Airborne Thematic Mapper Research Articles

The Sensitivity of Mapping Methods to Reference Data Quality: Training Supervised Image Classifications with Imperfect Reference Data

The accuracy of a map is dependent on the reference dataset used in its construction. Classification analyses used in thematic mapping can, for example, be sensitive to a range of sampling and data quality concerns. With particular focus on the latter, the effects of reference data quality on land cover classifications from airborne thematic mapper data are explored. Variations in sampling intensity and effort are highlighted in a dataset that is widely used in mapping and modelling studies; these may need accounting for in analyses. The quality of the labelling in the reference dataset was also a key variable influencing mapping accuracy. Accuracy varied with the amount and nature of mislabelled training cases with the nature of the effects varying between classifiers. The largest impacts on accuracy occurred when mislabelling involved confusion between similar classes. Accuracy was also typically negatively related to the magnitude of mislabelled cases and the support vector machine (SVM), which has been claimed to be relatively insensitive to training data error, was the most sensitive of the set of classifiers investigated, with overall classification accuracy declining by 8% (significant at 95% level of confidence) with the use of a training set containing 20% mislabelled cases.

Relationship between spatial heterogeneity and wavelength in multisensor airborne images

Variogram has been utilized to exploring the spatial heterogeneity of remote sensing images, especially its association with spatial resolution. However, very few attentions have been drawn on evaluating the spatial heterogeneity of multisensor airborne imagery and its relationship with spectral wavelength. Therefore, an investigation was carried out on multisensor airborne images to determine the relation between spatial heterogeneity and spectral wavelength for woodland, grass, and urban landscapes by applying variogram modeling. The airborne thematic mapper (ATM), compact airborne spectrographic imager (CASI), and Specim AISA Eagle airborne images at Harwood Forest, Monks wood, Cambridge, and River Frome areas, UK, were utilized. Results revealed that (1) the red band contained greater spatial variability than near-infrared wavelengths and other visible wavebands; (2) there was a steep gradient at the red edge in reference to its spatial variability of multisensor airborne images; (3) only for natural landscape such as woodland and grass, near-infrared waveband contains greater within-scene variations than the blue and green bands; (4) compared with the discrepancy of spatial resolution introduced by multisensor images (ATM, CASI, and Eagle), the specific landscape and spectral bands were more important in determining heterogeneity by means of original visible, near-infrared bands, and normalized difference vegetation index (NDVI). These findings remained us to be caution when combining and interpreting spatial variability and spatial structures derived from airborne images with different spatial resolution and spectral wavelength. Additionally, the outcomes of this study also have considerable implications in terms of designing and choosing suitable images for different applications.

Comparative spatio-spectral heterogeneity analysis using multispectral and hyperspectral airborne images

Knowledge of spatio-spectral heterogeneity within multisensor remote sensing images across visible, near-infrared and short wave infrared spectra is important. Till now, little comparative research on spatio-spectral heterogeneity has been conducted on real multisensor images, especially on both multispectral and hyperspectral airborne images. In this study, four airborne images, Airborne Thematic Mapper, Compact Airborne Spectrographic Imager, Specim AISA Eagle and AISI Hawk hyperspectral airborne images of woodland and heath landscapes at Harwood, UK, were applied to quantify and evaluate the differences in spatial heterogeneity through semivariogram modelling. Results revealed that spatial heterogeneity of multisensor airborne images has a close relationship with spatial and spectral resolution and wavelength. Within the visible, near-infrared spectra and short wave infrared spectra, greater spatial heterogeneity is generally observed from the relatively longer wavelength in short wave infrared spectra. There are dramatic changes across the red and red edge spectra, and the peak value is generally examined in the red middle or red edge wavelength across the visible and near-infrared spectra for vegetation or non-vegetation landscape respectively. In all, for real multisensor airborne images, the change in spatial heterogeneity with spatial resolution will accord with the change of support theory depending on whether dramatic change exists across the corresponding wavelength. Besides, if with close spatial resolution, the spatial heterogeneity of multispectral images might be far from the overall integration of these bands from the hyperspectral images involved. A comparative assessment of spatio-spectral heterogeneity using real hyperspectral and multispectral airborne images provides practical guidance for designing the placement and width of a spectral band for different applications and also makes a contribution to the understanding of how to reconcile spatial patterns generated by multisensors.

Using Airborne Remote Sensing to Study the Physical Dynamics of Lakes and the Spatial Distribution of Phytoplankton

Airborne remote sensing provides an ideal means of mapping the physical structure of lakes and studying the effects of mixing on the horizontal distribution of phytoplankton. Airborne surveys are easier to integrate with field studies than satellite measurements and currently provide the only means of collecting high resolution spectral and spatial data. In this review, I present some results from a series of airborne campaigns organised in the English Lake District between 1986 and 1991. The main instrument used for the surveys was a Daedalus 1268 Airborne Thematic Mapper. This instrument acquires data in 11 spectral bands ranging from the blue visible to the thermal infrared. The visible band data was used to study the spatial distribution of the phytoplankton and the thermal data to map the surface temperature of the lakes. Examples are presented to show how the instrument was used to track the trajectory of surface currents, study exchange processes with the shallow littoral and map the spatial distribution of phytoplankton. Some of the acquired images highlight patterns and processes that could not have been identified by any other means. These included the development of poorly mixed areas (‘dead zones’) in the littoral zone and the effect of power-boat movements on the surface temperature of Windermere. The review describes some of the methods used to process the acquired imagery and includes an introduction to the procedures used to analyse the spectral data now being gathered by a new generation of airborne instruments.

Evaluation of SVM, RVM and SMLR for Accurate Image Classification With Limited Ground Data

The accuracy of a conventional supervised classification is in part a function of the training set used, notably impacted by the quantity and quality of the training cases. Since it can be costly to acquire a large number of high quality training cases, recent research has focused on methods that allow accurate classification from small training sets. Previous work has shown the potential of support vector machine (SVM) based classifiers. Here, the potential of the relevance vector machine (RVM) and sparse multinominal logistic regression (SMLR) approaches is evaluated relative to SVM classification. With both airborne and spaceborne multispectral data sets, the RVM and SMLR were able to derive classifications of similar accuracy to the SVM but required considerably fewer training cases. For example, from a training set comprising 600 cases acquired with a conventional stratified random sampling design from an airborne thematic mapper (ATM) data set, the RVM produced the most accurate classification, 93.75%, and needed only 7.33% of the available training cases. In comparison, the SVM yielded a classification that had an accuracy of 92.50% and needed 4.5 times more useful training cases. Similarly, with a Landsat ETM+ (Littleport, Cambridgeshire, UK) data set, the SVM required 4.0 times more useful training cases than the RVM. For each data set, however, the classifications derived by each classifier were of similar magnitude, differing by no more than 1.25%. Finally, for both the ATM and ETM+ (Littleport) data sets, the useful training cases by SVM and RVM had distinct and potentially predictable characteristics. Support vectors were generally atypical but lay in the boundary region between classes in feature space while the relevance vectors were atypical but anti-boundary in nature. The SMLR also tended to mostly, but not always, use extreme cases that lay away from class boundary. The results, therefore, suggest a potential to design classifier-specific intelligent training data acquisition activities for accurate classification from small training sets, especially with the SVM and RVM.

Mapping tree species in temperate deciduous woodland using time‐series multi‐spectral data

AbstractQuestions: What is the optimum combination of image dates across a growing season for tree species differentiation in multi‐spectral data and how does species composition affect overstorey canopy density?Location: Monks Wood, Cambridgeshire, eastern England, UK.Methods: Six overstorey tree species were mapped using five Airborne Thematic Mapper images acquired across the 2003 growing season (17 March, 30 May, 16 July, 23 September, 27 October). After image pre‐processing, supervised maximum likelihood classification was performed on the images and on all two‐, three‐, four‐ and five‐date combinations. Relationships between tree species composition and canopy density were assessed using regression analyses.Results: The image with the greatest tree species discrimination was acquired on 27/10 when the overstorey species were in different stages of leaf tinting and fall. In this image, tree species were mapped with an overall classification accuracy (OCA) of 71% (kappa 0.63). A similar OCA was achieved from the other four images combined (OCA 72%, kappa 0.64). The highest classification accuracy was achieved by combining three images: 17 March, 16 July, 27 October. This achieved an OCA of 84% (kappa 0.79), increasing to 88% (kappa 0.85) after a post‐classification clump and sieve procedure. Canopy height and percentage cover of oak explained 72% of variance in canopy density.Conclusions: The ability to discriminate and map temperate deciduous tree species in airborne multi‐spectral imagery is increased using time‐series data. An autumn image supplemented with an image from both the green‐up and full‐leaf phases was optimum. The derived tree species map provides a more powerful ecological tool for determining woodland structural/compositional relationships than field‐based measures.

Use of airborne remote sensing to detect riverside Brassica rapa to aid in risk assessment of transgenic crops

High resolution descriptions of plant distribution have utility for many ecological applications but are especially useful for predictive modeling of gene flow from transgenic crops. Difficulty lies in the extrapolation errors that occur when limited ground survey data are scaled up to the landscape or national level. This problem is epitomized by the wide confidence limits generated in a previous attempt to describe the national abundance of riverside Brassica rapa (a wild relative of cultivated rapeseed) across the United Kingdom. Here, we assess the value of airborne remote sensing to locate B. rapa over large areas and so reduce the need for extrapolation. We describe results from flights over the river Nene in England acquired using Airborne Thematic Mapper (ATM) and Compact Airborne Spectrographic Imager (CASI) imagery, together with ground truth data. It proved possible to detect 97% of flowering B. rapa on the basis of spectral profiles. This included all stands of plants that occupied >2m square (>5 plants), which were detected using single-pixel classification. It also included very small populations (<5 flowering plants, 1-2m square) that generated mixed pixels, which were detected using spectral unmixing. The high detection accuracy for flowering B. rapa was coupled with a rather large false positive rate (43%). The latter could be reduced by using the image detections to target fieldwork to confirm species identity, or by acquiring additional remote sensing data such as laser altimetry or multitemporal imagery.

Monitoring transparency in inland water bodies using multispectral images

The aim of this research was to monitor transparency in inland water bodies from water reflectance field measurements and also from high and medium spatial resolution sensors. The most suitable band to measure transparency was determined by sampling irradiance, water reflectance and transparency at different depths. The results of this sampling show that the wavelength range 520–600 nm is the most suitable band for measuring transparency. Reflectance in this band was regressed against transparency field data to derive an empirical equation for transparency estimation. This equation was assessed using data from Landsat Thematic Mapper (TM) images from 1992 and Daedalus 1268 Airborne Thematic Mapper (ATM) data from 1997. The difference in the errors in the data observed between the two dates was attributed to the fact that in 1997 the sampling time was less than 3 h (1.5 h before and after the acquisition of the image), which allowed similar solar illumination conditions to be maintained, while in 1992 the sampling time was much longer. Once the equation was verified, water transparency monitoring was performed for a set of satellite images obtained from 1984 to 2000. The maps show that transparency in the water bodies responds to seasonal trends. Furthermore, this analysis enabled trophic state and Ecological Quality Ratio (EQR) transparency maps to be generated, which are very useful for the management of water bodies.

Airborne remote sensing of valley floor geoarchaeology using Daedalus ATM and CASI

AbstractThis paper investigates the potential of multispectral methods of airborne remote sensing for geological and archaeological prospection in river valleys, specifically the archaeologically rich, well‐documented aggregate landscape of the middle Trent Valley in the East Midlands of England. The paper reviews the systematic examination of Daedalus 1268 Airborne Thematic Mapper (ATM) and Compact Airborne Spectrographic Imager (CASI) multispectral data. Data from a single flight along the Trent Valley in 1996 were subject to a comprehensive suit of analytical techniques including the calculation of vegetation indices and multivariate analysis. The results suggest that both ATM and CASI data have great potential for detecting and characterizing archaeological and geological features. Vegetation indices relying on ratios of red to infrared reflectance were particularly effective at highlighting cropmarks, in some cases revealing crop variation not apparent in the visible spectrum. Copyright © 2008 John Wiley &amp; Sons, Ltd.

Monitoring active subglacial volcanoes: a case study using airborne remotely sensed imagery of Grímsvötn, Iceland

Grímsvötn is located beneath Vatnajökull, Europe's largest temperate ice cap. As a part of ongoing research on heat flux, morphological changes and volcanic processes at Grímsvötn, Airborne Thematic Mapper (ATM) imagery and aerial photographs were acquired in 2001. The thermal images illuminated distinct areas of geothermal activity along the southern caldera wall. In combination with meteorological data the images were used to estimate surface temperatures and heat flux from patches of open water along the margin of the ice shelf covering the Grímsvötn subglacial lake. It was found that water temperatures varied from 0°C to ∼45°C and that the heat flux to the atmosphere from open water varied from slightly negative values (net energy gain) up to 1000 W m−2. The total heat output from the ∼0.1 km2 of open water was estimated as ∼18 MW, about 1% of the base heat output of Grímsvötn. The aerial photographs were used to produce geomorphological maps of the caldera wall, including areas that cannot be safely mapped from the ground. This work indicates that thermal imagery can be an important supplement to ground‐based measurements, and that combined optical and thermal remote sensing is a useful tool for spatially detailed monitoring of inaccessible and partly ice‐covered volcanoes.

Remote thermal IR surveying to detect abandoned mineshafts in former mining areas

In former mining areas it is critical to locate unknown, abandoned mineshafts prior to the development of a site. Abandoned mineshafts are ground disturbances that have very localized effects on the morphology and the physical, chemical, drainage and moisture properties of the surface geological materials and thus thermo-physical properties. Remotely sensed thermal IR surveys provide the potential for a rapid, inexpensive and non-intrusive technique for mineshaft detection. The key parameters of thermal IR radiation and the application of remote thermal IR surveys to planning are described, using case histories from former mining areas in Lancashire, Yorkshire and Nottinghamshire. Field-measured IR temperature differences correlated well with different ground conditions caused by changes in vegetation, disturbance, compaction and moisture-drainage regimes. A thermal anomaly over an area of c . 6 m 2 above a known mineshaft was characterized by traces of methane and temperatures higher by 0.5–1 °C than those of the adjacent ground surface. Using thermal IR images, collected with the Daedalus 1260 Airborne Thematic Mapper, a scheme was developed to classify and map mineshafts with and without any observed visual characteristics. When applied using thermal imagery obtained from commercial flights the scheme identified several potential sites of abandoned mineshafts in an area designated for the redevelopment of the Nottingham Business Park, East Midlands. The thermal anomalies were associated with minor topographic features such as mounds, depressions and dereliction, as well as compositional features caused by coal enrichment and Coal Measures mudstone infill. These features had very little surface expression and were confirmed only using soil stripping.

An assessment of the use of remote sensing to map habitat features important to sustaining lamprey populations

Abstract This study describes the use of colour aerial photography and Airborne Thematic Mapper (ATM) multispectral imagery (420–1050 nm) to map gravel‐bed river habitats important to sustaining river lamprey (Lampetra fluviatilis) — an aquatic species with high conservation interest in Europe. The accuracy of the remote sensing approach was assessed by comparing the output from automated mapping of physical habitat features using image analysis and GIS and habitat maps derived from field survey. Unsupervised classification of the aerial photography and ATM imagery mapped hydraulic habitat such as pools and riffles with overall accuracies of 57% and 65.5% respectively. Submerged sand and silt deposits, which are important to river lamprey ammocoetes could not be identified on the imagery by simple unsupervised or supervised classification. They could be mapped indirectly, however, by association between their presence and morphological attributes such as backwaters and embayments in the bank profile detectable on the imagery. Remotely determining the likely presence and location of lamprey ammocoete habitat can be useful in planning detailed electrofishing campaigns to estimate lamprey densities, to assess conservation status and any change over time, and to determine the need for restoration strategies. The results suggest that on rivers wider than 20 m without a continuous wooded riparian fringe, and where there is good quality imagery, it is possible to automate mapping of physical habitat features important to many river species with high nature conservation interest. As a result organizations charged with monitoring and conserving river species should consider investing in airborne imagery covering river systems of high conservation value. Copyright © 2007 John Wiley &amp; Sons, Ltd.

An experimental approach to the measurement of the effects of water depth and substrate on optical and near infra‐red reflectance: a field‐based assessment of the feasibility of mapping submerged instream habitat

This study represent an assessment of the feasibility of using optical and near infra‐red wavelengths to map water depth and substrate type which are two of the primary components of river physical habitat. The objective was met by measuring reflectance using field spectroscopy of exposed and then progressively submerged (0–1 m) artificial substrates during rising tides on a tidal reach of the River Forth; this approach provided absolute control for substrate type. The field data were used to simulate multi‐spectral response as measured using Airborne Thematic Mapper (bands 1 to 7; 420nm to 900nm) since airborne remote sensing provides the way forward for synoptic mapping of small streams (<20 m), such as those typically found in Scotland. In the presence of sunlit conditions significant differences in reflectance over a broad range of wavelengths was evident with strong correlations with water depth and substrate. Differences in reflectance were most marked under shallow conditions (<0.75 m) and then progressively diminished. Application of simple linear correlation, band ratios, the Lyzenga algorithm and multiple discriminant analysis all suggested the possibility of mapping shallow river channel water depths. The results also showed that substrate type and the analysis technique selected also affected the optimum spectral wavelength for detection of channel substrate type and water depth. Overall the work suggests mapping channel depth and substrate type is possible using remotely sensed data in optical and near infra‐red wavelengths, using high spatial and spectral resolution multi‐spectral imagery. However it may be restricted to depths of less than 1 m where there is high organic colour and/or a significant periphyton cover.

Comparison and validation of Airborne Thematic Mapper thermal imagery using ground-based temperature data for Grímsvötn caldera, Vatnajökull, Iceland

Abstract Grímsvötn, Iceland's most active volcano, is also one of the most powerful geothermal areas in Iceland. This subglacial volcano is located in the centre of Vatnajökull, Europe's largest temperate ice cap, and it erupted most recently in 1998 and 2004. As part of continuing research on heat flux, morphological changes and volcanic processes at Grímsvötn, thermal anomalies were mapped using remote sensing Natural Environmental Research Council (NERC) Airborne Research and Survey Facility (ARSF) data. The 2001 Airborne Thematic Mapper (ATM) thermal images of the Grímsvötn subglacial caldera reveal distinct areas of geothermal activity and provide an overview of the thermal anomalies associated with water and rock exposures. A crater lake located on the 1998 eruption site is shown to have a surface temperature of 30–35 °C. There is a good correlation between the ARSF data and ground-based temperature measurements. The thermal images also revealed previously undetected areas of high heat flow. Factors that complicate the interpretation and comparison of different datasets from an ice-covered area include recent cornice collapses and variations in atmospheric humidity. To reduce uncertainty in future missions, temperature measurements should be made at points whose position is well constrained using differential global positioning system. In addition, humidity and temperature measurements should be made at the time of flight.

Spatial dynamics of estuarine water surface temperature from airborne remote sensing

This paper examines the applicability of airborne remote sensing to the characterization and quantification of the spatial dynamics of water surface temperature (WST) within estuaries. For this, a series of successive airborne Thematic Mapper thermal images was acquired of Kirkcudbright Bay, an estuary in the United Kingdom, on two dates. Spatial dynamics were determined by two techniques: (1) qualitatively, which involved interpreting features ‘by eye’; and (2) by the velocimetric technique of maximum cross correlation (MCC). Qualitatively, complex spatial dynamics were identified over a wide range of spatial and temporal scales, associated with the creation and dissipation of eddies, and the convergence and divergence of fronts, all of which were superimposed on the synoptic tidal flow. MCC was effective for producing synoptic velocity fields, but the minimum cell size was constrained by image noise, limiting the ability to analyze small-scale spatial dynamics. The effectiveness of MCC was related to the ratio of spatially correlated variance to spatially independent variance, as estimated from the variogram. Suggestions for optimizing the application of airborne remote sensing to estuarine spatial dynamics are presented.

Airborne remote sensing for landslide hazard assessment: a case study on the Jurassic escarpment slopes of Worcestershire, UK

This paper describes the application of airborne remote sensing to the study of landslides on the clay-dominated slopes of the Cotswolds Hills between the towns of Broadway, Worcestershire, and Snowshill, Gloucestershire, in the UK. The project involved an initial desk study, airphoto interpretation and field survey in order to provide detailed information about the nature and extent of the landsliding in the area. High-resolution Airborne Thematic Mapper (ATM) imagery was acquired by the NERC Airborne Remote Sensing Facility, which was subsequently processed in order to develop a remote sensing method for landslide identification using airborne multispectral data. A range of image processing methods are described including colour composite enhancement and thermal imaging, while the focus of the paper will be on the development of a semi-automated method of landslide identification using image classification and texture analysis. Results from the first stage of the study have shown that the use of image processing techniques such as colour composites and thermal imaging can provide information on the ground surface not visible in conventional aerial photography. In this case study, this has included more detailed geomorphological information on landsides in the area and the nature of the cambering and gulls at the top of the escarpment. The second part of the study has investigated the use of image texture enhancement as a method of landslide identification, applied in isolation and incorporated into an image classification scheme as a semi-automated method of landslide identification. Results from this investigation indicate that landslides can be identified automatically with a high degree of accuracy (83%) and that by using image texture, the image classification technique is able to successfully differentiate between areas of landslide activity and stable slopes in the airborne imagery. Its is clear from the results of this study that in order to identify landslides in these types of clay-dominated terrains, image texture must be used. Inland landslides, like those on the Cotswolds escarpment, do not have a spectral signature but they do exhibit a distinct spatial signature that allows them to be identified in airborne imagery using textural analysis. The semi-automated method of landslide identification described in this paper represents a rapid method of terrain evaluation and landslide hazard assessment, which can be undertaken prior to more detailed field mapping.

Spatial and temporal changes in salt marsh distribution in the Dee estuary, NW England, determined from aerial photographs

Vegetation changes in salt marsh communities of the Dee estuary, northwest England, were analysed with a combination of remote sensing techniques using data dating back to the 1950s. The distribution of communities in 1997 was classified using Airborne Thematic Mapper data and used to develop a methodology for the analysis of black and white photographs of the marsh. These methods were then applied retrogressively to a time sequence of monochrome photographs running from 1955 to 1975. At the apex of the salt marshes on the English shore of the Dee estuary, the marsh expanded dramatically to 1975, and consisted predominantly of pioneer and low marsh vegetation types. Between 1975 and 1997, however, there was only a slight increase in salt marsh area, but with an increase in mid and high marsh vegetation, replacing pioneer marsh. In a second area of the salt marsh on the English shore, a different pattern of salt marsh expansion was observed. The area occupied by marsh continued to increase right up to 1997, with extensive pioneer vegetation suggesting a process of continuing expansion. However, the pattern of marsh colonisation appeared to be different in 1997 compared to 1975. The significance of the changes in salt marsh distribution within the Dee estuary are discussed in relation to the historical pattern of salt marsh colonisation, the importance of Spartina anglica in the process and the implications for strategic management of the estuarine resources.

Supervised image classification by MLP and RBF neural networks with and without an exhaustively defined set of classes

The absence of assumptions about the dataset to be classified is one of the major attractions of neural networks for supervised image classification applications. Classification by a neural network does, however, make assumptions about the classes. One key assumption typically made is that the set of classes has been defined exhaustively. If this assumption is unsatisfied, cases of an untrained class will be present and commissioned into the set of trained classes to the detriment of classification accuracy. This was observed in land cover classifications derived with multi-layer perceptron (MLP) and radial basis function (RBF) neural networks in which the presence of an untrained class resulted in a ∼12.5% decrease in the accuracy of crop classifications derived from airborne thematic mapper data. However, since the RBF network partitions feature space locally rather than globally as with the MLP, it was possible to reduce the commission of atypical cases into the set of trained classes through the setting of post-classification thresholds on the RBF network's outputs. As a result it was possible to identify and exclude some cases of untrained classes from a classification with a RBF network which resulted in an increase in classification accuracy.

Mapping of water constituents in Lake Constance using multispectral airborne scanner data and a physically based processing scheme

A physically based method to determine water constituents in case II waters from imaging spectrometer data is presented. The processing scheme includes modules for sunglitter correction, aerosol retrieval, atmospheric correction, and the retrieval of water constituents from subsurface reflectance spectra in case II waters. For the aerosol retrieval, the sun glitter, and the atmospheric correction modules, an azimuthally resolved radiative transfer model for a multilayer atmosphere-ocean system with a flat water surface is used. The retrieval of water constituents is achieved by an iterative fitting algorithm to adjust modeled and measured underwater reflectances. The processing scheme is applied to airborne multispectral scanner Daedalus AADS1268 in ATM (airborne thematic mapper) modus. Owing to the sensor properties (reduced radiometric sensitivity of the blue channel), it was not possible to work completely without additional data; Gelbstoff absorption could not be retrieved. It had to be inputted and was kept constant over the mission area. The processing system has been tested for multitemporal mapping of water constituents at Lake Constance. The results show the system as an efficient and reliable tool for the inversion of remote sensing data from natural waters.

Mapping intertidal estuarine sediment grain size distributions through airborne remote sensing

The intertidal environments of estuaries represent critical exchange environments of sediment and sediment bound contaminants. Ecological and sedimentological related investigations of these environments require monitoring methods that provide rapid spatially representative data on sediment grain size distribution. Remote sensing has the potential to provide synoptic information of intertidal environments. Previous in situ and laboratory-based reflectance investigations have demonstrated that for effective quantification of sediment grain size distributions, remote sensing platforms must include measurements within the short-wave infrared (SWIR). In addition, the timing of image acquisition, in relation to tidal cycles and sediment moisture content, is critical in optimising the spectral differences between the coarser sand and finer ‘mud’ fraction of sediments. Daedalus 1268 Airborne Thematic Mapper (ATM) has been identified as an appropriate platform and sensor for providing accurate synoptic maps of estuarine sediment distributions. This paper presents the results from the application of ATM 1.75 m resolution data to the mapping of surface sediment grain-size distributions across intertidal areas of Ribble Estuary, Lancashire, UK. ATM imagery was acquired after the intertidal area was exposed to strong summer drying conditions. Pre-processing and linear unmixing of the imagery collected of the intertidal zone following a period of drying allowed accurate sub-pixel determinations (1.75 m resolution) of sediment clay ( r 2=0.79) but less accurate for sand ( r 2=0.60). The results also demonstrate deterioration in the image calibration with increasing sediment moisture content and microphytobenthos cover. However, recombining the subpixel end member abundances through multivariate regression analysis improved the image calibration significantly for both sediment clay and sand content ( r 2>0.8) for imagery collected in both dryer and wetter conditions. These results demonstrate that ATM data, or similar, can be used to gain quantitative information on intertidal sediment distributions and such data has application to a wide variety of estuarine research.