This site is using cookies to collect anonymous visitor statistics and enhance the user experience. OK | Find out more

NERC logo
Skip to content

Recent ARSF-supported Science: 2005

2005 Normal UK flying:

2005 Hyperspectral UK flying:

GB05/01 William Hanson: The application of airborne remote sensing techniques in archaeology: a comparative study

Aerial photography has made the single most important contribution to our improved appreciation of the density, diversity and distribution of archaeological sites since WWII. This is particularly the case for areas of intensive lowland agriculture where ploughed-out sites are known only from marks in crops growing above them. However, reconnaissance for such cropmarks is not equally effective throughout the lowlands because of the particular conditions of drier weather, well-drained soils and arable agriculture required before they become visible. There is, therefore, considerable bias in the discovery and, consequently, known distribution of archaeological sites in favour of the drier eastern side of Scotland with its higher percentage of arable agriculture, as opposed to the west with its wetter climate and greater proportion of grazing land. Given that the appearance of cropmarks is linked to moisture stress in growing plants, they are potentially detectable at bandwidths outside the visible and before they become apparent therein. Using carefully selected case study sites in Lowland Scotland, one in the east and one in the west for the purposes of comparison, this project seeks to examine the extent to which multispectral and hyperspectral imagery can reveal otherwise invisible archaeological sites, particularly in pasture land.

GB05/02 Alex Souza: Air-sea and Sediment processes within the Hilbre Coastal Observatory

The Project is part of the POL science plan and aims to observe sediment transport dynamics within the Dee Estuary. As part of the POL Coastal Observatory an X-Band radar has been deployed at Hilbre Island to monitor bathymetric changes at the mouth of the Dee Estuary. The study will also make a set of marine in-situ observations which are focused on studying the small scale processes, turbulence and SPM resuspension. Understanding these processes is important to predict the large scale variability. Three-dimensional numerical modelling will be used to help interpret the dynamics within the Estuary. The project will also study air-sea interaction and exchanges of momentum at the sea surface and sea-bed.

The use of the NERC Aircraft will be of great advantage in giving a synoptic view of the SPM/CDOM and chlorophyll distributions within the Estuary, as well as providing us with high-resolution bathymetry of the intertidal area. The overlapping aerial photographs will also give information about wave refraction over this area: if the wind speed is sufficiently high they will give white cap coverage and the speed of propagation of the breaking wave fronts.

GB05/03 Mat Disney: Retrieval of structural and biochemical parameters from multiangle, multispectral airborne data

The project aims to relate airborne reflectance data to canopy structural and biochemical parameters relating to canopy state and productivity. In situ measurements of CO2 flux will be used to quantify net (daily and annual) fluxes of energy and carbon, which will in turn be compared to EO measures known to be related to vegetation productivity such as light-use-efficiency (LUE) and photochemical reflectance index (PRI). Canopy structural parameters will be extracted from airborne data using canopy scattering models driven by estimates of 3D canopy architecture. This work will augment existing development of 3D forest modelling and reflectance simulation methods. Biochemical parameters of interest include leaf pigment concentration, photosynthetic activity, LUE and PRI; structural parameters include canopy height, leaf area index (LAI), standing biomass and canopy clumping. To obtain both biochemical and structural parameters, multispectral multiangular reflectance data are required. EO data will also be combined with ecosystem models in an assimilation scheme aimed at describing canopy productivity and hence carbon sequestration. The results of this project will contribute to understanding how terrestrial carbon fluxes can be related to EO data and, in particular, how uncertainties in the calculation of such fluxes can be reduced.

GB05/05 Geoffrey Duxbury: Airborne infrared quantum cascade laser absorption spectrometer for detection of variations of methane concentrations in the troposphere

To adapt a novel mid-infrared Quantum Cascade laser spectrometer for aircraft operation and to exploit its high sensitivity and rapid response for trace gas detection for measuring the variation of the concentration of key molecules found in the natural and polluted troposphere. In laboratory trials we have demonstrated that both nitrous oxide and methane can be detected simultaneously in the ambient atmosphere, and that ppbv variations in the methane concentration may be easily detected. Following installation and testing in the aircraft we plan to use the instrument in conjunction with Mr Andrew Wilson, Head of the Underpinning Technologies Group of CEH Monks Wood in two of his projects involving methane detection. These are the monitoring of emissions from urban traffic in Cambridge, and detecting methane hot spots associated with emissions from landfill sites.

GB05/07 Paul Bates: High spatial and temporal resolution mapping of flood dynamics for numerical model validation

Remotely sensed flood maps have been shown to be invaluable in assessing the performance of numerical model predictions of inundation extent, but previous studies have been limited by the temporal resolution of the satellite and airborne sensors used. In particular, information on the wetting phase of the floodwave is difficult to acquire due to the short rise time of many hydrographs, and hence the dynamic performance of numerical models has not been fully tested. A mission is therefore proposed to acquire a sequence of images of a dynamic flood event on a UK lowland river reach in order to test models' ability to reproduce accurately spatially and temporally variable patterns of inundation. The mission will acquire ATM data and investigate the spectral signatures of open water, dry land, waterlogged areas and emergent vegetation, and use these to derive a series of high resolution maps of flood extent through time. These will be used to compare the different performance of 1-D and 2-D numerical codes, investigate the implementation of wetting and drying algorithms in these models, and explore issues of calibration and uncertainty in fluvial hydraulic models.

GB05/08 Stuart Barr: Integrated Remote Sensing Techniques for Urban Catchment Flood Modelling

UK river catchments are experiencing increasing development which, in association with climate change, is placing many urban areas at an increased risk from flooding. In order to understand the likely damage and cost of urban flooding, as well as the influence of urban areas on channel discharge, fluvial flood models need to explicitly represent the complex topography and hydraulically important features of urban areas; information difficult to obtain by traditional survey methods. This research will employ an integrated remote sensing approach to derive urban topographic features and hydraulic parameters required for an explicit urban fluvial flood model. LIDAR, digital aerial photography and multispectral image data will be employed in an integrated framework to derive information on surface topography, 3-D features, surface type and surface texture/roughness for model parameterisation. These will be used to model known historical urban flood inundation for the river Ouseburn; for which data on previous flood events is available. The model will also be compared to models that do not include urban topography and features. The output of this research will be a better understanding of how urban areas should be managed in flood models and the utility of remote sensing for the parameterisation of these.

GB05/12 Jim McQuaid: Boundary Layer Mixing and Land Surface Properties (BLMix)

Solar heating of the planet's surface leads to heating which results in the generation of rising air currents (convection). In general convection initiates the majority of the transport and mixing in the mixed layer. Surface heating is a function of the surface properties (albedo). Small scale variability due to differing surface properties causes localised circulations which contribute to the overall mixing processes. These individual circulations are usually less than frictionally derived circulations which lead to the formation of the boundary layer. This project will use surface properties and atmospheric measurements to estimate the contributions to this motion from different land surfaces. Clearly small scale motion will enhance boundary layer motion and may play a key role in redistribution of pollutants. Modern remote sensing techniques means it is possible to map out the land surface. The D-CALM aircraft is well equipped to make such remote sensing measurements in particular with the Airborne Thematic Mapper. In conjunction with the aerial surveying by the ARSF aircraft the project will make use of the Chilbolton clear air Doppler radar during the NERC CSIP field campaign Summer 2005 which produces high resolution vertical cross sections of winds along a flight path.

GB05/13 Peter Land: Integrated Land-Sea Atmospheric Correction Experiment

The aim is to develop and validate atmospheric correction algorithms for CASI and the new hyperspectral sensor over water and land. We will measure optical and environmental characteristics of water, land and atmosphere while the aircraft passes over the same scene at several heights to evaluate the height dependence of atmospheric correction. The marine site will be the case 2 waters from the Tamar plume in Plymouth Sound, while the terrestrial site will be Plymouth Hoe. Surface measurements will include:
. surface reflectance factors, radiance and irradiance over land based calibration/validation site, and over the sea surface using a hyperspectral radiometer;
.radiance and irradiance profiles using the PML profiling radiometer;
.absorption and backscatter profiles using the PML ac9, VSF-3 and HydroScat-6;
.particulate and dissolved absorption spectra using spectrophotometry at PML and/or UoP; pigment analysis using HPLC at PML;
.suspended particulate material measurements including size distribution at PML and/or UoP;
.local aerosol optical depth using Microtops sun photometers;
.aerosol type (optical depth, size distribution, refractive index) using sun/sky photometers.
This dataset will allow detailed modelling of both in-water and atmospheric optics in support of the development and validation of atmospheric correction algorithms .In addition, the project aims to solve one of the major problems associated with application of such algorithms to images collected of the coastal zone. This issue is of great importance in the UK, given the dynamic nature of the coastal zone and the need for accurate monitoring of it. Remote sensing provides an ideal tool for monitoring change in such areas, however, an important step in this process is the ability to compare images from different dates and sites in different scenes. These comparisons require the digital counts from different scenes to be calibrated to common reference values, thus highlighting the need to establish suitable algorithms for use in this environment.The project is unique in terms of the collaborative effort between various higher education and research institutions in the South-West region.

HY05/02 Claire Fleming: Characterising Contaminated Land using Integrated Hyperspectral and Geophysical data

Industrial and mining activity in the UK has left a legacy of contaminated land and associated management challenges. Before these can be met successfully, suitable data are required that characterise the chemical and physical properties of the surface, the shallow subsurface and the ground water and pollutants transported through them. Airborne surveys provide a cost-effective, rapid, safe and non-invasive means of collecting much of the information. This proposal aims to develop the novel, integrated airborne surveying techniques by using the full range of available technologies. An airborne thermal and hyperspectral survey is proposed over an area of environmental management challenges for which airborne geophysics has just been acquired by BGS.

High-resolution airborne geophysical surveys provide detailed information on the structure, chemical and physical properties of the surface and the shallow sub-surface from electro-magnetic, magnetic and radiometric data. Hyperspectral scanning techniques can provide complimentary information on the composition of the land surface in the visible to infrared parts of the spectrum. By integrating these complimentary data types it will in future be possible to provide holistic 3D chemical and physical characterisation of the land surface and shallow subsurface. This will make it possible to define human impacts and processes in the shallow geosphere and on river catchment and aquifer systems in particular. Before this can happen, however, these two complimentary airborne technologies need to be integrated into a single survey methodology.

HY05/03 Meredith Williams: Comparison of CASI-2 and VIS-SWIR hyperspectral sensor capabilities for detection and characterisation of vegetation anomalies associated with buried gas pipelines

Comparison of CASI-2 and VIS-SWIR hyperspectral sensor capabilities for detection and characterisation of vegetation anomalies associated with buried gas pipelines

Airborne hyperspectral imaging offers a potential solution for operational monitoring of soil disturbance associated with the instatement and remediation of land adjacent to a buried gas pipeline. However, suitable methodologies have yet to be developed to reliably detect vegetation stress associated with buried pipeline soil disturbance, due to constraints imposed by the available sensors on spatial resolution, signal to noise ratio, bandwidth, and spectral range. This project will compare the performance of the new hyperspectral sensor with CASI-2 for detection and characterisation of vegetation stress associated with a buried gas pipeline in Aberdeenshire, for which CASI & ATM data were acquired in 2004.

Laboratory spectroscopy experiments from previous studies have established that VIS-NIR techniques based on the red-edge cannot reliably distinguish stress caused by gas from waterlogging effects. SWIR sensors have the potential to detect vegetation stress on the basis of absorption and reflectance features, but airborne experiments have established that the spatial resolution achievable with HyMAP may not be sufficient.

A generic methodology for detection of vegetation stress associated with buried pipeline soil disturbance will be developed, utilising the improved capabilities of the new sensor, integrating identified stress indicator absorption and reflectance features from the full VIS-SWIR range. Data processing will be carried out by an EPSRC CASE PhD student, supported by Shell International Exploration and Production (SIEP).

HY05/04 Andrew Wilson: Remote Sensing of Green-House Gases and Hydrocarbon Emissions Using an Airborne Hyperspectral SWIR Imaging Sensor

Methane and carbon dioxide are important green-house gases that are produced by a variety of natural and anthropogenic sources, many of which are amenable to mitigation or reduction strategies, if sources can be located precisely. Hyperspectral imaging spectrometry in the Shortwave Infrared (SWIR) provides a technique to derive qualitative and potentially quantitative values of column methane and carbon dioxide gas concentrations at high spatial resolution. The availability of a new full data cube hyperspectral sensor covering the 400-2500nm wavelength range will be tested over a flightline previously flown by an older, and less capable, HyMap sensor for comparative purposes, using a diverse range of sources, and at a new site flown to quantify minimum detection limits and sensitivity against a well characterised methane emission source, using ground based measurements at an abandoned coal mine.

HY05/06 Mark Cutler: Hyperspectral Remote Sensing Of Peat Physico-Chemical Properties In Glenogil, Angus: Spatial Transferability Of Pennine Results

Hyperspectral sensors offer an alternative to time-consuming wet chemistry techniques for measuring humification (degree of decomposition) of peat. SHAC results for the Dark Peak showed strong correlations between spectral indices (particularly in the SWIR) developed for HyMap and the degree of humification of exposed peat (McMorrow et al., 2002; Cutler et al., 2002; McMorrow et al., 2004a, b). This application for the new hyperspectral sensor aims to test the transferability of the relationships to another site containing exposed peat but with different properties. A related bid by McMorrow (Manchester) will test the transferability of the relationships to another date and sensor in the Pennines. Humifcation indices developed for HyMap and moisture indices developed with the ASD (McMorrow et al., 2003), will be calculated for the new sensor. Correlation, stepwise regression and artificial nueral networks will be used to investigate relationships with humification and gravimetric moisture content. The results observed at this site will be compared to those observed for different peat properties at the Dark Peak site. Extending the range of peat types will help further understand the radiation / peat properties relationship and test the usefulness of remote sensing as a tool for monitoring a range of peat types.

HY05/07 Julia McMorrow: Hyperspectral Remote Sensing Of Peat Physico-Chemical Properties In The Dark Peak: transferability of HyMap results

Hyperspectral sensors offer an alternative to time-consuming wet chemistry techniques for measuring humification (degree of decomposition). SHAC results for the Dark Peak showed strong correlations between spectral indices developed for HyMap and the degree of humification of exposed peat, measured colorimetrically as percent transmission (McMorrow et al., 2002; Cutler et al., 2002; McMorrow et al,. 2004a, b). This application for the new hyperspectral sensor, with supporting LiDAR and aerial photos, aims to test the transferability of the relationships to another date and sensor at the Dark Peak site. A related bid by Cutler (Dundee) will test spatial transferability to a Scottish site. We will use data collected for SHAC and CASI-SWIR flights, plus additional sites within resources available. Humification indices developed for HyMap, and moisture indices developed with the ASD (McMorrow et al., 2003), will be calculated for the new sensor. Correlation, stepwise regression and artificial neural networks will be used to investigate relationships with humification and gravimetric moisture content. The analysis will be repeated with ASD spectra in contact probe and field mode to investigate sensitivity to spatial resolution. The best models will be used to produce images of exposed peat humification and surface moisture content. Results will be compared with those for the Scottish site.

HY05/08 Zoe Carroll: Sub-Grid Spatial Heterogeneity of Soil Carbon Stocks in the Plynlimon Catchment

An accurate knowledge of soil carbon stores is fundamental to understanding and predicting the role of terrestrial systems in the global carbon cycle. For the UK, soil carbon is currently mapped at a relatively coarse (1 km) grid resolution, but field data clearly demonstrate a high level of heterogeneity at fine spatial scales. A new method for estimating the carbon stocks of upland organic-rich soils will therefore be developed, in which remotely sensed data will be related to an intensive ground-based survey of soil carbon in the Plynlimon catchments, mid-Wales. The project will utilise spatial high-resolution elevation (LiDAR) and test novel high spectral resolution hyperspectral data to develop fine-scale topographic and vegetation maps, across a heterogeneous area of upland, organic-rich moorland soils. The relationship between topographic position and vegetation parameters to field measurements of soil carbon content will be examined using geostatistical techniques. Geostatistical prediction will then be used to map C stocks at a previously unachievable resolution for moorland catchments.

HY05/09 Sara Long: Asssessing the dispersion of contaminants to vegetation and soil around chemical plants using indicators of plant health

Many industrial sites in the UK have had a long history of environmental pollution that is only recently being quantified through ground sampling campaigns and laboratory analyses. The sampling strategy is generally limited in nature and cannot adequately provide the spatial variation of contaminant concentrations required in estimating environmental risk or in informing mitigation and restoration strategies. The project will test the use of novel high spectral resolution hyperspectral data to predict soil and vegetation contamination through developing a geostatistical relationship between indicators of plant health and ground-based chemical analysis. In the first instance, this will be against the concentration of Polycyclic aromatic hydrocarbons (PAHs). A well characterised chemical fire, generating high volumes of PAHs, has provided the opportunity to test the methods in a semi-controlled environment, with adequate ground sampling to validate geostatistical interpolation techniques.