Integrating Landsat TM and SRTM-DEM derived variables with decision trees for habitat classification and change detection in complex neotropical environments

Remotely sensed images and processing techniques are a primary tool for mapping changes in tropical forest types important to biodiversity and environmental assessment. Detailed land cover data are lacking for most wet tropical areas that present special challenges for data collection. For this study, we utilize decision tree (DT) classifiers to map 32 land cover types of varying ecological and economic importance over an 8000 km² study area and biological corridor in Costa Rica. We assess multivariate QUEST DTs with unbiased classification rules and linear discriminant node models for integrated vegetation mapping and change detection. Predictor variables essential to accurate land cover classification were selected using importance indices statistically derived with classification trees. A set of 35 variables from SRTM-DEM terrain variables, WorldClim grids, and Landsat TM bands were assessed.

Of the techniques examined, QUEST trees were most accurate by integrating a set of 12 spectral and geospatial predictor variables for image subsets with an overall cross-validation accuracy of 93% ± 3.3%. Accuracy with spectral variables alone was low (69% ± 3.3%). A random selection of training and test set pixels for the entire landscape yielded lower classification accuracy (81%) demonstrating a positive effect of image subsets on accuracy. A post-classification change comparison between 1986 and 2001 reveals that two lowland forest types of differing tree species composition are vulnerable to agricultural conversion. Tree plantations and successional vegetation added forest cover over the 15-year time period, but sometimes replaced native forest types, reducing floristic diversity. Decision tree classifiers, capable of combining data from multiple sources, are highly adaptable for mapping and monitoring land cover changes important to biodiversity and other ecosystem services in complex wet tropical environments.     

Land-cover classification in a moist tropical region of Brazil with Landsat Thematic Mapper imagery

This research aims to improve land-cover classification accuracy in a moist tropical region in Brazil by examining the use of different remote-sensing-derived variables and classification algorithms. Different scenarios based on Landsat Thematic Mapper (TM) spectral data and derived vegetation indices and textural images and different classification algorithms, maximum likelihood classification (MLC), artificial neural network (ANN), classification tree analysis (CTA) and object-based classification (OBC), were explored. The results indicate that a combination of vegetation indices as extra bands into Landsat TM multi-spectral bands did not improve the overall classification performance, but the combination of textural images was valuable for improving vegetation classification accuracy. In particular, the combination of both vegetation indices and textural images into TM multi-spectral bands improved the overall classification accuracy (OCA) by 5.6% and the overall kappa coefficient (OKC) by 6.25%. Comparison of the different classification algorithms indicated that CTA and ANN have poor classification performance in this research, but OBC improved primary forest and pasture classification accuracies. This research indicates that use of textural images or use of OBC are especially valuable for improving the vegetation classes such as upland and liana forest classes that have complex stand structures and large patch sizes.     

Land-cover classification in a moist tropical region of Brazil with Landsat TM imagery

This research aims to improve land-cover classification accuracy in a moist tropical region in Brazil by examining the use of different remote sensing-derived variables and classification algorithms. Different scenarios based on Landsat Thematic Mapper (TM) spectral data and derived vegetation indices and textural images, and different classification algorithms - maximum likelihood classification (MLC), artificial neural network (ANN), classification tree analysis (CTA), and object-based classification (OBC), were explored. The results indicated that a combination of vegetation indices as extra bands into Landsat TM multispectral bands did not improve the overall classification performance, but the combination of textural images was valuable for improving vegetation classification accuracy. In particular, the combination of both vegetation indices and textural images into TM multispectral bands improved overall classification accuracy by 5.6% and kappa coefficient by 6.25%. Comparison of the different classification algorithms indicated that CTA and ANN have poor classification performance in this research, but OBC improved primary forest and pasture classification accuracies. This research indicates that use of textural images or use of OBC are especially valuable for improving the vegetation classes such as upland and liana forest classes having complex stand structures and having relatively large patch sizes.     

Application of spectral mixture analysis to Amazonian land-use and land-cover classification

Abundant vegetation species and associated complex forest stand structures in moist tropical regions often create difficulties in accurately classifying land-use and land-cover (LULC) features. This paper examines the value of spectral mixture analysis (SMA) using Landsat Thematic Mapper (TM) data for improving LULC classification accuracy in a moist tropical area in Rondônia, Brazil. Different routines, such as constrained and unconstrained least-squares solutions, different numbers of endmembers, and minimum noise fraction transformation, were examined while implementing the SMA approach. A maximum likelihood classifier was also used to classify fraction images into seven LULC classes: mature forest, intermediate secondary succession, initial secondary succession, pasture, agricultural land, water, and bare land. The results of this study indicate that reducing correlation between image bands and using four endmembers improve classification accuracy. The overall classification accuracy was 86.6% for the seven LULC classes using the best SMA processing routine, which represents very good results for such a complex environment. The overall classification accuracy using a maximum likelihood approach was 81.4%. Another finding is that use of constrained or unconstrained solutions for unmixing the atmospherically corrected or raw Landsat TM images does not have significant influence on LULC classification performances when image endmembers are used in a SMA approach.     

基于ENVISat ASAR影像的湿地植被冠层下淹水范围提取以扎龙保护区为例

以扎龙自然保护区湿地为例,结合ENVISat ASAR多极化(HH/HV)雷达影像与传统的光学影像Landsat TM (band1~5,7),分析雷达影像后向散射系数与Landsat TM影像不同波段反射率在淹水植被、非淹水植被、明水面和裸土不同地表覆被类型的差异。选择训练样本,采用分类回归树(Classification and Regression Tree,CART)模型,分别对两种影像进行分类,可视化表达湿地植被淹水范围空间分布情况。基于实测的植被冠层下淹水范围与非淹水范围样本点对两种数据源的分类结果进行精度验证。结果表明:HH/HV极化影像中,植被覆盖下水体的后向散射系数与其他地表覆被类型有明显区别,分类结果总精度为79.49%,Kappa系数为0.70,湿地植被淹水范围提取精度较高。而TM影像分类结果中,由于部分地区植被覆盖水体,淹水植被分类误差较高。将雷达影像引入沼泽湿地研究,提高了植被淹水范围提取效果,为有效分析湿地生态水文过程提供基础,对湿地水资源合理利用及生物多样性保护具有重要意义。     

Spectral analysis and classification accuracy of coffee crops using Landsat and a topographic‐environmental model

Coffee is an extremely important cash crop, yet previous work indicates that satellite mapping of coffee has produced low classification accuracy. This research examines spectral band combinations and ancillary data for evaluating the classification accuracy and the nature of spectral confusion between coffee and other cover types in a Costa Rican study area. Supervised classification using Landsat Enhanced Thematic Mapper (ETM+) with only red, near‐infrared, and mid‐infrared bands had significantly lower classification accuracy compared to datasets that included more spectral bands and ancillary data. The highest overall accuracy achieved was 65%, including a coffee environmental stratification model (CESM). Producer's and user's accuracy was highest for shade coffee plantations (91.8 and 61.1%) and sun coffee (86.2 and 68.4%) with band combination ETM+ 34567, NDVI, cos (i), and including the use of the CESM. Post‐classification stratification of the optimal coffee growing zone based on elevation and precipitation data did not show significant improvement in land cover classification accuracy when band combinations included both the thermal band and NDVI. A forward stepwise discriminant analysis indicated that ETM+ 5 (mid‐infrared band) had the highest discriminatory power. The best discriminatory subset for all woody cover types including coffee excluded ETM+ 3 and 7; however, the land cover accuracy assessment indicated that overall accuracy, as well as producer's and user's accuracy of shade and sun coffee, were slightly improved with the inclusion of these bands. Although spectral separation between coffee crops and woodland areas was only moderately successful in the Costa Rica study, the overall accuracy, as well as the sun and shade coffee producer's and user's accuracy, were higher than reported in previous research.     

Coastal wetland vegetation classification with a Landsat Thematic Mapper image

Coastal wetland vegetation classification with remotely sensed data has attracted increased attention but remains a challenge. This paper explored a hybrid approach on a Landsat Thematic Mapper (TM) image for classifying coastal wetland vegetation classes. Linear spectral mixture analysis was used to unmix the TM image into four fraction images, which were used for classifying major land covers with a thresholding technique. The spectral signatures of each land cover were extracted separately and then classified into clusters with the unsupervised classification method. Expert rules were finally used to modify the classified image. This research indicates that the hybrid approach employing sub-pixel information, an analyst's knowledge and characteristics of coastal wetland vegetation distribution shows promise in successfully distinguishing coastal vegetation classes, which are difficult to separate with a maximum likelihood classifier (MLC). The hybrid method provides significantly better classification results than MLC.     

Improvement of tropical vegetation mapping using a remote sensing technique: A case of Khao Yai National Park,Thailand

The forest ecosystems of Thailand are characterized by a diverse and complex vegetation structure. Classification of vegetation types of such forest ecosystems has been experienced as a difficult task, even with large-scale aerial photography. Satellite remote sensing, the digital technique in particular, has not been widely used for vegetation mapping in Thailand until now. The objective of this study was to explore the potential of digital image processing over the existing technique of visual interpretation of Landsat Thematic Mapper (TM) false colour composite (BGR-2, 3, 4) to produce forest cover maps in Thailand. Supervised and unsupervised classification methods were employed with different band combinations to discriminate vegetation types in the Khao Yai National Park using Landsat TM data. The results indicated that thematic classes derived from supervised classification produced higher overall accuracy than unsupervised classification. In addition, the combination of ratio bands R4/3, R5/2, R5/4 and R5/7 ranked the highest in terms of accuracy (65% for unsupervised and 79% for supervised) and the combination of bands 2, 3 and 4 gave the lowest (56% for both methods). Finally, it was concluded that, even within the limit of spectral information available in the image, the digital classification can improve the result of visual interpretation.     

A non-parametric,supervised classification of vegetation types on the Kaibab National Forest using decision trees

Traditional land classification techniques for large areas that use Landsat Thematic Mapper (TM) imagery are typically limited to the fixed spatial resolution of the sensors (30 m). However, the study of some ecological processes requires land cover classifications at finer spatial resolutions. We model forest vegetation types on the Kaibab National Forest (KNF) in northern Arizona to a 10-m spatial resolution with field data, using topographical information and Landsat TM imagery as auxiliary variables. Vegetation types were identified by clustering the field variables total basal area and proportion of basal area by species, and then using a decision tree based on auxiliary variables to predict vegetation types. Vegetation types modelled included pinyon-juniper, ponderosa pine, mixed conifer, spruce- and deciduous-dominated mixes, and openings. To independently assess the accuracy of the final vegetation maps using reference data from different sources, we used a post-stratified, multivariate composite estimator. Overall accuracy was 74.5% (Kappa statistic = 49.9%). Sources of error included differentiating between mixed conifer and spruce-dominated types and between openings in the forest and deciduous-dominated mixes. Overall, our non-parametric classification method successfully identified dominant vegetation types on the study area at a finer spatial resolution than can typically be achieved using traditional classification techniques.     

基于时序Landsat数据的三江平原植被地表类型变化遥感探测研究

Landsat 卫星遥感数据具有分辨率较高,数据积累时间长的特点,在探测地表覆盖变化和地物分类中得到广泛应用。首先,对获取的Landsat TM/ETM+时间序列数据进行了定量化处理,获取了三江平原七台河市1989~2012年时间序列Landsat地表反射率图像。其次,设计了林地指数和湿地指数,提取了三江平原七台河区域地物光谱和时序特征,同时设计构建了地表覆盖分类和植被地表类型变化探测的决策树算法,实现了1989~2012年七台河区域的植被地表覆盖变化的动态监测,提取了森林覆盖变化的空间分布与变化时间。最后,对七台河区域地表覆盖与植被地表类型变化进行了精度检验,分类总体精度达到90.04%,Kappa系数达0.88。研究结果表明:基于定量化的Landsat时间序列数据的分类算法能克服单时相影像分类的缺陷,实现区域地物自动分类和地表覆盖变化的动态监测。
     

Analysis of different polarimetric target decomposition methods in forest density classification using C band SAR data

The ability of synthetic aperture radar (SAR) C-band microwave energy to penetrate within forest vegetation makes it possible to extract information on crown components, which in turn gives a better approximation of relative canopy density than optical data-derived canopy density. Many studies have been reported to estimate forest biomass from SAR data, but the scope of C-band SAR in characterizing forest canopy density has not been adequately understood with polarimetric techniques. Polarimetric classification is one of the most significant applications of polarimetric SAR in remote sensing. The objective of the present study was to evaluate the feasibility of different polarimetric SAR data decomposition methods in forest canopy density classification using C-band SAR data. Landsat (Land Satellite) 5 TM (Thematic Mapper) data of the same area has been used as optical data to compare the classification result. RADARSAT (Radar Satellite)-2 image with fine quad-pol obtained on 27 October 2011 over tropical dry forests of Madhav National Park, India, was used for the analysis of full polarimetric data. Six decomposition methods were selected based on incoherent decomposition for generating input images for classification, i.e. Huynen, Freeman and Durden, Yamaguchi, Cloude, Van zyl, and H/A/α. The performance of each decomposition output in relation to each land cover unit present in the study area was assessed using a support vector machine (SVM) classifier. Results show that Yamaguchi 4-component decomposition (overall accuracy 87.66% and kappa coefficient (κ) 0.86) gives better classification results, followed by Van Zyl decomposition (overall accuracy 87.20% and κ 0.85) and Freeman and Durden (overall accuracy 86.79% and κ 0.85) in forest canopy density classification. Both model-based decompositions (Freeman and Durden and Yamaguchi4) registered good classification accuracy. In eigenvector or eigenvalue decompositions, Van zyl registered the second highest accuracy among different decompositions. The experimental results obtained with polarimetric C-band SAR data over a tropical dry deciduous forest area imply that SAR data have significant potential for estimating canopy density in operational forestry. A better forest density classification result can be achieved within the forest mask (without other land cover classes). The limitations associated with optical data such as non-availability of cloud-free data and misclassification because of gregarious occurrence of bushy vegetation such as Lantana can be overcome by using C-band SAR data.     

TM digital processing of a tropical forest region in southeastern Mexico

A humid forest in the neotropical area of Los Tuxtlas, in southeastern Mexico has been used as a test area (900km²) for classification of landscape and vegetation by means of Landsat Thematic Mapper (TM) data, aerial photography and 103 ground samples. The area presents altitudinal variations from sea level to 1640m, providing a wide variety of vegetation types. A hybrid (supervised/unsupervised) classification approach was used, defining spectral signatures for 14 clustering areas with data from the reflective bands of the TM. The selected clustering areas ranged from vegetation of the highlands and the rain forest to grassland, barren soil, crops and secondary vegetation. The digital classification compared favourably with results from aerial photography and with those from a multivariate analysis of the 103 ground data. The statistical evaluation (error matrix) of the classified image indicated an overall 84·4 per cent accuracy with a kappa coefficient of agreement of 0·83. A geographical information system (GIS) was used to compile a land unit and a vegetation map. The TM data allowed for delineation of boundaries in the land unit map, and for a finer differentiation of vegetation types than those identified during field work. Digital value patterns of several information classes are shown and discussed as an indirect guide of the spectral behaviour of vegetation of highlands, rain forest, secondary vegetation and crops. The method is considered applicable to the inventory of other forested areas, especially those with significant variations in vegetation.     

Mapping montane tropical forest successional stage and land use with multi-date Landsat imagery

When mapping land cover with satellite imagery in montane tropical regions, varying illumination angles and ecological zones can obscure the differences between spectral responses of old-growth forest, secondary forest and agricultural lands. We used multi-date, Landsat Thematic Mapper (TM) imagery to map secondary forests, agricultural lands and old-growth forests in the Talamanca Mountain Range in southern Costa Rica. With stratification by illumination and ecological zone, the overall accuracy for this classification was 87% with a Kappa coefficient of 0.83. We also examined spectral responses to forest successional stage, ecological zone and aspect illumination for the TM Tasselled Cap indices, TM (2 x 6)/7, TM 4/5 and TM difference bands, and whether using digital data from multiple decades improved classification accuracy. Digital maps of ecological zones should be useful for large-scale mapping of land use and forest successional stage in complex montane regions such as those in Central America.     

Land cover mapping of large areas using chain classification of neighboring Landsat satellite images

Satellite imagery is the major data source for regional to global land cover maps. However, land cover mapping of large areas with medium-resolution imagery is costly and often constrained by the lack of good training and validation data. Our goal was to overcome these limitations, and to test chain classifications, i.e., the classification of Landsat images based on the information in the overlapping areas of neighboring scenes. The basic idea was to classify one Landsat scene first where good ground truth data is available, and then to classify the neighboring Landsat scene using the land cover classification of the first scene in the overlap area as training data. We tested chain classification for a forest/non-forest classification in the Carpathian Mountains on one horizontal chain of six Landsat scenes, and two vertical chains of two Landsat scenes each. We collected extensive training data from Quickbird imagery for classifying radiometrically uncorrected data with Support Vector Machines (SVMs). The SVMs classified 8 scenes with overall accuracies between 92.1% and 98.9% (average of 96.3%). Accuracy loss when automatically classifying neighboring scenes with chain classification was 1.9% on average. Even a chain of six images resulted only in an accuracy loss of 5.1% for the last image compared to a reference classification from independent training data for the last image. Chain classification thus performed well, but we note that chain classification can only be applied when land cover classes are well represented in the overlap area of neighboring Landsat scenes. As long as this constraint is met though, chain classification is a powerful approach for large area land cover classifications, especially in areas of varying training data availability.     

Detection of land use/land cover changes for the northern part of the Nile delta (Burullus region), Egypt

The monitoring of land use/land cover changes along the north part of the Nile delta, Egypt is very important for the planner, management, governmental and non-governmental organizations and the scientific community. This information is essential for planning and implementing policies to optimize the use of natural resources and accommodate development whilst minimizing the impact on the environment. To study these changes along the study area, two sets of Landsat Thematic Mapper (TM) data were used. TM data are useful for this type of study due to its high spatial resolution, spectral resolution and low repetitive acquisition (16 days). A post-classification technique is used in this study based on hybrid classification (unsupervised and supervised). Each method used was assessed, and checked in field. Nine land use/land cover classes are produced. The overall accuracy for a 1984 image is 78% and for a 1997 image is 80%. The objective of this study was to detect land use/land cover changes, and to assess the nature of future change following construction of the international coastal road which crosses the study area.     

Regional-scale boreal forest cover and change mapping using Landsat data composites for European Russia

Boreal forests are a critical component of the global carbon cycle, and timely monitoring allows for assessing forest cover change and its impacts on carbon dynamics. Earth observation data sets are an important source of information that allow for systematic monitoring of the entire biome. Landsat imagery, provided free of charge by the USGS Center for Earth Resources Observation and Science (EROS) enable consistent and timely forest cover updates. However, irregular image acquisition within parts of the boreal biome coupled with an absence of atmospherically corrected data hamper regional-scale monitoring efforts using Landsat imagery. A method of boreal forest cover and change mapping using Landsat imagery has been developed and tested within European Russia between circa year 2000 and 2005. The approach employs a multi-year compositing methodology adapted for incomplete annual data availability, within-region variation in growing season length and frequent cloud cover. Relative radiometric normalization and cloud/shadow data screening algorithms were employed to create seamless image composites with remaining cloud/shadow contamination of less than 0.5% of the total composite area. Supervised classification tree algorithms were applied to the time-sequential image composites to characterize forest cover and gross forest loss over the study period. Forest cover results when compared to independently-derived samples of Landsat data have high agreement (overall accuracy of 89%, Kappa of 0.78), and conform with official forest cover statistics of the Russian government. Gross forest cover loss regional-scale mapping results are comparable with individual Landsat image pair change detection (overall accuracy of 98%, Kappa of 0.71). The gross forest cover loss within European Russia 2000-2005 is estimated to be 2210 thousand hectares, and constitutes a 1.5% reduction of year 2000 forest cover. At the regional scale, the highest proportional forest cover loss is estimated for the most populated regions (Leningradskaya and Moskovskaya Oblast). Our results highlight the forest cover depletion around large industrial cities as the hotspot of forest cover change in European Russia.     

Land cover discrimination potential of radar multitemporal series and optical multispectral images in a Mediterranean cultural landscape

The aim of the present study is (1) to evaluate the performances of two series of European Remote Sensing (ERS) Synthetic Aperture Radar (SAR) images for land cover classification of a Mediterranean landscape (Minorca, Spain), compared with multispectral information from Système Pour l'Observation de la Terre (SPOT) and Landsat Thematic Mapper (TM) sensors, and (2) to test the synergy of SAR and optical data with a fusion method based on the Demspter–Shafer evidence theory, which is designed to deal with imprecise information. We have evaluated as a first step the contribution of multitemporal ERS data and contextual methods of classification, with and without filtering, for the discrimination of vegetation types. The present study shows the importance of time series of the ERS sensor and of the vectorial MMSE (minimum mean square error) filter based on segmentation for land cover classification. Fifteen land cover classes were discriminated (eight concerning different vegetation types) with a mean producer's accuracy of 0.81 for a five-date time series within 1998, and of 0.71 for another four-date time series for 1994/1995. These results are comparable to those from SPOT XS images: 0.69 for July, 0.67 for October (0.85 for July plus October), and also from TM data (0.81). These results are corroborated by the kappa coefficient of agreement. The fusion between the 1994 series of ERS and XS (July), based on a derived method of the Dempster–Shafer evidence theory, shows a slight improvement on overall accuracies: +0.06 of mean producer's accuracy and +0.04 of kappa coefficient.     

Mapping the invasive species,Chinese tallow,with EO1 satellite Hyperion hyperspectral image data and relating tallow occurrences to a classified Landsat Thematic Mapper land cover map

Our objective was to provide a realistic and accurate representation of the spatial distribution of Chinese tallow (Triadica sebifera) in the Earth Observing 1 (EO1) Hyperion hyperspectral image coverage by using methods designed and tested in previous studies. We transformed, corrected, and normalized Hyperion reflectance image data into composition images with a subpixel extraction model. Composition images were related to green vegetation, senescent foliage and senescing cypress‐tupelo forest, senescing Chinese tallow with red leaves (‘red tallow’), and a composition image that only corresponded slightly to yellowing vegetation. These statistical and visual comparisons confirmed a successful portrayal of landscape features at the time of the Hyperion image collection. These landscape features were amalgamated in the Landsat Thematic Mapper (TM) pixel, thereby preventing the detection of Chinese tallow occurrences in the Landsat TM classification. With the occurrence in percentage of red tallow (as a surrogate for Chinese tallow) per pixel mapped, we were able to link dominant land covers generated with Landsat TM image data to Chinese tallow occurrences as a first step toward determining the sensitivity and susceptibility of various land covers to tallow establishment. Results suggested that the highest occurrences and widest distribution of red tallow were (1) apparent in disturbed or more open canopy woody wetland deciduous forests (including cypress‐tupelo forests), upland woody land evergreen forests (dominantly pines and seedling plantations), and upland woody land deciduous and mixed forests; (2) scattered throughout the fallow fields or located along fence rows separating active and non‐active cultivated and grazing fields, (3) found along levees lining the ubiquitous canals within the marsh and on the cheniers near the coastline; and (4) present within the coastal marsh located on the numerous topographic highs.     

Mapping land use/cover distribution on a mountainous tropical island using remote sensing and GIS

The land use/cover distribution on Langkawi Island, Malaysia was mapped using remote sensing and a Geographic Information System (GIS). A Landsat Thematic Mapper (TM) satellite image taken in March 1995 was processed, geocorrected and analysed using IDRISI, raster-based GIS software. An unsupervised classification was performed based on spectral data from a composite image of the bands TM3, TM4 and TM5. Using this output, field data together with available secondary data consisting of topography, land use and soil maps were used to perform a maximum likelihood supervised classification. The overall accuracy of the output image was 90% and individual class accuracy ranged from 74% for rubber to 100% for paddy fields. The classified areas on the image were mainly confined to the mountainous and hilly regions on the island. A shaded relief map, simulating sunshine conditions, showed that the unclassified areas are located in the shadowed slopes, i.e. the slopes facing west. Consequently, the imagery was subdivided on the basis of slope aspect and a stratified classification was performed. As a result of this procedure, the overall accuracy increased to 92% and the individual class accuracy for the inland forest class increased by 9% to 90% . Using IDRISI, individual class areas as well as percentages were calculated. The kappa coefficient for the classified image was 0.90. Qualitative analysis indicates that topography is the main control on the spatial distribution of land use/cover types on the island. As Langkawi Island has been developing rapidly over the last decade, successful planning will require reliable information about land use/cover distribution and change. This study illustrates that remote sensing and GIS techniques are capable of providing such information.