NDVI time series for monitoring RUSLE cover management factor in a tropical watershed

Land cover, an important factor for monitoring changes in land use and erosion risk, has been widely monitored and evaluated by vegetation indices. However, a study that associates normalized difference vegetation index (NDVI) time series to climate parameters to determine soil cover has yet to be conducted in the Atlantic Rainforest of Brazil, where anthropogenic activities have been carried out for centuries. The objective of this paper is to evaluate soil cover in a Brazilian Atlantic rainforest watershed using NDVI time series from Thematic Mapper (TM) Landsat 5 imagery from 1986 to 2009, and to introduce a new method for calculating the cover management factor (C-factor) of the Revised Universal Soil Loss Equation (RUSLE) model. Twenty-two TM Landsat 5 images were corrected for atmospheric effects using the 6S model, georeferenced using control points collected in the field and imported to a GIS database. Contour lines and elevation points were extracted from a 1:50,000-scale topographic map and used to construct a digital elevation model that defined watershed boundaries. NDVI and RUSLE C-factor values derived from this model were calculated within watershed limits with 1 km buffers. Rainfall data from a local weather station were used to verify NDVI and C-factor patterns in response to seasonal rainfall variations. Our proposed method produced realistic values for RUSLE C-factor using rescaled NDVIs, which highly correlated with other methods, and were applicable to tropical areas exhibiting high rainfall intensity. C-factor values were used to classify soil cover into different classes, which varied throughout the time-series period, and indicated that values attributed to each land cover cannot be fixed. Depending on seasonal rainfall distribution, low precipitation rates in the rainy season significantly affect the C-factor in the following year. In conclusion, NDVI time series obtained from satellite images, such as from Landsat 5, are useful for estimating the cover management factor and monitoring watershed erosion. These estimates may replace table values developed for specific land covers, thereby avoiding the cumbersome field measurements of these factors. The method proposed is recommended for estimating the RUSLE C-factor in tropical areas with high rainfall intensity.     

Remote sensing of selective logging in Amazonia: Assessing limitations based on detailed field observations, Landsat ETM+, and textural analysis

We combined a detailed field study of forest canopy damage with calibrated Landsat 7 Enhanced Thematic Mapper Plus (ETM+) reflectance data and texture analysis to assess the sensitivity of basic broadband optical remote sensing to selective logging in Amazonia. Our field study encompassed measurements of ground damage and canopy gap fractions along a chronosequence of postharvest regrowth of 0.5–3.5 years. We found that canopy damage and regrowth rates varied according to the logging method used, either conventional logging or reduced impact logging. Areas used to stage felled trees prior to transport, log decks, had the largest gap fractions immediately following cutting. Log decks were quickly colonized by early successional plant species, resulting in significant gap fraction decreases within 1.5 years after site abandonment. Although log decks were the most obvious damage areas on the ground and in satellite imagery, they accounted for only 1–2% of the total harvested area of the blocks studied. Other forest damage features such as tree-fall gaps, skid trails, and roads were difficult to recognize in Landsat reflectance data or through textural analysis. These landscape features could be only crudely resolved in the most intensively logged forests and within about 0.5 years following harvest. We found that forest damage within any of the landscape strata (decks, roads, skids, tree falls) could not be resolved with Landsat reflectance or texture data when the canopy gap fraction was <50%. The basic Landsat ETM+ imagery lacks the resolution of forest structural features required for quantitative studies of logging damage. Landsat textural analyses may be useful for broad delineation of logged forests, but detailed ecological and biogeochemical studies will probably need to rely on other remote sensing approaches. Until spatial gradients of canopy damage and regrowth resulting from selective logging operations in tropical forests in the Amazon region are resolved, the impacts of this land use on a continental scale will remain poorly understood.     

Mapping burned areas from Landsat TM/ETM+ data with a two-phase algorithm: Balancing omission and commission errors

Maps of burned area have been obtained from an automatic algorithm applied to a multitemporal series of Landsat TM/ETM+ images in two Mediterranean sites. The proposed algorithm is based on two phases: the first one intends to detect the more severely burned areas and minimize commission errors. The second phase improves burned patches delimitation using a hybrid contextual algorithm based on logistic regression analysis, and tries to minimize omission errors. The algorithm was calibrated using six study sites and it was validated for the whole territory of Portugal (89,000 km²) and for Southern California (70,000 km²). In the validation exercise, 65 TM/ETM+ scenes for Portugal and 35 for California were used, all from the 2003 fire season. A good agreement with the official burned area perimeters was shown, with kappa values close to 0.85 and low omission and commission errors (< 16.5%). The proposed algorithm could be operationally used for historical mapping of burned areas from Landsat images, as well as from future medium resolution sensors, providing they acquire images in two bands of the Short Wave Infrared (1.5-2.2 μm).     

A refined classification approach by integrating Landsat Operational Land Imager (OLI) and RADARSAT-2 imagery for land-use and land-cover mapping in a tropical area

Producing accurate land-use and land-cover (LULC) mapping is a long-standing challenge using solely optical remote-sensing data, especially in tropical regions due to the presence of clouds. To supplement this, RADARSAT images can be useful in assisting LULC mapping. The fusion of optical and active remote-sensing data is important for accurate LULC mapping because the data from different parts of the spectrum provide complementary information and often lead to increased classification accuracy. Also, the timeliness of using synthetic aperture radar (SAR) fills information gaps during overcast or hazy periods. Therefore, this research designed a refined classification procedure for LULC mapping for tropical regions. Determining the best method for mapping with a specific data source and study area is a major challenge because of the wide range of classification algorithms and methodologies available. In this study, different combinations and the potential of Landsat Operational Land Imager (OLI) and RADARSAT-2 SAR data were evaluated to select the best procedure for LULC classification. Results showed that the best filter for SAR speckle reduction is the 5 × 5 enhanced Lee. Furthermore, image-sharpening algorithms were employed to fuse Landsat multispectral and panchromatic bands and subsequently these algorithms were analysed in detail. The findings also confirmed that Gram–Schmidt (GS) performed better than the other techniques employed. Fused Landsat data and SAR images were then integrated to produce the LULC map. Different classification algorithms were adopted to classify the integrated Landsat and SAR data, and the maximum likelihood classifier (MLC) was considered the best approach. Finally, a suitable classification procedure was designed and proposed for LULC as mapping in tropical regions based on the results obtained. An overall accuracy of 98.62% was achieved from the proposed methodology. The proposed methodology is a useful tool in industry for mapping purposes. Additionally, it is also useful for researchers, who could extend the method for different data sources and regions.     

Modelling and mapping potential hooded warbler (Wilsonia citrina) habitat using remotely sensed imagery

Modelling and mapping of hooded warbler (Wilsonia citrina) nesting habitat in forests of southern Ontario were conducted using Ikonos and Landsat data. The study began with an analysis of skyward hemispherical photography to determine canopy characteristics associated with nest sites. It showed that nest sites had significantly less overhead canopy cover and larger maximum gap size than in non-nest areas. These findings led to the hypothesis that brightness variability in high to moderate resolution remotely sensed imagery may be greater at nest sites than in non-nest areas due to larger shadows and greater shadow variability related to these gap characteristics. This was confirmed when, in addition to some spectral band brightness variables, several image texture and spectrally unmixed fraction (shadow, bare soil) variables were found to be significantly different for nest and non-nest sites in Ikonos and Landsat imagery. These significantly different variables were used in maximum likelihood classification (MLC) and logistic regression (LR) to produce maps of potential nesting habitat. Mapping was conducted with Ikonos and Landsat in a local area where most known nest sites occur, and regionally using Landsat data for almost all of the hooded warbler range in southern Ontario. For the local area mapping using Ikonos data, a posteriori probabilities for both the MLC and LR methods showed that about 62% of the nest sites set aside for validation had been classified with high probability (p > 0.70) in the nest class. MLC mapping accuracy was 70% for the validation nest sites and 87% of validation nest sites were within 10 m of classified nesting habitat, a distance approximately equivalent to expected positional error in the data. LR accuracy was slightly lower. Nest site MLC mapping accuracy in the local area using Landsat data was 87% but the map was much coarser due to the larger pixel size. Regional mapping with Landsat imagery produced lower classification accuracy due to high errors of commission for the habitat class. This resulted from a poor spatial distribution and low number of observations of nest sites throughout the region compared to the local area, while the non-nest site data distribution was too narrow, having been defined and assessed (using standard accepted methods) as areas with no ground shrubs. If either of these problems can be ameliorated, regional mapping accuracy may improve.     

Forest regeneration on abandoned clearances in central Amazonia

A time-series of Landsat and SPOT sensor data was used to approximate the age of tropical forests regenerating on abandoned agricultural land north of Manaus, Brazil, and to estimate the period of active land use prior to abandonment. Based on field data, two distinct regeneration pathways, dominated by the pioneer genera Cecropia and Vismia, respectively, were described, with the former regenerating on the least intensively used sites. Forests of mixed species composition and lacking numerical dominance by pioneer species were also observed. Transformed Divergence Analysis of Landsat Thematic Mapper (TM) radiance data revealed that forests of varying age and following different regeneration pathways were best discriminated using mid infrared (1.55-1.74 w m) wavelengths. As rates of carbon sequestration by forests vary with age and regeneration pathway, the potential exists for refining spatial estimates of the carbon balance of tropical forests regenerating on abandoned agricultural lands.     

Shuttle radar mapping with diverse incidence angles in the rainforest of Borneo

Abstract

Shuttle Imaging Radar-B (SIR-B) images, obtained at two different incidence angles were analysed for discrimination and mapping of vegetation in the rainforest of Borneo. In an area of coastal lowland three units of forest canopy and two units of open surface cover were distinguished and mapped. The backscatter characteristics of the units were compared and analysed by ratioing digital image data of sample sites against digital data of reference sites. The ratios show that the backscatter for each surface unit is clustered in separate, mostly non-overlapping, domains at both incidence angles. Different rates of change with incidence angle indicate a strong angular dependence in the dominant backscatter mechanism of swamp that is not apparent for the other units. A corresponding digitally coregistered Landsat multispectral scanner (MSS) image helped to verify the spatial distribution of the mapped units in the coastal lowland. In the interior upland three major forest species associations that have contrasted canopy structures were not discriminated on the SIR-B images. Owing to perennial cloud cover the distribution and extent of the different canopies cannot be determined from Landsat MSS or other images obtained at optical wavelengths. The ability to discriminate and map different forest canopies in the interior of Borneo requires a wider radar response capability which may be obtainable at shorter wavelengths and with multipolariz-ation states.     

Landsat TM image texture and moisture variations of the soil surface under the rainforest of the Rondonia state,Brazil

A Landsat TM image, part of a rainforest located in the state of Rondonia, Brazil, was analysed using Fourier transform techniques in order to extract textural features. The lattice-like texture pattern recorded was then compared with the geological structural trends of the area, and measures of the soil surface moisture treated by geostatistical analysis. The consistency between the different sets of analysis suggests that the spatial distribution of moisture at the soil surface is related to the structure of the bedrock. The near infrared channel of the Landsat TM may be used to record trends in water content distributions at the soil surface under homogeneous forest canopies.     

Mapping northern land cover fractions using Landsat ETM+

The goal of fractional mapping is to obtain land cover fraction estimates within each pixel over a region. Using field, Ikonos and Landsat data at three sites in northern Canada, we evaluate a physical unmixing method against two modeling approaches to map five land cover fractions that include bare, grass, deciduous shrub, conifer, and water along an 1100 km north-south transect crossing the tree-line of northern Canada. Error analyses are presented to assess factors that affect fractional mapping results, including modeling method (linear least squares inversion (LLSI) vs. linear regression vs. regression trees), number of Landsat spectral bands (3 vs. 5), local and distant fraction estimation using locally and globally calibrated models, and spatial resolution (30 m vs. 90 m). The ultimate purpose of this study is to determine if reliable land cover fractions can be obtained for biophysical modeling over northern Canada from a three band, resampled 90 m Landsat ETM+ mosaic north of the tree-line. Of the three modeling methods tested, linear regression and regression trees with five spectral bands produced the best local fraction estimates, while LLSI produced comparable results when unmixing was sufficiently determined. However, distant fraction estimation using both locally and globally calibrated models was most accurate using the three spectral bands available in the Landsat mosaic of northern Canada at 30 m resolution, and only slightly worse at 90 m resolution. While local calibrations produced more accurate fractions than global calibrations, application of local calibration models requires stratification of areas where local endmembers and models are representative. In the absence of such information, globally calibrated linear regression and regression trees to estimate separate fractions is an acceptable alternative, producing similar root mean square error, and an average absolute bias of less than 2%.     

Measurement of oak tree density with Landsat TM data for estimating biogenic isoprene emissions in Tennessee,USA

Isoprene emissions from oak trees in the eastern USA play an important role in tropospheric ozone pollution. Oak trees (Quercus) emit an order of magnitude more isoprene than most other emitting tree species, and are by far the largest source of biogenic isoprene in the eastern US. We used Landsat TM data to measure oak tree abundance near Oak Ridge, Tennessee, to estimate fluxes of isoprene. The Landsat classification was performed using multi-date data, supervised classification techniques, and an iterative approach. Training sites were selected based on transect data, and ten vegetation classes were mapped. A supervised classification algorithm called the Spectral Angle Mapper was used to classify the data. Empirical vegetation emission data were used to estimate the isoprene flux from each of the vegetation classes. The resultant isoprene flux maps were compared with concentrations measured in the field, and a good correspondence was observed. We also compare the Landsat classification with three other landcover schemes including the USGS's Global Landcover Classification, which is based on AVHRR data. Results from these landcover classifications are used as input for models that predict tropospheric ozone production and are used to investigate ozone control strategies.     

An automated approach for segmenting and classifying a large sample of multi-date Landsat imagery for pan-tropical forest monitoring

The TREES-3 project of the Joint Research Centre aims at assessing tropical forest cover changes for the periods 1990-2000 and 2000-2010 using a sample-based approach. This paper refers to the 1990-2000 assessment. Extracts of Landsat satellite imagery (20 km × 20 km) are analyzed for these reference dates for more than 4000 sample sites distributed systematically across the tropical belt. For the processing and analysis of such a large amount of satellite imagery a robust methodological approach for forest mapping and change detection has been developed. This approach comprises two automated steps of multi-date image segmentation and object-based land cover classification (based on a supervised spectral library), followed by an intense phase of visual control and expert refinement. Image segmentation is done at two spatial scales, introducing the concept of a minimum mapping unit via the automated selection of a site-specific scale parameter. The automated segmentation of land cover polygons and the pre-classification of land cover types mainly aim at avoiding manual delineation and at reducing the efforts of visual interpretation of land cover to a reasonable level, making the analysis of 4000 sample sites feasible. The importance of visual control and correction can be perceived when comparing to the initial automatic classification result: about 20% of the polygon labels were changed through expert knowledge by visual interpretation. The component of visual refinement of the mapping results had also a notable impact on forest area and change estimates: for a set of sample sites in Southeast Asia (~ 90% of all sites of SE-Asia) the rate of change in tree cover (deforestation) was assessed at 0.9% and 1.6% before and after visual control, respectively.     

The multispectral separability of Costa Rican rainforest types with support vector machines and Random Forest decision trees

Estimating the extent of tropical rainforest types is needed for biodiversity assessment and carbon accounting. In this study, we used statistical comparisons to determine the ability of Landsat Thematic Mapper (TM) bands and spectral vegetation indices to discriminate composition and structural types. A total of 144 old-growth forest plots established in northern Costa Rica were categorized via cluster analysis and ordination. Locations for palm swamps, forest regrowth and tree plantations were also acquired, making 11 forest types for separability analysis. Forest types classified using support vector machines (SVM), a theoretically superior method for solving complex classification problems, were compared with the random forest decision tree classifier (RF). Separability comparisons demonstrate that spectral data are sensitive to differences among forest types when tree species and structural similarity is low. SVM class accuracy was 66.6% for all forest types, minimally higher than the RF classifier (65.3%). TM bands and the Normalized Difference Vegetation Index (NDVI) combined with digital elevation data notably increased accuracies for SVM (84.3%) and RF (86.7%) classifiers. Rainforest types discriminated here are typically limited to one or two categories for remote sensing classifications. Our results indicate that TM bands and ancillary data combined via machine learning algorithms can yield accurate and ecologically meaningful rainforest classifications important to national and international forest monitoring protocols.     

A method for calibrated maximum likelihood classification of forest types

A modification to the maximum likelihood algorithm was developed for classification of forest types in Sweden's part of the CORINE land cover mapping project. The new method, called the “calibrated maximum likelihood classification” involves an automated and iterative adjustment of prior weights until class frequency in the output corresponds to class frequency as calculated from objective (field-inventoried) estimates. This modification compensates for the maximum likelihood algorithm's tendency to over-represent dominant classes and under-represent less frequent ones. National forest inventory plot data measured from a five-year period are used to estimate relative frequency of class occurrence and to derive spectral signatures for each forest class. The classification method was implemented operationally within an automated production system which allowed rapid production of a country-wide forest type map from Landsat TM/ETM+ satellite data. The production system automated the retrieval and updating of forest inventory plots, a plot-to-image matching routine, illumination and haze correction of satellite imagery, and classification into forest classes using the calibrated maximum likelihood classification. This paper describes the details of the method and demonstrates the result of using an iterative adjustment of prior weights versus unadjusted prior weights. It shows that the calibrated maximum likelihood algorithm adjusts for the overclassification of classes that are well represented in the training data as well as for other classes, resulting in an output where class proportions are close to those as expected based on forest inventory data.     

Vegetation mapping of a tropical freshwater swamp in the Northern Territory,Australia: A comparison of aerial photography,Landsat TM and SPOT satellite imagery

The tropical wetland environments of northern Australia have ecological, social, cultural and economic values. Additionally, these areas are relatively pristine compared to the many other wetland environments in Australia, and around the world, that have been extensively altered by humans. However, as the remote northern coastline of Australia becomes more populated, environmental problems are beginning to emerge that highlight the need to manage the tropical wetland environments. Lack of information is currently considered to be a major factor restricting the effective management of many ecosystems and for the expansive wetlands of the Northern Territory, this is especially the case, as these areas are generally remote and inaccessible. Remote sensing is therefore an attractive technique for obtaining relevant information on variables such as land cover and vegetation status. In the current study, Landsat TM, SPOT (XS and PAN) and large-scale, true-colour aerial photography were evaluated for mapping the vegetation of a tropical freshwater swamp in Australia's Top End. Extensive ground truth data were obtained, using a helicopter survey method. Fourteen cover types were delineated from 1:15 000 air photos (enlarged to 1:5000 in an image processing system) using manual interpretation techniques, with 89% accuracy. This level of detail could not be extracted from any of the satellite image data sets, with only three broad land-cover types identified with accuracy above 80%. The Landsat TM and SPOT XS data provided similar results although superior accuracy was obtained from Landsat, where the additional spectral information appeared to compensate in part for the coarser spatial resolution. Two different classification algorithms produced similar results.     

Mapping forest patches and scattered trees from SPOT images and testing their ecological importance for woodland birds in a fragmented agricultural landscape

Scattered trees represent an important element within the agricultural matrix for birds. The aims of this study were to develop methods for mapping isolated trees from satellite imagery and to assess the importance of isolated trees for bird species richness. Field sampling of birds and plants was conducted at 120 sites in south-east Australia. We mapped tree cover from Landsat and SPOT images using a combination of spectral and segmentation based methods. Mapping of isolated trees as point objects was highly accurate (80–90%). Tree cover at spatial extents of 3–28 ha around sites explained 60% of the variability in woodland–dependent bird species richness. However, isolated trees in agricultural areas made just a small contribution to explaining the spatial variability in overall avian richness. This approach can be used for more extensive assessment of avian habitat quality from high spatial resolution images across a range of human modified landscapes.     

Cover Oceanic fronts in the northeastern Caribbean revealed by satellite ocean colour imaging

This paper is an attempt to assess the changes taking place in the vegetation communities found on the fringes of tropical rainforest in Nigeria over a 28-year period. The study made use of aerial photographs (1963 and 1973) and Landsat TM (1991). Landsat TM was digitally analysed, while aerial photos were subjected to an analogue procedure. The results show that the forest community receded at an annual rate of 1.90% between 1963 and 1991, while regressive succession was already in progress in the area. The intensity of agricultural activity was noted to have made the area vulnerable to savanna incursion. This paper highlights the prospects and problems of using remote sensing techniques in surveying Africa's rural landscape with its bewildering complexities. Some policy recommendations are suggested in the paper.     

A Neural Network Method for Efficient Vegetation Mapping

This article describes the application of a neural network method designed to improve the efficiency of map production from remote sensing data. Specifically, the ARTMAP neural network produces vegetation maps of the Sierra National Forest, in Northern California, using Landsat Thematic Mapper (TM) data. In addition to spectral values, the data set includes terrain and location information for each pixel. The maps produced by ARTMAP are of comparable accuracy to maps produced by a currently used method, which requires expert knowledge of the area as well as extensive manual editing. In fact, once field observations of vegetation classes had been collected for selected sites, ARTMAP took only a few hours to accomplish a mapping task that had previously taken many months. The ARTMAP network features fast online learning, so that the system can be updated incrementally when new field observations arrive, without the need for retraining on the entire data set. In addition to maps that identify lifeform and Calveg species, ARTMAP produces confidence maps, which indicate where errors are most likely to occur and which can, therefore, be used to guide map editing.     

Exploring the temporal density of Landsat observations for cropland mapping: experiments from Egypt,Ethiopia, and South Africa

ABSTRACT

Accurate land-use/land-cover mapping based on remote-sensing images depends on clear and frequent observations. This study aimed to explore how many Landsat images were needed within a year and when they should be acquired, for cropland mapping in Africa. Three Landsat footprints in Egypt (Path/Row: 177/039, 127 images), Ethiopia (Path/Row: 168/054, 98 images), and South Africa (Path/Row: 170/078, 207 images) from 1984 to 2016 were used together with spectral indices and a 30-m digital elevation model in a random forest-based supervised classification. Detailed exploration was conducted into the number and temporal distribution of Landsat images required. Our results indicated that average cropland mapping accuracies for these three sites ranged from 81.17% to 87.59% (Egypt), 54.43% to 79.72% (Ethiopia), and 28.11% to 59.35% (South Africa) using different numbers of images within a year. The overall cropland accuracies were improved with an increase in available Landsat images within a year and reached a relatively stable stage when more than five images were acquired in all three sites. Growing season images played a key role in identifying cropland, accounting for a 13.22% average accuracy improvement compared with non-growing season images. Therefore, at least five images are recommended from a computational efficiency perspective, although fewer images, as low as two growing season images, can also achieve good results in specific regions.     

Comparison of burned area estimates derived from SPOT-VEGETATION and Landsat ETM+ data in Africa: Influence of spatial pattern and vegetation type

An algorithm for burned area mapping in Africa based on classification trees was developed using SPOT-VEGETATION (VGT) imagery. The derived 1 km spatial resolution burned area maps were compared with 30 m spatial resolution maps obtained with 13 Landsat ETM+ scenes, through linear regression analysis. The procedure quantifies the bias in burned area estimation present in the low spatial resolution burned area map. Good correspondence was observed for seven sites, with values of the coefficient of determination (R²) ranging from 0.787 to 0.983. Poorer agreement was observed in four sites (R² values between 0.257 and 0.417), and intermediate values of R² (0.670 and 0.613) were obtained for two sites. The observed variation in the level of agreement between the Landsat and VGT estimates of area burned results from differences in the spatial pattern and size distribution of burns in the different fire regimes encompassed by our analysis. Small and fragmented burned areas result in large underestimation at 1 km spatial resolution. When large and compact burned areas dominate the landscape, VGT estimates of burned area are accurate, although in certain situations there is some overestimation. Accuracy of VGT burned area estimates also depends on vegetation type. Results showed that in forest ecosystems VGT maps underestimate substantially the amount of burned area. The most accurate estimates were obtained for woodlands and grasslands. An overall linear regression fitted with the data from the 13 comparison sites revealed that there is a strong relationship between VGT and Landsat estimates of burned area, with a value of R² of 0.754 and a slope of 0.803. Our findings indicate that burned area mapping based on 1 km spatial resolution VGT data provides adequate regional information.