Conghe Song

Classification and change detection using landsat TM data : When and how to correct atmospheric effects?

Abstract The electromagnetic radiation (EMR) signals collected by satellites in the solar spectrum are modified by scattering and absorption by gases and aerosols while traveling through the atmosphere from the Earths surface to the sensor. When and how to correct the atmospheric effects depend on the remote sensing and atmospheric data available, the information desired, and the analytical methods used to extract the information. In many applications involving classification and change detection, atmospheric correction is unnecessary as long as the training data and the data to be classified are in the same relative scale. In other circumstances, corrections are mandatory to put multitemporal data on the same radiometric scale in order to monitor terrestrial surfaces over time. A multitemporal dataset consisting of seven Landsat 5 Thematic Mapper (TM) images from 1988 to 1996 of the Pearl River Delta, Guangdong Province, China was used to compare seven absolute and one relative atmospheric correction algorithms with uncorrected raw data. Based on classification and change detection results, all corrections improved the data analysis. The best overall results are achieved using a new method which adds the effect of Rayleigh scattering to conventional dark object subtraction. Though this method may not lead to accurate surface reflectance, it best minimizes the difference in reflectances within a land cover class through time as measured with the Jeffries–Matusita distance. Contrary to expectations, the more complicated algorithms do not necessarily lead to improved performance of classification and change detection. Simple dark object subtraction, with or without the Rayleigh atmosphere correction, or relative atmospheric correction are recommended for classification and change detection applications.

Monitoring land-use change in the Pearl River Delta using Landsat TM

The Pearl River Delta in the Peoples Republic of China is experiencing rapid rates of economic growth. Government directives in the late 1970s and early 1980s spurred economic development that has led to widespread land conversion. In this study, we monitor land-use through a nested hierarchy of land-cover. Change vectors of Tasseled Cap brightness, greenness and wetness of Landsat Thematic Mapper (TM) images are combined with the brightness, greenness, wetness values from the initial date of imagery to map four stable classes and five changes classes. Most of the land-use change is conversion from agricultural land to urban areas. Results indicate that urban areas have increased by more than 300% between 1988 and 1996. Field assessments confirm a high overall accuracy of the land-use change map (93.5%) and support the use of change vectors and multidate Landsat TM imagery to monitor land-use change. Results confirm the importance of field-based accuracy assessment to identify problems in a land-use map and to improve area estimates for each class.

Widespread decline of Congo rainforest greenness in the past decade

Tropical forests are global epicentres of biodiversity and important modulators of climate change, and are mainly constrained by rainfall patterns. The severe short-term droughts that occurred recently in Amazonia have drawn attention to the vulnerability of tropical forests to climatic disturbances. The central African rainforests, the second-largest on Earth, have experienced a long-term drying trend whose impacts on vegetation dynamics remain mostly unknown because in situ observations are very limited. The Congolese forest, with its drier conditions and higher percentage of semi-evergreen trees, may be more tolerant to short-term rainfall reduction than are wetter tropical forests, but for a long-term drought there may be critical thresholds of water availability below which higher-biomass, closed-canopy forests transition to more open, lower-biomass forests. Here we present observational evidence for a widespread decline in forest greenness over the past decade based on analyses of satellite data (optical, thermal, microwave and gravity) from several independent sensors over the Congo basin. This decline in vegetation greenness, particularly in the northern Congolese forest, is generally consistent with decreases in rainfall, terrestrial water storage, water content in aboveground woody and leaf biomass, and the canopy backscatter anomaly caused by changes in structure and moisture in upper forest layers. It is also consistent with increases in photosynthetically active radiation and land surface temperature. These multiple lines of evidence indicate that this large-scale vegetation browning, or loss of photosynthetic capacity, may be partially attributable to the long-term drying trend. Our results suggest that a continued gradual decline of photosynthetic capacity and moisture content driven by the persistent drying trend could alter the composition and structure of the Congolese forest to favour the spread of drought-tolerant species.

Monitoring forest succession with multitemporal Landsat images: factors of uncertainty

This study evaluates uncertainty factors in using multitemporal Landsat images for subtle change detection, including atmosphere, topography, phenology, and sun and view angles. The study is based on monitoring forest succession with a set of multiple Landsat Thematic Mapper/Enhanced Thematic Mapper Plus (TM/ETM+) images spanning 15 years over the H.J. Andrews Experimental Forest in the Western Cascades of Oregon. The algorithms for removing atmospheric effects from remotely sensed images evaluated include a new version of the dark object subtraction (DOS3) method, the dense dark vegetation (DDV) method, the path radiance (PARA) approach, and the 6S radiative transfer codes. We found that the DOS3 approach undercorrects the image, and the recently developed DDV and PARA approaches can produce surface reflectance values closely matching those produced by 6S using in situ measurements of atmospheric aerosol optical depth. Atmospheric effects reduce normalized difference vegetation index (NDVI) and greenness, and increase brightness and wetness. Topography modifies brightness and greenness, but has minimal effects on NDVI and wetness, and it interacts with sun angle. Forest stands at late successional stages are more sensitive to topography than younger stands. Though the study areas are covered predominantly by evergreen needleleaf forests, phenological effect is significant. Sun angle effects are confounded with phenology, and reflectance values for stands at different successional stages are related to sun angles nonlinearly. Though Landsat has a small field of view angle, the view angle effects from overlapping Landsat scenes for a mountainous forested landscape may not be ignored when monitoring forest succession with multitemporal images.

Forest mapping with a generalized classifier and Landsat TM data

Monitoring landcover and landcover change at regional and global scales often requires Landsat data to identify and map landscape features and patterns with sufficient detail. Analytical methods based on image-by-image interpretation are too time-consuming and labor-intensive for studies of large areas to be undertaken with any degree of frequency. One potential solution is to develop algorithms or classifiers that can be generalized beyond the arena of the initial training to new images from different spatial, temporal or sensor domains. Building upon earlier success with a generalized classifier to monitor forest change, we now address the question of generalization for classifications of stable landcovers. We evaluate the ability of a supervised neural network, Fuzzy ARTMAP, to identify conifer forest across time and space with Landsat Thematic Mapper (TM) images for a region in northwest Oregon. We also assess the effects of atmospheric corrections on generalized classification accuracies. Using midsummer images atmospherically corrected with a simple dark-object-subtraction (DOS) method, there is no statistically significant loss of accuracy as the classification is extended from the initial training image to other images from the same scene (path and row): temporal generalization is successful. Extending the classifier across space and time to nearby scenes results in a mean decline of 8–13% accuracy depending on the atmospheric correction used. Obvious sources of error, such as seasonality, solar angle variation, and complexity of landcover identification, do not explain the decline in error. Additionally, the patterns in generalization accuracies are complex, and the relationship between pairs of training and testing images is not necessarily reciprocal, i.e., good training data are not necessarily good testing data. Simple DOS atmospheric corrections produce classifications with comparable accuracies as classifications from the more complex radiative transfer corrections. These findings are based on over 200 classifications. A high degree of variability in the classification accuracies underscores the importance of extensive, in-depth analysis of remote sensing techniques and applications, and highlights the potential problem for misleading results based on just a few tests. Generalization is well suited for multitemporal classifications of one Landsat scene. Using simple DOS and midsummer images, generalization offers the opportunity for frequent landcover mapping of a Landsat scene without having to retrain the classifier for each time period of interest. However, at this point, the utility of regional landcover mapping with a generalized classifier remains limited.

A regional forest ecosystem carbon budget model: impacts of forest age structure and landuse history

Abstract This study investigated the impacts of landuse history and forest age structure on regional carbon fluxes for the forests in the Pacific Northwest of the United States based on a two-stage modeling strategy. In the first stage, an individual-based forest ecosystem carbon flux model (IntCarb) at stand scale is developed. IntCarb combines components from the ZELIG and CENTURY models to simulate forest development and heterotrophic respiration, respectively. Stand scale carbon fluxes simulated by IntCarb strongly depend on stand age. A forest stand can be a carbon sink for up to 200 years old with a peak at 30–40 years old. Old-growth stands are carbon neutral to the atmosphere in the long term. For any particular year, an old-growth stand can be either a carbon sink or source. The interannual variation of Net Ecosystem Productivity (NEP) for an old-growth stand is primarily determined by heterotrophic respiration. Due to the high spatial variability of stand ages, forest age structure needs to be taken into account to improve estimation of carbon budgets of forest ecosystems over large areas. In the stand stage, a regional carbon budget model (RegCarb) is developed to estimate regional carbon fluxes over large areas based on forest age structure, adjusting for the nonrespiratory carbon losses (timber harvesting). Our initial estimate with RegCarb for the Pacific Northwest of the United States indicates that this region was a tremendous carbon source to the atmosphere from 1890 to 1990 due to extensive logging of old-growth forest. Projection for the role of forests in this region in global carbon cycle in the future strongly depend on the amount of timber to be harvested, i.e. how the age structure of forests in this region is to be altered.

Vegetation temperature condition index and its application for drought monitoring

An index called vegetation temperature condition index (VTCI) was developed for drought monitoring in this study. The index can be used to monitor drought occurrences at a regional level for a special period (e.g. 10 days) of a year, and can be also used to study the spatial distribution of drought within the region. VTCI is not only related to NDVI changes in the region, but also related to land surface temperature changes of pixels with the same NDVI value. A pilot study was carried out for drought monitoring in the Guanzhong Plain area of the Loess Plateau in the Northwest China. The results showed that VTCI had better performances in classifying the relative drought occurrence levels and in studying the distribution of drought occurrences.

The spatial manifestation of forest succession in optical imagery: The potential of multiresolution imagery

Forest succession is a fundamental ecological phenomenon, which has significant implications for sustainable ecosystem management as well as biological, biophysical, and biogeochemical processes. Remote sensing is perhaps the only viable option for monitoring changes in forest ecosystems over large areas in a timely and cost efficient manner. This study investigates the spatial manifestation of forest succession in optical imagery through three types of models: a two-component spatial model, a canopy reflectance model (Geometric–Optical and Radiative Transfer, GORT) and a forest ecosystem dynamics model (ZELIG). The latter two models provide inputs to the former one to predict the spatial properties of images as a function of the combined effects of tree size and density, the spectral signatures of scene components and pixel size. An important source of information that is diagnostic of canopy structure has been identified: the spatial properties of multiresolution imagery. The sill of variograms of images of forest stands decrease with regularization, and in particular the rate of decrease is related to the size of trees. For stands with larger trees the sills of variograms decrease more slowly with increasing regularization than for stands with smaller trees. However, the spatial patterns for a scene with multiresolution imagery are also dependent on tree cover. This implies that the use of spatial patterns to estimate tree size will require independent estimates of tree cover as a preliminary step. Concept verification with an Ikonos 1-m panchromatic image for stands at the H.J. Andrews Experimental Forest in the Cascade Range of Oregon indicates the simulated spatial patterns exist in multiresolution imagery. This study demonstrates the potential to map tree size automatically from multiresolution imagery.

Spatiotemporal pattern of urbanization in Shanghai, China between 1989 and 2005

Quantifying the spatiotemporal pattern of urbanization is necessary to understand urban morphology and its impacts on biodiversity and ecological processes, and thus can provide essential information for improving landscape and urban planning. Recent studies have suggested that, as cities evolve, certain general patterns emerge along the urban–rural gradient although individual cities always differ in details. To help better understand these generalities and idiosyncrasies in urbanization patterns, we analyzed the spatiotemporal dynamics of the Shanghai metropolitan area from 1989 to 2005, based on landscape metrics and remote sensing data. Specifically, the main objectives of our study were to quantitatively characterize the spatiotemporal patterns of urbanization in Shanghai in recent decades, identify possible spatial signatures of different land use types, and test the diffusion coalescence hypotheses of urban growth. We found that, similar to numerous cities around the world reported in previous studies, urbanization increased the diversity, fragmentation, and configurational complexity of the urban landscape of Shanghai. In the same time, however, the urban–rural patterns of several land use types in Shanghai seem unique—quite different from previously reported patterns. For most land use types, each showed a distinctive spatial pattern along a rural–urban transect, as indicated by landscape metrics. Furthermore, the urban expansion of Shanghai exhibited an outward wave-like pattern. Our results suggest that the urbanization of Shanghai followed a complex diffusion–coalescence pattern along the rural–urban transect and in time.

The spectral/temporal manifestation of forest succession in optical imagery: The potential of multitemporal imagery

Abstract Forest succession is a fundamental ecological process, which has significant implications for sustainable natural resource management as well as ecosystem biological, biophysical, and biogeochemical processes. Remote sensing is perhaps the only viable option to monitor succession in forest ecosystems over large areas in a timely and cost efficient manner. This study integrates the geometric-optical and radiative transfer (GORT) canopy reflectance model with the ZELIG forest ecosystem dynamics model to study the manifestation of forest succession in optical imagery. The spectral/temporal trajectories associated with forest succession are highly nonlinear. The rate of change in spectral space through time is faster during the early years of succession. The succession trajectories do not proceed linearly in the spectral space, but rather curvilinearly, and are strongly influenced by initial background/understory conditions and topography. The nonlinear nature of the trajectories implies that adequate characterization of forest development with remote sensing requires multiple images through time. Validation using multiple Landsat TM images for the H.J. Andrews Experimental Forest (HJA) in the Cascade Range of Oregon found the expected successional trajectories in successfully regenerated young stands. The simulated effects of topography on successional trajectories exist in the Landsat data. The primary challenge in using multitemporal imagery to monitor forest succession is to minimize the effects of various kinds of noise, including varying atmospheric effects, phenology, changes in sun and viewing angles, and topography. The current linkage of GORT–ZELIG only captures the spectral/temporal changes for young stands. Successional changes associated with mature and old-growth stands due to changes in the materials comprising the canopy are not captured. Further research is necessary to include changes associated with mature and old-growth forests, and to develop operational algorithms to utilize information from multitemporal imagery to monitor forest succession.