRR NDVI datasets. This was accomplished by comparison with atmospherically corrected AVHRR data at AErosol RObotic NETwork (AERONET) sunphotometer sites in 1999. Of the datasets included in this study, Long Term Data Record (LTDR) was found to have least errors (precision=0.02 to 0.037 for clear and average atmospheric conditions) followed by Pathfinder AVHRR Land (PAL) (precision=0.0606 to 0.0418), and Top of Atmosphere (TOA) (precision=0.0613 to 0.0684). ` Although the use of field data is the most direct type of validation and is used extensively by the remote sensing community, it results in a single uncertainty estimate and does not account for spatial heterogeneity and the impact of spatial and temporal aggregation. These shortcomings were addressed by using Moderate Resolution Imaging Spectrometer (MODIS) data to estimate uncertainty in AVHRR NDVI data. However, before AVHRR data could be compared with MODIS data, the nonstationarity introduced by inter-annual variations in AVHRR NDVI data due to orbital drift had to be removed. This was accomplished by using a Bidirectional Reflectance Distribution Function (BRDF) correction technique originally developed for MODIS data. The results from the evaluation of AVHRR data using MODIS showed that in many regions minimal spatial aggregation will improve the precision of AVHRR NDVI data significantly. However temporal aggregation improved the precision of the data to a limited extent only. The research presented in this dissertation indicated that the NDVI change of ~0.03 to ~0.08 NDVI units in 10 to 20 years, frequently reported in recent literature, can be significant in some cases. However, unless spatially explicit uncertainty metrics are quantified for the specific spatiotemporal aggregation schemes used by these studies, the significance of observed differences between sites and temporal trends in NDVI will remain unknown.Item Improving Predictive Capabilities of Environmental Change with GLOBE Data(2006-07-25) Robin, Jessica; Dubayah, Ralph; Geography; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation addresses two applications of Normalized Difference Vegetation Index (NDVI) essential for predicting environmental changes. The first study focuses on whether NDVI can improve model simulations of evapotranspiration for temperate Northern (> 35) regions. The second study focuses on whether NDVI can detect phenological changes in start of season (SOS) for high Northern (> 60) environments. The overall objectives of this research were to (1) develop a methodology for utilizing GLOBE data in NDVI research; and (2) provide a critical analysis of NDVI as a long-term monitoring tool for environmental change. GLOBE is an international partnership network of K-12 students, teachers, and scientists working together to study and understand the global environment. The first study utilized data collected by one GLOBE school in Greenville, Pennsylvania and the second utilized phenology observations made by GLOBE students in Alaska. Results from the first study showed NDVI could predict transpiration periods for environments like Greenville, Pennsylvania. In phenological terms, these environments have three distinct periods (QI, QII, and QIII). QI reflects onset of the growing season (mid March - mid May) when vegetation is greening up (NDVI < 0.60) and transpiration is less than 2mm/day. QII reflects end of the growing season (mid September - October) when vegetation is greening down and transpiration is decreasing. QIII reflects height of the growing season (mid May - mid September) when transpiration rates average between 2 and 5 mm per day and NDVI is at its maximum (>0.60). Results from the second study showed that a climate threshold of 153 ± 22 growing degree days was a better predictor of SOS for Fairbanks than a NDVI threshold applied to temporal AVHRR and MODIS datasets. Accumulated growing degree days captured the inter-annual variability of SOS better than the NDVI threshold and most closely resembled actual SOS observations made by GLOBE students. Overall, biweekly composites and effects of clouds, snow, and conifers limit the ability of NDVI to monitor phenological changes in Alaska. Both studies did show that GLOBE data provides an important source of input and validation information for NDVI research.Item Effect of Relative Spectral Response on Multi-Spectral Measurements and NDVI from Different Remote Sensing Systems(2006-01-12) Fleming, David J.; Goward, Samuel N; Geography; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Spectrally derived metrics from remotely sensed data measurements have been developed to improve understanding of land cover and its dynamics. Today there are an increasing number of remote sensing systems with varying characteristics that provide a wide range of data that can be synthesized for Earth system science. A more detailed understanding is needed on how to correlate measurements between sensors. One factor that is often overlooked is the effect of a sensor's relative spectral response (RSR) on broadband spectral measurements. This study examined the variability in spectral measurements due to RSR differences between different remote sensing systems and the implications of these variations on the accuracy and consistency of the normalized difference vegetation index (NDVI). A theoretical model study and a sensor simulation study of laboratory and remotely sensed hyper-spectral data of known land cover types was developed to provide insight into the effect on NDVI due to differences in RSR measurements of various land cover signatures. This research has shown that the convolution of RSR, signature reflectance and solar irradiance in land cover measurements leads to complex interactions and generally small differences between sensor measurements. Error associated with cross-senor calibration of signature measurements and the method of band radiance conversion to reflectance also contributed to measurement discrepancies. The effect of measurement discrepancies between sensors on the accuracy and consistency of NDVI measurements of vegetation was found to be dependent on the increasing sensitivity of NDVI to decreasing band measurements. A concept of isolines of NDVI error was developed as a construct for understanding and predicting the effect of differences in band measurements between sensors on NDVI. NDVI difference of less than 0.05 can be expected for many sensor comparisons of vegetation, however, some cases will lead to higher differences. For vegetation signatures used in this study, maximum effect on NDVI from measurement differences was 0.063 with an average of 0.023. For sensors with well aligned RSRs such as Landsat 7 ETM+ and MODIS, NDVI differences in the range of 0.01 are possible.