This thesis examines the occurrence of hydrologic variability in subarctic ponds within the Churchill region of the Hudson Bay Lowland (HBL) and investigates the utility of using remote sensing studies to characterize changes in pond surface area. The thesis also characterizes hydro-climatic change over the past ~60 years, and compares this to pond sustainability within the region of Churchill. A multiple-methods approach incorporating field research, simple water balance modeling and remote sensing is used to address these objectives. Research findings demonstrate the occurrence of natural fluctuations in pond surface area and water levels in the Canadian subarctic. These fluctuations in pond water levels (and thus surface area) are caused by differences in antecedent hydrologic conditions, which are easily detected using remotely sensed imagery and may produce unrepresentative estimates of pond surface area change. Resulting from a 4.5 - 11.8 cm variation in water depth, pond surface areas were significantly altered by antecedent precipitation (average: 3,711 m²), intra-seasonal variability (average: 2,049 m²) and inter-annual climatic variations (average: 1,977 m²). These noteworthy pond boundary and water level differences reinforce the importance of accounting for hydrologic variability when delineating representative pond coverage and sustainability. Contemporary pond sustainability findings reveal significant regional climatic change, changing pond hydrologic conditions and overall pond physical stability between 1947 and 2008. Specifically, the Churchill region has become warmer and wetter. Occurring at a rate of 1.37 mm/yr over the study period, changing atmospheric conditions caused a decrease in open water pond hydrologic deficits. During the hydrologic recharge period, modeled pond water levels exhibited an increasing trend (August +0.72 mm/yr, September 0.51 mm/yr), which suggests ponds are filling closer to their maximum storage capacity prior to freeze-up. A remote sensing analysis of pond boundary modifications in mid-summer revealed no change in contemporary physical pond sustainability. Detected surface area changes from imagery were mainly attributed to naturally induced hydrologic variability. Overall, this thesis suggests a new methodological approach for conducting remote sensing pond sustainability research within the arctic/subarctic environment. As well, this study determined pond sustainability within the Churchill region over the last ~60 years.

Mixed grasslands in south Canada serve a variety of economic, environmental and ecological purposes. Numerical modeling has become a major method used to identify potential grassland ecosystem responses to environment changes and human activities. In recent years, the focus has been on process models because of their high accuracy and ability to describe the interactions among different environmental components and the ecological processes. At present, two commonly-used process models (CENTURY and BIOME-BGC) have significantly improved our understanding of the possible consequences and responses of terrestrial ecosystems under different environmental conditions. However, problems with these models include only using site-based parameters and adopting different assumptions on interactions between plant, environmental conditions and human activities in simulating such complex phenomenon. In light of this shortfall, the overall objective of this research is to integrate remote sensing products into ecosystem process model in order to simulate productivity for the mixed grassland ecosystem in the landscape level. Data used includes 4-years of field measurements and diverse satellite data (System Pour lObservation de la Terre (SPOT) 4 and 5, Landsat TM and ETM, Advanced Very High Resolution Radiometer (AVHRR) imagery). <p>Using wavelet analyses, the study first detects that the dominant spatial scale is controlled by topography and thus determines that 20-30 m is the optimum resolution to capture the vegetation spatial variation for the study area. Second, the performance of the RDVI (Renormalized Difference Vegetation Index), ATSAVI (Adjusted Transformed Soil-Adjusted Vegetation Index), and MCARI2 (Modified Chlorophyll Absorption Ratio Index 2) are slightly better than the other VIs in the groups of ratio-based, soil-line-related, and chlorophyll-corrected VIs, respectively. By incorporating CAI (Cellulose Absorption Index) as a litter factor in ATSAVI, a new VI is developed (L-ATSAVI) and it improves LAI estimation capability by about 10%. Third, vegetation maps are derived from a SPOT 4 image based on the significant relationship between LAI and ATSAVI to aid spatial modeling. Fourth, object-oriented classifier is determined as the best approach, providing ecosystem models with an accurate land cover map. Fifth, the phenology parameters are identified for the study area using 22-year AVHRR data, providing the input variables for spatial modeling. Finally, the performance of popular ecosystem models in simulating grassland vegetation productivity is evaluated using site-based field data, AVHRR NDVI data, and climate data. A new model frame, which integrates remote sensing data with site-based BIOME-BGC model, is developed for the mixed grassland prairie. The developed remote sensing-based process model is able to simulate ecosystem processes at the landscape level and can simulate productivity distribution with 71% accuracy for 2005.

Wildlife habitat mapping strongly supports applications in natural resource management, environmental conservation, impacts of anthropogenic activity, perturbed ecosystem restoration, species-at-risk recovery and species inventory. Remote sensing has long been identified as a feasible and effective technology for large-area wildlife habitat mapping. However, existing and future uncertainties in remote sensing will definitely have a significant effect on relevant scientific research, such as the limitation of Landsat-series data; the negative impact of cloud and cloud shadows (CCS) in optical imagery; and landscape pattern analysis using remote sensing classification products. This thesis adopted a manuscript-style format; it addresses these challenges (or uncertainties) and opportunities through exploring the state-of-the-art optical and radar remotely sensed data for large-area wildlife habitat mapping, and investigating their feasibility and applicability primarily by comparison either on the level of direct remote sensing products (e.g. classification accuracy) or indirect ecological model (e.g. presence/absence and frequency of use model based on landscape pattern analysis). A framework designed to identify and investigate the potential remotely sensed data, including Disaster Monitoring Constellation (DMC), Landsat Thematic Mapper (TM), Indian Remote Sensing (IRS), and RADARSAT-2, has been developed. The chosen DMC and RADARSAT-2 imagery have acceptable capability of addressing the existing and potential challenges (or uncertainties) in remote sensing of large-area habitat mapping, in order to produce cloud-free thematic maps for the study of wildlife habitat. A quantitative comparison between Landsat-based and IRS-based analyses showed that the characteristics of remote sensing products play an important role in landscape pattern analysis to build grizzly bear presence/absence and frequency of use models.

The primary objective of this research was to evaluate the potential for monitoring forest change using Landsat ETM and Aster data. This was accomplished by performing eight change detection algorithms: pixel post-classification comparison (PCC), image differencing Normalized Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), Transformed Difference Vegetation Index (TDVI), principal component analysis (PCA), multivariate alteration detection (MAD), change vector analysis (CVA) and tasseled cap analysis (TCA). Methods, Post-Classification Comparison and vegetation indices are straightforward techniques and easy to apply. In this study the simplified classification with only 4 forest classes namely close forest, open forest, bare land and grass land was used The overall classification accuracy obtained were 88.4%, 91.9% and 92.1% for the years 2000, 2003 and 2006 respectively. The Tasseled Cap green layer (GTC) composite of the three images was proposed to detect the change in vegetation of the study area. We found that the RBG-TCG worked better than RGBNDVI. For instance, the RBG-TCG detected some areas of changes that RGB-NDVI failed to detect them, moreover RBG-TCG displayed different changed areas with more strong colours. Change vector analysis (CVA) based on Tasseled Cap transformation (TCT) was also applied for detecting and characterizing land cover change. The results support the CVA approach to change detection. The calculated date to date change vectors contained useful information, both in their magnitude and their direction. A powerful tool for time series analysis is the principal components analysis (PCA). This method was tested for change detection in the study area by two ways: Multitemporal PCA and Selective PCA. Both methods found to offer the potential for monitoring forest change detection. A recently proposed approach, the multivariate alteration detection (MAD), in combination with a posterior maximum autocorrelation factor transformation (MAF) was used to demonstrate visualization of vegetation changes in the study area. The MAD transformation provides a way of combining different data types that found to be useful in change detection. Accuracy assessment is an important final step addressed in the study to evaluate the different change detection techniques. A quantitative accuracy assessment at level of change/no change pixels was performed to determine the threshold value with the highest accuracy. Among the various accuracy assessment methods presented the highest accuracy was obtained using the post-classification comparison based on supervised classification of each two time periods (2000 -2003 and 2003-2006), which were 90.6% and 87% consequently.

The potential of atmospheric dust aerosols to modify the Earth's environment and climate has been recognized for some time. However, predicting the diverse impact of dust has several significant challenges. One is to quantify the complex spatial and temporal variability of dust burden in the atmosphere. Another is to quantify the fraction of dust originating from human-made sources. This thesis focuses on the spatiotemporal characterization of sources and dust outbreaks in Central and East Asia by integrating ground-based data, satellite multi-sensor observations, and modeling. A new regional dust modeling system capable of operating over a span of scales was developed. The modeling system consists of a dust module DuMo, which incorporates several dust emission schemes of different complexity, and the PSU/NCAR mesoscale model MM5, which offers a variety of physical parameterizations and flexible nesting capability. The modeling system was used to perform for the first time a comprehensive study of the timing, duration, and intensity of individual dust events in Central and East Asia. Determining the uncertainties caused by the choice of model physics, especially the boundary layer parameterization, and the dust production scheme was the focus of our study. Implications to assessments of the anthropogenic dust fraction in these regions were also addressed. Focusing on Spring 2001, an analysis of routine surface meteorological observations and satellite multi-sensor data was carried out in conjunction with modeling to determine the extent to which to this integrated data set can be used to characterize the spatiotemporal distribution of dust plumes at a range of temporal scales, addressing the active dust sources in China and Mongolia, mid-range transport and trans-Pacific, long-range transport of dust outbreaks on a case-by-case basis. This work demonstrates that adequate and consistent characterization of individual dust events is central to establishing a reliable climatology, ultimately leading to improved assessments of dust impacts on the environment and climate. This will also help to identify the appropriate temporal and spatial scales for adequate intercomparison between model results and observational data as well as for developing an integrated analysis methodology for dust studies.

Environmental satellites represent an economic and easily accessible monitoring means for a plethora of environmental variables, the most important of which is arguably precipitation. While precipitation can also be measured by conventional rain gages and radar, in most world regions, satellites provide the only reliable and sustainable monitoring system. This thesis presents a methodology for estimating precipitation using information from the satellite-borne precipitation radar of the Tropical Rainfall Measurement Mission (TRMM). The methodology combines the precise, but infrequent, TRMM data with the infrared (IR) and visible (VIS) images continuously produced by geostationary satellites to provide precipitation estimates at a variety of temporal and spatial scales. The method is based on detecting IR patterns associated with convective storms and characterizing their evolution phases. Precipitation rates are then estimated for each phase based on IR, VIS, and terrain information. This approach improves the integration of TRMM precipitation rates and IR/VIS data by differentiating major storms from smaller events and noise, and by separating the distinct precipitation regimes associated with each storm phase. Further, the methodology explicitly quantifies the uncertainty of the precipitation estimates by computing their full probability distributions instead of just single optimal values. Temporal and spatial autocorrelation of precipitation are fully accounted for by using spatially optimal estimator methods (kriging), allowing to correctly assess precipitation uncertainty over different spatial and temporal scales. This approach is tested in the Lake Victoria basin over the period 1996-1998 against precipitation data from more than one hundred rain gages representing a variety of precipitation regimes. The precipitation estimates were shown to exhibit much lower bias and better correlation with ground data than commonly used methods. Furthermore, the approach reliably reproduced the variability of precipitation over a range of temporal and spatial scales.

The sustainability of the Texas agricultural and timber sectors is measured using the ratio of human appropriation of net primary productivity (HANPP) to available net primary productivity (NPP) on a county-by-county basis for the entire state. By combining NPP and HANPP, a measure of ecologic sustainability in terms of carbon dynamics is achieved. This is based on a six-year average from 2000 to 2005 obtained from the NASA MODIS sensor, as well as the calculated NPP harvested from agricultural and timber activities reported by USDA Agricultural and Texas Forest Department timber statistics covering the same years. The spatial pattern of NPP in Texas is strongly influenced by moisture availability and is naturally highest in the Gulf Coastal Plains, and parts of east Texas. Areas of artificially-high NPP can often rival or surpass naturally occurring NPP and occur primarily due to irrigation, such as in the Panhandle and lower Rio Grande Valley. Human appropriation of this carbon is greatest in the Panhandle and lower Rio Grande Valley where, in many counties, >45% of all carbon produced is appropriated. HANPP values throughout the rest of the state are moderate (10-24%) corresponding well with global and national HANPP literature. These results support two conflicting findings: increased HANPP indicates decreased ecological sustainability, but is also a measure of increased agricultural efficiency.

Accurate areal measurements of snow-cover extent are important for hydrological and climate modeling. The traditional method of mapping snow cover is binary where a pixel is approximated to either snow-covered or snow-free. Fractional snow cover (FSC) mapping achieves a more precise estimate of areal snow-cover extent by determining the fraction of a pixel that is snow-covered. The two most common FSC methods using Moderate Resolution Imaging Spectroradiometer (MODIS) images are linear spectral unmixing and the empirical Normalized Difference Snow Index (NDSI) method. Machine learning is an alternative to these approaches for estimating FSC, as Artificial Neural Networks (ANNs) have been used for estimating the subpixel abundances of other surfaces. The advantages of ANNs over the other approaches are that they can easily incorporate auxiliary information such as land-cover type and are capable of learning nonlinear relationships between surface reflectance and snow fraction. ANNs are especially applicable to mapping snow-cover extent in forested areas where spatial mixing of surface components is nonlinear. This study developed an ANN approach to snow-fraction mapping. A feed-forward ANN was trained with backpropagation to estimate FSC from MODIS surface reflectance, NDSI, Normalized Difference Vegetation Index (NDVI) and land cover as inputs. The ANN was trained and validated with high spatial-resolution FSC derived from Landsat Enhanced Thematic Mapper Plus (ETM+) binary snow-cover maps. ANN achieved best result in terms of extent of snow-covered area over evergreen forests, where the extent of snow cover was slightly overestimated. Scatter plot graphs of the ANN and reference FSC showed that the neural network tended to underestimate snow fraction in high FSC and overestimate it in low FSC. The developed ANN compared favorably to the standard MODIS FSC product with the two methods estimating the same amount of total snow-covered area in the test scenes.

The Corredor Bioceanico is a major transportation project connecting the agricultural heartlands of South America to the Atlantic and Pacific coasts. The final link is in southeastern Bolivia - an underdeveloped area that is home to two indigenous groups and globally-significant woodlands and wetlands. Infrastructure developments - comprising a major highway upgrade, revitalized railway services and increased flows along gas pipelines to Brazil - pose major threats to livelihoods and the region's ecological integrity. There are two broad objectives: (i) to map and quantify the spatial patterns of land change using a time-series of coarse and medium resolution satellite imagery; and (ii) to understand the socio-economic and political drivers of change by linking household surveys and interviews with farmers; environmental, climatic, and political data; and classified satellite imagery. Overall, large-scale deforestation has occurred along the Corredor Bioceanico for mechanized commercial production of oil-seed crops such as soybeans and sunflower. The significance of these findings is that agriculture-driven deforestation is pushing into sensitive areas threatening world-renowned ecosystems such as the Chaco, Chiquitano and Pantanal as well as noteworthy national parks. Though quantity remains relatively small compared to other parts of South America, rates of forest loss match or exceed those of more publicized regions such as Rondonia or Mato Grosso, Brazil. Moreover, rates of forest loss are accelerating linearly with time due to policies implemented by incumbent president Evo Morales. Results also show that in the first years of cultivation, pasture is the dominant land-use, but it quickly gives way to intensively cropped farmland. The main findings in terms of percentage area cleared according to forest type is that farmers appear to be favoring transitional forest types on deep and poorly drained soils of alluvial plains. Semi-structured interviews with farmers and representatives of key institutions illustrate that price determined by the global market is not proportionally the most dominant motive driving LULCC in the lowlands of Santa Cruz, Bolivia - an area seen as a quintessential neoliberal frontier.

Crown fires, the fastest spreading of all forest fires, can occur in any forest type throughout the United States and the world. The occurrence of crown fires has become increasingly frequent and severe in recent years. The overall aim of this study is to estimate the forest canopy fuel parameters including crown base height (CBH) and crown bulk density (CBD), and to investigate the potential of using airborne lidar data in east Texas. The specific objectives are to: (1) propose allometric estimators of CBD and CBH and compare the results of using those estimators to those produced by the CrownMass/FMAPlus software at tree and stand levels for 50 loblolly pine plots in eastern Texas, (2) develop a methodology for using airborne light detection and ranging (lidar) to estimate CBD and CBH canopy fuel parameters and to simulate fire behavior using estimated forest canopy parameters as FARSITE inputs, and (3) investigate the use of spaceborne ICEsat /GLAS (Ice, Cloud, and Land Elevation Satellite/Geoscience Laser Altimeter System) lidar for estimating canopy fuel parameters. According to our results from the first study, the calculated average CBD values, across all 50 plots, were 0.18 kg/m³ and 0.07 kg/m³, respectively, for the allometric equation proposed herein and the CrownMass program. Lorey’s mean height approach was used in this study to calculate CBH at plot level. The average height values of CBH obtained from Lorey’s height approach was 10.6 m and from the CrownMass program was 9.1 m. The results obtained for the two methods are relatively close to each other; with the estimate of CBH being 1.16 times larger than the CrownMass value. According to the results from the second study, the CBD and CBH were successfully predicted using airborne lidar data with R² values of 0.748 and 0.976, respectively. The third study demonstrated that canopy fuel parameters can be successfully estimated using GLAS waveform data; an R² value of 0.84 was obtained. With these approaches, we are providing practical methods for quantifying these parameters and making them directly available to fire managers. The accuracy of these parameters is very important for realistic predictions of wildfire initiation and growth.