Remote sensing imagery in vegetation mapping: a review

Aims Mapping vegetation through remotely sensed images involves various considerations, processes and techniques. Increasing availability of remotely sensed images due to the rapid advancement of remote sensing technology expands the horizon of our choices of imagery sources. Various sources of imagery are known for their differences in spectral, spatial, radioactive and temporal characteristics and thus are suitable for different purposes of vegetation mapping. Generally, it needs to develop a vegetation classification at first for classifying and mapping vegetation cover from remote sensed images either at a community level or species level. Then, correlations of the vegetation types (communities or species) within this classification system with discernible spectral characteristics of remote sensed imagery have to be identified. These spectral classes of the imagery are finally translated into the vegetation types in the image interpretation process, which is also called image processing. This paper presents an overview of how to use remote sensing imagery to classify and map vegetation cover.Methods Specifically, this paper focuses on the comparisons of popular remote sensing sensors, commonly adopted image processing methods and prevailing classification accuracy assessments.Important findings The basic concepts, available imagery sources and classification techniques of remote sensing imagery related to vegetation mapping were introduced, analyzed and compared. The advantages and limitations of using remote sensing imagery for vegetation cover mapping were provided to iterate the importance of thorough understanding of the related concepts and careful design of the technical procedures, which can be utilized to study vegetation cover from remote sensed images.     

Spatial patterns of the Chaco vegetation of central Argentina: Integration of remote sensing and phytosociology

Abstract. In this study we report the first application of Landsat TM imagery to Chaco vegetation studies at a regional scale in Argentina. We produced a map showing 13 clearly differentiated land‐cover types, and described the composition and structure of the plant communities, in an area of almost 42002 km² in central Argentina. The land‐cover map obtained shows that the Chaco vegetation in central Argentina is highly disturbed. In the lowland part of the area the dominant land‐cover types are largely cultural landscapes and substitute shrublands, which have displaced the original Chaco forests, leaving only small isolated remnants generally confined to sites with some kind of constrain for agriculture. The use of TM images and the multivariate analysis of phytosociological data showed a qualified, high accuracy mapping capability for land‐cover types in the Chaco region (ca. 85% overall accuracy). Our results highlight the utility of TM and field data in a subtropical to warm‐temperate region, which is promising where other ancillary data are not available and a rapid acquisition of reliable vegetation data is required, so constituting a starting point for an imperative and more extensive classification and mapping of the endangered Chaco region.     

Classifying the Neotropical savannas of Belize using remote sensing and ground survey

Aim This paper evaluates a method of combining data from GPS ground survey with classifications of medium spatial resolution LANDSAT imagery to distinguish variations within Neotropical savannas and to characterize the boundaries between savanna areas and the associated gallery forests, seasonally dry forests and wetland communities. Location Rio Bravo Conservation Area, Orange Walk District, Belize, Central America. Methods Dry season LANDSAT data for 10 April 1993 and 9 March 2001 covering a conservation area of 240,000 acres (97,459 ha), were rectified to sub‐pixel accuracy using ground control points positioned by GPS ground survey. The 1993 image was used to assess the accuracy with which the boundaries between the savanna matrix and gallery forests, high forests, wetlands and water bodies could be discriminated. The image was classified by a maximum likelihood (ML) classifier and the shapes and areas of forest and wetland classes were compared with an interpretation of these land cover types from 1 : 24,000 aerial photography, mapped at 1 : 50,000 scale in 1993. The 2001 image was used to assess whether different subtypes of savanna could be distinguished from LANDSAT data. This required the creation of a reference (‘ground truth’) data set for testing classifications of the image. One hundred and sixty sample patches (650 ha, distributed over an area of 7000 ha) of ten sub‐types of savanna vegetation and associates identified using a physiognomic classification scheme, were delineated on the ground by GPS and divided into two subsets for training and testing. Continuous classifications of LANDSAT data covering the savannas were developed that estimated potential contributions from up to five sub‐types of land cover (grassland, wetland, pine woodland, gallery forest and palmetto). The accuracy of each classification was assessed by comparison against ground data. An ML classification was also produced for the 2001 image using the same areas for training. This allowed a comparison of the relative accuracy of both continuous and Boolean ML methods for classifying savanna areas. Results The boundary between savannas and evergreen forests, gallery forests and open water in the study region could be delineated by the ML classifier to within 2 pixels (60 m) using LANDSAT imagery. However, the constituent sub‐types within the savanna were poorly discriminated. Whilst the shape and extent of closed canopy forest, gallery forest, wetlands and water bodies agreed closely with the distributions interpreted from aerial photography, classes such as ‘open pine savanna’ or ‘grassland’ were only 45–65% accurate when tested against ground data. A continuous classification, estimating the proportions of three savanna vegetation subtypes (grassland, marshland and woodland) present in each pixel, correctly classified more of the ground data for these cover types than the comparable ML result. Proportional mixtures of the land cover estimated by the continuous classifier also compared realistically with the vegetation formations observed along ground transects. Main conclusions By using GPS, a ground survey of vegetation cover was accurately matched to remotely sensed imagery and the accuracy of delineating boundaries and classifying areas of savanna was assessed directly. This showed that ML classification techniques can reliably delineate the boundaries of savannas, but continuous classifiers more accurately and realistically represent the distribution of the subtypes comprising savanna land cover. By combining these ground survey and image classification methods, medium spatial resolution satellite sensor data can provide an affordable means for land managers to assess the nature, extent and distribution of savanna formations. Over time, using the archives of LANDSAT (and SPOT) data together with marker sites surveyed in the field, quantitative changes in the extents and boundaries of savannas in response to both natural (e.g. fire, hurricane and drought) and anthropogenic (e.g. cutting and disturbance) factors can be assessed.     

Object-based classification of land cover and tree species by integrating airborne LiDAR and high spatial resolution imagery data

We evaluated the effectiveness of integrating discrete return light detection and ranging (LiDAR) data with high spatial resolution near-infrared digital imagery for object-based classification of land cover types and dominant tree species. In particular we adopted LiDAR ratio features based on pulse attributes that have not been used in past studies. Object-based classifications were performed first on land cover types, and subsequently on dominant tree species within the area classified as trees. In each classification stage, two different data combinations were examined: LiDAR data integrated with digital imagery or digital imagery only. We created basic image objects and calculated a number of spectral, textural, and LiDAR-based features for each image object. Decision tree analysis was performed and important features were investigated in each classification. In the land cover classification, the overall accuracy was improved to 0.975 when using the object-based method and integrating LiDAR data. The mean height value derived from the LiDAR data was effective in separating “trees” and “lawn” objects having different height. As for the tree species classification, the overall accuracy was also improved by object-based classification with LiDAR data although it remained up to 0.484 because spectral and textural signatures were similar among tree species. We revealed that the LiDAR ratio features associated with laser penetration proportion were important in the object-based classification as they can distinguish tree species having different canopy density. We concluded that integrating LiDAR data was effective in the object-based classifications of land cover and dominant tree species.     

Remote sensing of upland vegetation: the potential of high spatial resolution satellite sensors

Aim Traditional methodologies of mapping vegetation, as carried out by ecologists, consist primarily of field surveying or mapping from aerial photography. Previous applications of satellite imagery for this task (e.g. Landsat TM and SPOT HRV) have been unsuccessful, as such imagery proved to have insufficient spatial resolution for mapping vegetation. This paper reports on a study to assess the capabilities of the recently launched remote sensing satellite sensor Ikonos, with improved capabilities, for mapping and monitoring upland vegetation using traditional image classification methods. Location The location is Northumberland National Park, UK. Methods Traditional remote sensing classification methodologies were applied to the Ikonos data and the outputs compared to ground data sets. This enabled an assessment of the value of the improved spatial resolution of satellite imagery for mapping upland vegetation. Post‐classification methods were applied to remove noise and misclassified pixels and to create maps that were more in keeping with the information requirements of the NNPA for current management processes. Results The approach adopted herein for quick and inexpensive land cover mapping was found to be capable of higher accuracy than achieved with previous approaches, highlighting the benefits of remote sensing for providing land cover maps. Main conclusions Ikonos imagery proved to be a useful tool for mapping upland vegetation across large areas and at fine spatial resolution, providing accuracies comparable to traditional mapping methods of ground surveys and aerial photography.     

A new land-cover map of Africa for the year 2000

Aim In the framework of the Global Land Cover 2000 (GLC 2000), a land‐cover map of Africa has been produced at a spatial resolution of 1 km using data from four sensors on‐board four different Earth observing satellites. Location The map documents the location and distribution of major vegetation types and non‐vegetated land surface formations for the entire African continent plus Madagascar and the other surrounding islands. Methods The bulk of these data were acquired on a daily basis throughout the year 2000 by the VEGETATION sensor on‐board the SPOT‐4 satellite. The map of vegetation cover has been produced based upon the spectral response and the temporal profile of the vegetation cover. Digital image processing and geographical information systems techniques were employed, together with local knowledge, high resolution imagery and expert consultation, to compile a cartographic map product. Radar data and thermal sensors were also used for specific land‐cover classes. Results A total of 27 land cover categories are documented, which has more thematic classes than previously published land cover maps of Africa contain. Systematic comparison with existing land cover data and 30‐m resolution imagery from Landsat are presented, and the map is also compared with other pan‐continental land cover maps. The map and digital data base are freely available for non‐commercial uses from http://www.gvm.jrc.it/tem/africa/products.htm Main conclusions The map improves our state of knowledge of the land‐cover of Africa and presents the most spatially detailed view yet published at this scale. This first version of the map should provide an important input for regional stratification and planning purposes for natural resources, biodiversity and climate studies.     

Vegetation mapping using hierarchical object‐based image analysis applied to aerial imagery and lidar data

Aims: The primary objective of this study is to map the distribution and quantify the cover of vegetation alliances over the entirety of San Clemente Island (SCI). To this end, we develop and evaluate the mapping method of hierarchical object‐based classification with a rule‐based expert system. Location: San Clemente Island, California, USA. Methods: We developed and tested an approach based on hierarchical object‐based classification with a rule‐based expert system to effectively map vegetation communities on SCI following the Manual of California Vegetation classification system. In this mapping approach, the shrub species defining each vegetation community and non‐shrub growth forms were first mapped using aerial imagery and lidar data, then used as input in an automated mapping rule set that incorporates the percent cover rules of a field‐based mapping rule set. Results: The final vegetation map portrays the distribution of 19 vegetation communities across SCI, with the largest areas comprised of California Annual and Perennial Grassland (35%) and three types of coastal sage scrub and maritime succulent scrub, comprising a combined 53% of the area. Map accuracy was assessed to be 79% based on fuzzy methods and 61% with a traditional accuracy assessment. The accuracy of tree identification was assessed to be 81%, but species‐level tree accuracy was 45%. Conclusions: Semi‐automated approaches to vegetation community mapping can produce repeatable maps over large spatial extents that facilitate ecological management efforts. However, some low‐statured shrub community types were difficult to differentiate due to patchy canopies of co‐occurring species including abundant non‐native grasses characteristic of complex disturbance histories. Species‐level tree mapping accuracy was low due to the difficulty of identifying species within poorly illuminated canyons, resulting from sub‐optimal image acquisition timing.     

Summit-to-sea mapping and change detection using satellite imagery: tools for conservation and management of coral reefs

Continuous summit-to-sea maps showing both land features and shallow-water coral reefs have been completed in Puerto Rico and the U.S. Virgin Islands, using circa 2000 Landsat 7 Enhanced Thematic Mapper (ETM+) Imagery. Continuous land/sea terrain was mapped by merging Digital Elevation Models (DEM) with satellite-derived bathymetry. Benthic habitat characterizations were created by unsupervised classifications of Landsat imagery clustered using field data, and produced maps with an estimated overall accuracy of>75% (Tau coefficient >0.65). These were merged with Geocover-LC (land use/land cover) data to create continuous land/ sea cover maps. Image pairs from different dates were analyzed using Principle Components Analysis (PCA) in order to detect areas of change in the marine environment over two different time intervals: 2000 to 2001, and 1991 to 2003. This activity demonstrates the capabilities of Landsat imagery to produce continuous summit-to-sea maps, as well as detect certain changes in the shallow-water marine environment, providing a valuable tool for efficient coastal zone monitoring and effective management and conservation.     

Regression Tree Analysis of satellite and terrain data to guide vegetation sampling and surveys

Abstract. Monitoring of regional vegetation and surface biophysical properties is tightly constrained by both the quantity and quality of ground data. Stratified sampling is often used to increase sampling efficiency, but its effectiveness hinges on appropriate classification of the land surface. A good classification must be sufficiently detailed to include the important sources of spatial variability, but at the same time it should be as parsimonious as possible to conserve scarce and expensive degrees of freedom in ground data. As part of the First ISLSCP (International Satellite Land Surface Climatology Program) Field Experiment (FIFE), we used Regression Tree Analysis to derive an ecological classification of a tall grass prairie landscape. The classification is derived from digital terrain, land use, and land cover data and is based on their association with spectral vegetation indices calculated from single-date and multi-temporal satellite imagery. The regression tree analysis produced a site stratification that is similar to the a priori scheme actually used in FIFE, but is simpler and considerably more effective in reducing sample variance in surface measurements of variables such as biomass, soil moisture and Bowen Ratio. More generally, regression tree analysis is a useful technique for identifying and estimating complex hierarchical relationships in multivariate data sets.     

Satellite surface reflectance improves habitat distribution mapping: a case study on heath and shrub formations in the Cantabrian Mountains (NW Spain)

Aim We aim to map the distribution of four heath and shrub formations constituting habitats of high conservation priority in Europe, whose occurrence is strongly dependent on human activities. Specifically, we assess whether the use of LANDSAT data in habitat distribution modelling may account for land use management, allowing accurate mapping of real distribution patterns. In particular, we explore whether reflectance values may be a better alternative to other remote sensing data traditionally used in modelling approaches (i.e. spectral vegetation indices and classified land cover maps). Finally, we test whether modelling performance is affected by the ecological traits of the dominant species of the target formations. Location Cantabrian Mountains (NW Spain). Methods We generated maps for the four formations (two specialists vs. two generalists) using MaxEnt. First, we ran the models with environmental predictors only (topography, climate, lithology and human disturbances). Then, we compared the advantages of including, in turn, different data derived from LANDSAT imagery: reflectance values (corresponding to different wavelength channels of the multispectral image), a spectral index and a land cover map. We assessed changes in explanatory power and also in the formation’s predicted distribution patterns. Results Formations dominated by specialist species were accurately mapped on a base of environmental variables only, whereas those dominated by generalists were overpredicted. Average mean temperature, southness and distance to urban areas were the variables contributing most in predictions of environmental models. LANDSAT channels increased the accuracy of all models, but mainly those for formations dominated by generalist species. They showed advantages against other remote sensing data traditionally used in modelling approaches. Main conclusions Habitat distribution models allowed accurate mapping of heath and shrub formations. The use of reflectance values as predictors improved the accuracy of the models, particularly for formations dominated by generalist species, supplying environmental information that was otherwise unavailable.     

SPATIAL PATTERNS AND ECOLOGICAL DETERMINANTS OF BENTHIC ALGAL ASSEMBLAGES IN MID-ATLANTIC STREAMS, USA

We attempted to identify spatial patterns and determinants for benthic algal assemblages in Mid-Atlantic streams. Periphyton, water chemistry, stream physical habitat, riparian conditions, and land cover/use in watersheds were characterized at 89 randomly selected stream sites in the Mid-Atlantic region. Cluster analysis (TWINSPAN) partitioned all sites into six groups on the basis of diatom species composition. Stepwise discriminant function analysis indicated that these diatom groups can be best separated by watershed land cover/use (percentage forest cover), water temperature, and riparian conditions (riparian agricultural activities). However, the diatom-based stream classification did not correspond to Omernik's ecoregional classification. Algal biomass measured as chl a can be related to nutrients in habitats where other factors do not constrain accumulation. A regression tree model indicated that chl a concentrations in the Mid-Atlantic streams can be best predicted by conductivity, stream slope, total phosphorus, total nitrogen, and riparian canopy coverage. Our data suggest that broad spatial patterns of benthic diatom assemblages can be predicted both by coarse-scale factors, such as land cover/use in watersheds, and by site-specific factors, such as riparian conditions. However, algal biomass measured as chl a was less predictable using a simple regression approach. The regression tree model was effective for showing that ecological determinants of chl a were hierarchical in the Mid-Atlantic streams.     

Remote sensing imagery for resource inventories in central Africa: The importance of detailed field data

The rate of rain forest clearing throughout central Africa is of national and international interest because it affects both the region's contribution to global warming and impacts the sustainable productive capacity of its natural resource base. The size and inaccessibility of much of central Africa makes remote sensing imagery the most suitable data source for regional land cover mapping and land transformation monitoring. Present image availability is poor. Most regional studies have had to rely on coarse resolution AVHRR 1 km data that fails to detect the small-scale agricultural clearings that are the primary cause of land cover change throughout the region. This study demonstrates that higher spatial resolution Landsat MSS imagery, which comprises the most available, geographically comprehensive and longest time series dataset, is too coarse to map land cover in low population density areas typical of most of central Africa. Furthermore, this study cautions that the use of high resolution imagery without detailed collateral field data on population density and land use practices while generating superficially plausible results, will most probably produce highly inaccurate estimates of land cover and land transformation. Policies for future regional remote sensing surveys of central Africa should focus on acquisition of higher spatial, spectral, and radiometric resolution imagery and must be accompanied by detailed, systematic field data collection.     

A new forest cover map of continental southeast Asia derived from SPOT‐VEGETATION satellite imagery

Question: Can recent satellite imagery of coarse spatial resolution support forest cover assessment and mapping at the regional level? Location: Continental southeast Asia. Methods: Forest cover mapping was based on digital classification of SPOT4‐VEGETATION satellite images of 1 km spatial resolution from the dry seasons 1998/1999 and 1999/2000. Following a geographical stratification, the spectral clusters were visually assigned to land cover classes. The forest classes were validated by an independent set of maps, derived from interpretation of satellite imagery of high spatial resolution (Landsat TM, 30 m). Forest area estimates from the regional forest cover map were compared to the forest figures of the FAO database. Results: The regional forest cover map displays 12 forest and land cover classes. The mapping of the region's deciduous and fragmented forest cover remained challenging. A high correlation was found between forest area estimates obtained from this map and from the Landsat TM derived maps. The regional and sub‐regional forest area estimates were close to those reported by FAO. Conclusion: SPOT4‐VEGETATION satellite imagery can be used for mapping consistently and uniformly the extent and distribution of the broad forest cover types at the regional scale. The new map can be considered as an update and improvement on existing regional forest cover maps.     

Autumn olive (<Emphasis Type="Italic">Elaeagnus umbellata</Emphasis>) presence and proliferation on former surface coal mines in Eastern USA

Invasive plants threaten native plant communities. Surface coal mines in the Appalachian Mountains are among the most disturbed landscapes in North America, but information about land cover characteristics of Appalachian mined lands is lacking. The invasive shrub autumn olive (Elaeagnus umbellata) occurs on these sites and interferes with ecosystem recovery by outcompeting native trees, thus inhibiting re-establishment of the native woody-plant community. We analyzed Landsat 8 satellite imagery to describe autumn olive’s distribution on post-mined lands in southwestern Virginia within the Appalachian coalfield. Eight images from April 2013 through January 2015 served as input data. Calibration and validation data obtained from high-resolution aerial imagery were used to develop a land cover classification model that identified areas where autumn olive was a primary component of land cover. Results indicate that autumn olive cover was sufficiently dense to enable detection on approximately 12.6 % of post-mined lands within the study area. The classified map had user’s and producer’s accuracies of 85.3 and 78.6 %, respectively, for the autumn olive coverage class. Overall accuracy was assessed in reference to an independent validation dataset at 96.8 %. Autumn olive was detected more frequently on mines disturbed prior to 2003, the last year of known plantings, than on lands disturbed by more recent mining. These results indicate that autumn olive growing on reclaimed coal mines in Virginia and elsewhere in eastern USA can be mapped using Landsat 8 Operational Land Imager imagery; and that autumn olive occurrence is a significant landscape vegetation feature on former surface coal mines in the southwestern Virginia segment of the Appalachian coalfield.     

Dealing with vagueness in complex forest landscapes: A soft classification approach through a niche-based distribution model

The increasing interest in biodiversity conservation has led to the development of new approaches to facilitate ecologically based conservation policies and management plans. In this context, the development of effective methods for the classification of forest types constitutes a crucial issue as forests represent the most widespread vegetation structure and play a key role in ecosystem functioning. In this study a maximum entropy approach (Maxent) to forest type classification in a complex Mediterranean area, has been investigated. Maxent, a niche-based model of species/habitat distribution, allowed researchers to estimate the potential distribution of four forest types: Holm oak, Mixed oak, Mixed broadleaved and Riparian forests. The Maxent model's internal tests have proved a powerful tool for estimating the model's accuracy and analyzing the effects of the most important variables in the produced models. Moreover the comparison with a spectral response-based fuzzy classification, showed a higher accuracy in the Maxent outputs, demonstrating how the use of environmental variables, combined with spectral information in the classification of natural or semi-natural land cover classes, improves map accuracies. The modeling approach followed by this study, taking into account the uncertainty proper of the natural ecosystems and the use of environmental variables in land cover classification, can represent a useful approach to making more efficient and effective field inventories and to developing effective conservation policies.     

基于多源遥感信息融合的广东省土地利用分类方法——以雷州半岛为例

准确高效的获取土地利用信息对生态环境评价非常重要。广东省地处华南热带和亚热带季风气候区,经济作物种类繁多,土地覆盖破碎,为土地利用精确分类带来很大不确定性,而常年多云雨的天气也为有效光学影像的获取带来困难。为提高土地覆盖分类精度,以雷州半岛为实验区,综合应用Landsat-TM/ETM、多时相HJ光学影像,以及X波段Terra SAR数据,通过分析不同地物类型在光谱、极化以及多时相特征上的差别,对原始图像进行特征提取。在此基础上融合多源遥感信息的地物特征运用面向对象土地覆盖分类方法获取研究区高精度的土地利用信息。结果显示这一方法能有效提高土地覆盖利用信息获取精度,为研究生态环境变化提供更准确的数据支持。     

The application of multivariate land classification to vegetation survey in the Wicklow Mountains,Ireland

Multivariate land classification and land cover mapping by aerial photographic interpretation were used to model spatial variation of land cover in the Wicklow Mountains, Ireland and to structure a stratified random sampling programme of upland blanket bog vegetation. The total area of blanket bog with gully-erosion features was estimated as 33% of the area studied. Vegetation with hand peat-cutting patterns was estimated at 5%, and there was 35% undissected (intact) vegetation. There were differences between land classes in the estimated area of land cover with gully-erosion features or hand peat-cutting patterns.Sample vegetation quadrats, stratified by land class and aerial photographic land cover type, were grouped by their plant species composition. The groups represented ombrotrophic mire, soligenous mire and shrub heath vegetation. There was significant association between vegetation group and land class, related to variation in regional landscape type, but no significant association between vegetation group and the aerial photographic land cover types, undissected (intact) and dissected (gullied and cut-over) peats. It is proposed that the similarity of vegetation between undissected and dissected blanket bog is related to vegetation regeneration. The need to consider differences in vegetation distribution, composition and dynamics in ecological management strategies is emphasised. The study demonstrated the value of stratified random field sampling for cost-efficient regional ecological assessment in upland blanket bog landscapes typified by the Wicklow mountains, Ireland.