Landsat Thematic Mapper Imagery Research Articles

Short-term glacier area changes, glacier geometry dependence, and regional climatic variations forcing, King George Island, Antarctica.

This study investigates the transient snowline (TSL) altitude for summer 2020, as well as glacial area loss in King George Island Icefields since 1988 using Sentinel-1 and 2 and Landsat Thematic Mapper (TM) imagery. Trends and anomalies in atmospheric temperature, U-wind, and V-wind were examined using ERA5 solutions. Results show the wet-snow zone corresponds to values of ≤ -13dB, and 44.3% of the glacial area is located above the TSL (≥ 300 m). Glacial area for 2020 is 999.95 km², and losses in the period represent 104.9 km² (error <1%) - a retreat of 3.17 km² / year. Glaciers in Keller Peninsula and Bellingshausen Dome lost the most area (28% and 17%, respectively) and did not have a TSL in 2020; followed by Warszawa (15%), Kraków (13%), and Eastern (10%), where the TSL was verified. Percentage area loss values increased with decreases in dimensions, area above TSL, and maximum elevation. Calving glaciers with ice-flow toward deeper and steeper submarine sectors (Bransfield Strait) exhibited greater glacier variations. The trend in warming atmospheric temperature was greater in the Bransfield Strait than in the Drake Passage. TSL and retreat difference between glaciers were influenced by climatic and ocean input, as well as multiple environmental factors.

INFLUENCES OF URBAN GROWTH ON URBAN HEAT ISLAND IN KUCHING CITY IN YEAR OF 2005 UNTIL 2017

Urbanization has contributed significantly to global warming. Kuching city has grown rapidly in recent years as a result of favourable socioeconomic, political, and physical factors that will undoubtedly affect the urban climate and environment. Landsat Thematic Mapper (ETM+) imagery from 2005, 2010, 2016, and 2017 was used to determine the urban heat island effect by examining the relationship between land-use changes and land surface temperature (LST). The study discovered a significant increase in the built-up area of about 17.1 percent between 2005 and 2017 and a decline in barren land and vegetation areas of about 41.6 percent and 5.1 percent, respectively. Additionally, this study discovered that as land use changed from 2005 to 2017, the LST increased year after year. The findings of this study demonstrate that they are effective tools for urban planners and environmental scientists because they provide critical data for monitoring urban growth patterns and their impact on urban climate.

Assessing Drought Vegetation Dynamics in Semiarid Grass- and Shrubland Using MESMA

Drought intensity and duration are expected to increase over the coming century in the semiarid western United States due to anthropogenic climate change. Historic data indicate that megadroughts in this region have resulted in widespread ecosystem transitions. Landscape-scale monitoring with remote sensing can help land managers to track these changes. However, special considerations are required: traditional vegetation indices such as NDVI often underestimate vegetation cover in semiarid systems due to short and multimodal green pulses, extremely variable rainfall, and high soil fractions. Multi-endmember spectral mixture analysis (MESMA) may be more suitable, as it accounts for both green and non-photosynthetic soil fractions. To determine the suitability of MESMA for assessing drought vegetation dynamics in the western US, we test multiple endmember selection and model parameters for optimizing the classification of fractional cover of green vegetation (GV), non-photosynthetic vegetation (NPV), and soil (S) in semiarid grass- and shrubland in central New Mexico. Field spectra of dominant vegetation species were collected at the Sevilleta National Wildlife Refuge over six field sessions from May–September 2019. Landsat Thematic Mapper imagery from 2009 (two years pre-drought), and Landsat Operational Land Imager imagery from 2014 (final year of drought), and 2019 (five years post-drought) was unmixed. The best fit model had high levels of agreement with reference plots for all three classes, with R2 values of 0.85 (NPV), 0.67 (GV), and 0.74 (S) respectively. Reductions in NPV and increases in GV and S were observed on the landscape after the drought event, that persisted five years after a return to normal rainfall. Results indicate that MESMA can be successfully applied for monitoring changes in relative vegetation fractions in semiarid grass and shrubland systems in New Mexico.

Relación entre la erosión de los barrancos y el modelo de paisaje agrícola en la región de suelo negro típico del noreste de China

The black soil region of Northeast China is an important grain producing area that faces serious erosion problem. We used SPOT5 remote sensing imagery from 2008 to establish the gully distribution in a typical black soil region located in the Wuyuer and Nemoer River basin, and Landsat Thematic Mapper imagery to designate landscape patterns. Ninety-three sub-basins were determined based on digital elevation model data using a hydrological analysis module for which landscape pattern metrics and gully density were obtained using FRAGSTATS software and the Geographic Information System (GIS) spatial analysis function, respectively. The results show a gully density of 4,219.2 m2/km2 with 10,149.5 ha of eroded dry land and 25,261.2 ha of destroyed dry land. Correlation analysis between gully density and the landscape percentage of dry land and forest, perimeter-area fractal dimension, contagion index, and Shannon’s diversity index indicate that regulation of the proportion of forest, dry land, and grass, and reasonable optimization of landscape arrangement is very important for soil erosion management and control. Additionally, the findings indicate that landscape differences between each sub-basin are not the principal reason for differences in gully erosion. Further research on gully formation should integrate the effect of natural forces and human activities.

Examining the Relationship Between Spatial Configurations of Urban Impervious Surfaces and Land Surface Temperature

The urban heat island (UHI) effect has significant effects on the quality of life and public health. Numerous studies have addressed the relationship between UHI and the increase in urban impervious surface area (ISA), but few of them have considered the impact of the spatial configuration of ISA on UHI. Land surface temperature (LST) may be affected not only by urban land cover, but also by neighboring land cover. The aim of this research was to investigate the effects of the abundance and spatial association of ISAs on LST. Taking Harbin City, China as an example, the impact of ISA spatial association on LST measurements was examined. The abundance of ISAs and the LST measurements were derived from Landsat Thematic Mapper (TM) imagery of 2000 and 2010, and the spatial association patterns of ISAs were calculated using the local Moran’s I index. The impacts of ISA abundance and spatial association on LST were examined using correlation analysis. The results suggested that LST has significant positive associations with both ISA abundance and the Moran’s I index of ISAs, indicating that both the abundance and spatial clustering of ISAs contribute to elevated values of LST. It was also found that LST is positively associated with clustering of high-ISA-percentage areas (i.e.,>50%) and negatively associated with clustering of low-ISA-percentage areas (i.e.,<25%). The results suggest that, in addition to the abundance of ISAs, their spatial association has a significant effect on UHIs.

Inversion of Lake Bathymetry through Integrating Multi-Temporal Landsat and ICESat Imagery.

Lake bathymetry provides valuable information for lake basin planning and treatment, lake watershed erosion and siltation management, water resource planning, and environmental protection. Lake bathymetry has been surveyed with sounding techniques, including single-beam and multi-beam sonar sounding, and unmanned ship sounding. Although these techniques have high accuracy, most of them require long survey cycles and entail a high degree of difficulty. On the contrary, optical remote sensing inversion methods are easy to implement, but tend to provide less accurate bathymetry measures, especially when applied to turbid waters. The present study, therefore, aims to improve the accuracy of bathymetry measurements through integrating Landsat Thematic Mapper imagery, the Ice, Cloud, and Land Elevation Satellite’s Geoscience Laser Altimeter System (ICESat/GLAS) data, and water level data measured at hydrological stations. First, the boundaries of a lake at multiple dates were derived using water extraction, initial boundary extraction, and Landsat Thematic Mapper/Enhanced Thematic Mapper (TM/ETM+) strip removal processing techniques. Second, ICESat/GLAS data were introduced to obtain additional topographic information of a lake. The striped topography of a lake’s area was then obtained through eliminating and correcting erroneous points and interpolating the values of unknown points. Third, the entire bathymetry of the lake was obtained through interpolating water level values of lake boundary points in various dates. Experiments show that accurate bathymetry (±1 m) can be successfully derived.

Stand density and carbon storage in cypress-tupelo wetland forests of the Mississippi River delta

Forested wetlands play a vital role in the coastal zone, but their vulnerability to coastal change—and the impact to high-value ecosystem services—is not as well established as that of more seaward systems such as saltmarsh and mangroves. To address this need, we develop field-based stand density classes, then classify baldcypress-water tupelo (Taxodium distichum (L.) Rich var. distichum; Nyssa aquatica L.) stand density on the Mississippi River delta using a multitemporal ordination of reflectance from Landsat Thematic Mapper imagery. Approximately 29, 50, and 21% of forest was classified as full canopy, intermediate, and open canopy, respectively. We estimate stand-level live-stem carbon stocks and annual accumulation rates of 96, 67, and 39 t-C ha−1 and 1.9, 1.4 and 0.8 t-C ha−1 yr−1 for full-canopy, intermediate, and open-canopy forest, respectively. Regional live-stem carbon stocks are ∼ 69 t-C ha-1, and total carbon and annual increment for all forest analyzed are ∼17 Mt-C and ∼0.34 Mt-C yr-1. Much of the cypress-tupelo forest on the delta stores carbon at rates significantly below the potential of fully stocked stands, yet overall rates of accumulation are comparable to other forest types and coastal systems. Delta-wide, there is a well-defined coastward gradient of stand density, which has important implications for the future of deltaic forests and their ecosystem services. The similarities between forest density gradients and well-documented marsh loss patterns in the region suggest that the same processes driving coastal marsh loss—relative sea level rise exacerbated by human activity—are likely responsible for a second, more gradual interface of land loss at the marsh-forest boundary. Results highlight the potential response of coastal forests to continued environmental change—and associated impacts to the carbon cycle—and provide a baseline for detecting future change to forests on the Mississippi River delta.

Modelling LiDAR derived tree canopy height from Landsat TM, ETM+ and OLI satellite imagery—A machine learning approach

Understanding ecological changes in native vegetation communities often requires information over long time periods (multiple decades). Tropical cyclones can have a major impact on woody vegetation structure across northern Australia, however understanding the impacts on woody vegetation structure is limited. Woody vegetation structural attributes such as height are used in ecological studies to identify long term changes and trends. LiDAR has been used to measure woody vegetation structure, however LiDAR datasets cover relatively small areas and historical coverage is restricted, limiting the use of this technology for monitoring long-term change. The Landsat archive spans multiple decades and is suitable for regional/continental assessment. Advances in predictive modelling using machine learning algorithms have enabled complex relationships between dependent and independent variables to be identified. The aim of this study is to develop a predictive model to estimate woody vegetation height from Landsat imagery to assist in understanding change through space and time. A LiDAR canopy height model was produced covering a range of vegetation communities in northern Australia (Darwin region) for use as the dependent variable. A random forest regression model was developed to predict mean LiDAR canopy height (30 m spatial resolution) from Landsat-5 Thematic Mapper (TM). Validation of the random forest model was undertaken on independent data (n = 30,500) resulting in an overall R2 = 0.53, RMSE of 2.8 m. Assessment of the RMSE within four broad vegetation communities ranged from 2.5 to 3.7 m with the two dominant communities in the study area Mangrove forests and Eucalyptus communities recording an RMSE value of 2.9 m and 2.5 m respectively. The model was also applied to Landsat-7 Enhanced Thematic Mapper Plus (ETM+) resulting in an R2 of 0.49, RMSE of 2.8 m. The model was then applied to all cloud free Landsat-5 TM, Landsat-7 ETM+ and Landsat-8 Operational Land Imager (OLI) imagery (106/69 path/row) available between the months April, May and June for 1987 to 2016 to produce annual estimates (29 years) of canopy height. A number of time traces were produced to illustrate tree canopy height through time in the Darwin region which was severely impacted by cyclone (hurricane) Tracy on the 25th December 1974.

Comparative Analysis of Modeling Algorithms for Forest Aboveground Biomass Estimation in a Subtropical Region

Remote sensing–based forest aboveground biomass (AGB) estimation has been extensively explored in the past three decades, but how to effectively combine different sensor data and modeling algorithms is still poorly understood. This research conducted a comparative analysis of different datasets (e.g., Landsat Thematic Mapper (TM), ALOS PALSAR L-band data, and their combinations) and modeling algorithms (e.g., artificial neural network (ANN), support vector regression (SVR), Random Forest (RF), k-nearest neighbor (kNN), and linear regression (LR)) for AGB estimation in a subtropical region under non-stratification and stratification of forest types. The results show the following: (1) Landsat TM imagery provides more accurate AGB estimates (root mean squared error (RMSE) values in 27.7–29.3 Mg/ha) than ALOS PALSAR (RMSE values in 30.3–33.7 Mg/ha). The combination of TM and PALSAR data has similar performance for ANN and SVR, worse performance for RF and KNN, and slightly improved performance for LR. (2) Overestimation for small AGB values and underestimation for large AGB values are major problems when using the optical (e.g., Landsat) or radar (e.g., ALOS PALSAR) data. (3) LR is still an important tool for AGB modeling, especially for the AGB range of 40–120 Mg/ha. Machine learning algorithms have limited effects on improving AGB estimation overall, but ANN can improve AGB modeling when AGB values are greater than 120 Mg/ha. (4) Forest type and AGB range are important factors that influence AGB modeling performance. (5) Stratification based on forest types improved AGB estimation, especially when AGB was greater than 160 Mg/ha, using the LR approach. This research provides new insight for remote sensing-based AGB modeling for the subtropical forest ecosystem through a comprehensive analysis of different source data, modeling algorithms, and forest types. It is critical to develop an optimal AGB modeling procedure, including the collection of a sufficient number of sample plots, extraction of suitable variables and modeling algorithms, and evaluation of the AGB estimates.

Spatiotemporal pattern of urban forest leaf area index in response to rapid urbanization and urban greening

Rapid urbanization and urban greening have caused great changes to urban forests in China. Understanding spatiotemporal patterns of urban forest leaf area index (LAI) under rapid urbanization and urban greening is important for urban forest planning and management. We evaluated the potential for estimating urban forest LAI spatiotemporally by using Landsat TM imagery. We collected three scenes of Landsat TM (thematic mapper) images acquired in 1997, 2004 and 2010 and conducted a field survey to collect urban forest LAI. Finally, spatiotemporal maps of the urban forest LAI were created using a NDVI-based urban forest LAI predictive model. Our results show that normalized differential vegetation index (NDVI) could be used as a predictor for urban forest LAI similar to natural forests. Both rapid urbanization and urban greening contribute to the changing process of urban forest LAI. The urban forest has changed considerably from 1997 to 2010. Urban vegetated pixels decreased gradually from 1997 to 2010 due to intensive urbanization. Leaf area for the study area was 216.4, 145.2 and 173.7 km2 in the years 1997, 2004 and 2010, respectively. Urban forest LAI decreased sharply from 1997 to 2004 and increased slightly from 2004 to 2010 because of numerous greening policies. The urban forest LAI class distributions were skewed toward low values in 1997 and 2004. Moreover, the LAI presented a decreasing trend from suburban to downtown areas. We demonstrate the usefulness of TM remote-sensing in understanding spatiotemporal changing patterns of urban forest LAI under rapid urbanization and urban greening.

An enhanced water index in extracting water bodies from Landsat TM imagery

ABSTRACTThis paper combines the principal component analysis (PCA) with a modified normalized difference water index (MNDWI) to improve the accuracy in extracting water bodies from Landsat Thematic Mapper (TM) imagery. To this end, one Landsat TM image of Taiyuan, China, obtained on 23 September 2011 was used to perform PCA and extract MNDWI. The first two principal components of the PCA and extracted MNDWI were combined to compose a false-colour image, which was then used to develop the enhanced water index (EWI). This method was applied to extract water bodies from Landsat TM imagery for both urban areas and surrounding mountain areas of Taiyuan, which resulted in an overall accuracy of 95%. EWI was also compared with other common water indices in extracting water bodies including general water index, MNDWI and new water index. The results show that the proposed method has the highest accuracy and is recommended to extract water bodies using multispectral satellite imagery for urban areas with mountains surrounded.

An Intensity Gradient/Vegetation Fractional Coverage Approach to Mapping Urban Areas From DMSP/OLS Nighttime Light Data

Many studies have demonstrated the efficient extraction of the spatial extent of urban areas from Defense Meteorological Satellite Program/Operational Linescan System imagery using a fixed thresholding technique. These studies may underestimate and overestimate the extents of small and large cities, respectively. To overcome this problem, a new intensity gradient (IG) and vegetation fractional coverage (VFC) method is developed for identifying cities or towns, principally based on the assumption that there is a border around a city at which the nighttime light intensity decreases sharply. Using this method, the spatial extents of urban areas for two of the biggest countries in the world, namely China and the United States, were extracted in 2010. The urban areas thus identified are compared with the urban areas interpreted from Landsat Thematic Mapper imagery, and the results show that there is a significant linear relationship between the former and latter areas. This demonstrates that the IG/VFC model is effective for efficiently extracting the extent of urban areas from nighttime light imagery.

An analysis of oasis evolution based on land use and land cover change: A case study in the Sangong River Basin on the northern slope of the Tianshan Mountains

This study investigated oasis evolution and the changes of peripheral desert in the Sangong River Basin since the 1950s by rebuilding seven land cover maps derived from black-and-white aerial photographs (1958, 1968, and 1978), a color-infrared aerial photograph (1987), Landsat Thematic Mapper (TM) imagery (1998), Satellite Pour l’Observation de la Terre (SPOT) imagery (2004), and Landsat Operational Land Imager (OLI) imagery (2014). The results showed that: (1) Since 1950, the oasis consecutively expanded more than four times from an alluvial fan to an alluvial plain, causing the shrinkage of desert landscapes that were dominated by a Haloxylon ammodendron Bunge community (HBC) and a Tamarix chinensis Lour community (TLC). Furthermore, the primary (1958–1968) and final (2004–2014) stages were the most important periods, during which agricultural land experienced the most rapid expansion during the period 1958–1968, and the built-up area showed the most rapid expansion after the 2000s. (2) Two basic management modes, a “local mode” formed by the local governments and a “farm management mode” developed by Xinjiang Production and Construction Corps, together promoted oasis evolution under various land-use and landcover (LULC) stages. (3) The evolution of the modern oasis during the 1950s–2004 showed the general features of an arid oasis, while during the period of 2004–2014 it was characterized by a large-scale inter-basin water diversion or the import of new water sources. (4) The oasis expanded at the expense of desert vegetation, resulting in distinct variation in the structure of the desert plant community, which will make it more difficult to protect the desert ecosystem.

Impervious surface extraction in imbalanced datasets: integrating partial results and multi-temporal information in an iterative one-class classifier

ABSTRACTAccurate urban land use/cover monitoring is an essential step towards a sustainable future. As a key part of the classification process, the characteristics of reference data can significantly affect classification accuracy and quality of produced maps. However, ideal reference data is not always readily available; users frequently have difficulty generating sufficient reference data for some classes given time, cost, data availability, expertise level, or other limitations. This study aims at dealing with this lack of sufficiently balanced reference data by presenting a modified hybrid one-class support vector data description (SVDD) model. The underlying hypothesis is that the lack of balanced reference data can be overcome through integration of partially extracted results and multi-temporal spectral information. The partially extracted results, defined as highly accurate classified pixels identified in previous algorithmic iterations, allow a gradual increase of the available training data. Furthermore, the method incorporates a voting system that integrates multi-temporal images using the SVDD algorithm. We applied this hybrid method to binary impervious classification of multi-temporal Landsat Thematic Mapper imagery from Central New York with imbalanced reference data. The proposed hybrid one-class SVDD model achieved a 5–6% improvement in overall accuracy and 0.05–0.09 in kappa than the typical one-class SVDD benchmark. While the method was tested on a single site (albeit with an unusually high reference dataset size of >870,000 pixels) we feel confident to suggest implementation of our methodology in other sites over the traditional method. This is because our approach automatically reverts to the traditional method when voting is inconsistent or there is a limited number of highly accurately classified pixels to assist future iterations. Future work could explore the quantity and temporal specificity (e.g. benefits of specific months) of the multi-temporal image selection and/or test other one-class classifiers.

Land Use Change Impact on Flooding Areas: The Case Study of Cervaro Basin (Italy)

The main goal of this paper is to study the effect of the spatio-temporal changes of Land Use/Land Cover (LULC) within the hydrologic regime of the Cervaro basin in Southern Italy. LANDSAT Thematic Mapper (TM) imagery acquisition dates from 1984, 2003, 2009, and 2011 were selected to produce LULC maps covering a time trend of 28 years. Nine synthetic bands were processed as input data identified as the most effective for the Artificial Neural Network (ANN) classification procedure implemented in this case study. To assess the possible hydrological effects of the detected changes during rainfall events, a physically-based lumped approach for infiltration contribution was adopted within each sub-basin. The results showed an increase in flood peak and a decrease of the rangelands, forests, and bare lands between 1984 and 2011, indicating a good correlation between flooding areas and land use changes, even if it can be considered negligible in basins of large dimensions. These results showed that the impact of land use on the hydrological response is closely related to watershed scale.

Landsat Imagery Spectral Trajectories—Important Variables for Spatially Predicting the Risks of Bark Beetle Disturbance

Tree mortality caused by bark beetle infestation has significant effects on the ecology and value of both natural and commercial forests. Therefore, prediction of bark beetle infestations is critical in forest management. Existing predictive models, however, rarely consider the influence of long-term stressors on forest susceptibility to bark beetle infestation. In this study we introduce pre-disturbance spectral trajectories from Landsat Thematic Mapper (TM) imagery as an indicator of long-term stress into models of bark beetle infestation together with commonly used environmental predictors. Observations for this study come from forests in the central part of the Šumava Mountains, in the border region between the Czech Republic and Germany, Central Europe. The areas of bark beetle-infested forest were delineated from aerial photographs taken in 1991 and in every year from 1994 to 2000. The environmental predictors represent forest stand attributes (e.g., tree density and distance to the infested forest from previous year) and common abiotic factors, such as topography, climate, geology, and soil. Pre-disturbance spectral trajectories were defined by the linear regression slope of Tasseled Cap components (Wetness, Brightness and Greenness) calculated from a time series of 16 Landsat TM images across years from 1984 until one year before the bark beetle infestation. Using logistic regression and multimodel inference, we calculated predictive models separately for each single year from 1994 to 2000 to account for a possible shift in importance of individual predictors during disturbance. Inclusion of two pre-disturbance spectral trajectories (Wetness slope and Brightness slope) significantly improved predictive ability of bark beetle infestation models. Wetness slope had the greatest predictive power, even relative to environmental predictors, and was relatively stable in its power over the years. Brightness slope improved the model only in the middle of the disturbance period (1996). Importantly, these pre-disturbance predictors were not correlated with other predictors, and therefore bring additional explanatory power to the model. Generally, the predictive power of most fitted model decreases as time progresses and models describing the initial phase of bark beetle outbreaks appear more reliable for conducting near-future predictions. The pre-disturbance spectral trajectories are valuable not only for assessing the risk of bark beetle infestation, but also for detection of long-term gradual changes even in non-forest ecosystems.

Influence of Field-Based Species Composition and Understory Descriptions on Spectral Mixture Analysis of Tree Species in the Northwest Territories, Canada

Tree species information is a fundamental component of forest inventories that is challenging to obtain in northern boreal forests because inherently open stands with individual trees might be clumped or widely distributed and contain multiple tree species. These challenges result in a mixed pixel problem that was investigated using Multiple Endmember Spectral Mixture Analysis (MESMA), a technique for identifying the type and proportions of forest components (e.g., sunlit canopy, background vegetation, shadow) in remotely sensed imagery at the subpixel scale. This was tested using Landsat Thematic Mapper (TM) imagery from a study area in the Northwest Territories (NWT), Canada. How the dominant tree species was described in the field, and how the spectral properties of understory vegetation was measured, were both important factors that affected species discrimination. Landsat TM was most sensitive to dominant species defined by the fraction of total basal area of the dominant and codominant trees. Classification accuracies of 90% and 63% for very open and medium density forest stands, respectively, were achieved when understory diversity and abundance information was incorporated into the definition of the background component. It was concluded that this approach warrants further investigation for northern boreal forest species classification.

Changes in irrigated land and agricultural water use in the Lake Urmia basin

ABSTRACTHesami A, Amini A. 2016. Changes in irrigated land and agricultural water use in the Lake Urmia basin. Lake Reserv Manage. 32:288–296.In the Urmia basin, intensive agriculture relies heavily on irrigation systems using river-fed irrigation canals and groundwater-supplied pressure systems, causing Lake Urmia to shrink in recent years. The loss of lake area has diminished the productivity of surrounding agricultural lands and caused environmental out-migration and substantial economic losses. Because water resources in the Lake Urmia basin are limited, the government has prohibited new development plans and determination of new water rights. In this study, remote sensing, geographic information systems, and data gathered from official organizations were used to quantify the change in area under irrigation and water consumption. Multi-date Landsat Thematic Mapper imagery was employed to derive the land use map. The study was conducted in the Miandoab alluvial plain, a 1025 km2 area in the southeast part of the Lake Urmia basin, located in northwest Iran. Crop data gathered from field observations were used to verify the land cover map. To quantify the change in agricultural water use, the net irrigation water requirements for each crop were determined using the CROPWAT model, and the reference evapotranspiration was estimated using the Penman–Monteith equation. The results of this study indicated a 19,363 ha increase in total irrigated area (≈20%) and a 226 million cubic meter increase in water demand (≈35%) for irrigation during the study period. This research highlighted the role of integrated water resources management as a framework to develop the agricultural lands in the basins.

Examining Spectral Reflectance Saturation in Landsat Imagery and Corresponding Solutions to Improve Forest Aboveground Biomass Estimation

The data saturation problem in Landsat imagery is well recognized and is regarded as an important factor resulting in inaccurate forest aboveground biomass (AGB) estimation. However, no study has examined the saturation values for different vegetation types such as coniferous and broadleaf forests. The objective of this study is to estimate the saturation values in Landsat imagery for different vegetation types in a subtropical region and to explore approaches to improving forest AGB estimation. Landsat Thematic Mapper imagery, digital elevation model data, and field measurements in Zhejiang province of Eastern China were used. Correlation analysis and scatterplots were first used to examine specific spectral bands and their relationships with AGB. A spherical model was then used to quantitatively estimate the saturation value of AGB for each vegetation type. A stratification of vegetation types and/or slope aspects was used to determine the potential to improve AGB estimation performance by developing a specific AGB estimation model for each category. Stepwise regression analysis based on Landsat spectral signatures and textures using grey-level co-occurrence matrix (GLCM) was used to develop AGB estimation models for different scenarios: non-stratification, stratification based on either vegetation types, slope aspects, or the combination of vegetation types and slope aspects. The results indicate that pine forest and mixed forest have the highest AGB saturation values (159 and 152 Mg/ha, respectively), Chinese fir and broadleaf forest have lower saturation values (143 and 123 Mg/ha, respectively), and bamboo forest and shrub have the lowest saturation values (75 and 55 Mg/ha, respectively). The stratification based on either vegetation types or slope aspects provided smaller root mean squared errors (RMSEs) than non-stratification. The AGB estimation models based on stratification of both vegetation types and slope aspects provided the most accurate estimation with the smallest RMSE of 24.5 Mg/ha. Relatively low AGB (e.g., less than 40 Mg/ha) sites resulted in overestimation and higher AGB (e.g., greater than 140 Mg/ha) sites resulted in underestimation. The smallest RMSE was obtained when AGB was 80–120 Mg/ha. This research indicates the importance of stratification in mitigating the data saturation problem, thus improving AGB estimation.

Using NDVI to measure precipitation in semi-arid landscapes

Measuring precipitation in semi-arid landscapes is important for understanding the processes related to rainfall and run-off; however, measuring precipitation accurately can often be challenging especially within remote regions where precipitation instruments are scarce. Typically, rain-gauges are sparsely distributed and research comparing rain-gauge and RADAR precipitation estimates reveal that RADAR data are often misleading, especially for monsoon season convective storms. This study investigates an alternative way to map the spatial and temporal variation of precipitation inputs along ephemeral stream channels using Normalized Difference Vegetation Index (NDVI) derived from Landsat Thematic Mapper imagery. NDVI values from 26 years of pre- and post-monsoon season Landsat imagery were derived across Yuma Proving Ground (YPG), a region covering 3,367 km2 of semiarid landscapes in southwestern Arizona, USA. The change in NDVI from a pre-to post-monsoon season image along ephemeral stream channels explained 73% of the variance in annual monsoonal precipitation totals from a nearby rain-gauge. In addition, large seasonal changes in NDVI along channels were useful in determining when and where flow events have occurred.