Rosa Maria Cavalli

Remote sensing water observation for supporting Lake Victoria weed management.

This paper aims to assess the suitability of remote sensing for enhancing the management of water body resources and for providing an inexpensive way to gather, on a wide area, weed infestation extent and optical parameter linked to the water body status. Remotely sensed satellite images and ancillary ground true data were used to produce land cover maps, trough classification techniques, and water compounds maps, applying radiative transfer models. The study proposed within the framework of the cooperation between Italian Foreign Affair Ministry (through the University of Rome) and Kenyan Authorities has been carried out on the Kenyan part of the Lake Victoria. This lake is one of the largest freshwater bodies of the world where, over the last few years environmental challenges and human impact have perturbed the ecological balance affecting the biodiversity. The objective of this research study is to define the thematic products, retrievable from satellite images, like weed abundance maps and water compound concentrations. These products, if provided with an appropriate time frequency, are useful to identify the preconditions for the occurrence of hazard events like abnormal macrophyte proliferation and to develop an up-to-date decision support system devoted to an apprised territory, environment and resource management.

Experimental Approach to the Selection of the Components in the Minimum Noise Fraction

An experimental method to select the number of principal components in minimum noise fraction (MNF) is proposed to process images measured by imagery sensors onboard aircraft or satellites. The method is based on an experimental measurement by spectrometers in dark conditions from which noise structure can be estimated. To represent typical land conditions and atmospheric variability, a significative data set of synthetic noise-free images based on real Multispectral Infrared and Visible Imaging Spectrometer images is built. To this purpose, a subset of spectra is selected within some public libraries that well represent the simulated images. By coupling these synthetic images and estimated noise, the optimal number of components in MNF can be obtained. In order to have an objective (fully data driven) procedure, some criteria are proposed, and the results are validated to estimate the number of components without relying on ancillary data. The whole procedure is made computationally feasible by some simplifications that are introduced. A comparison with a state-of-the-art algorithm for estimating the optimal number of components is also made.

Hyperspectral Sensor Data Capability for Retrieving Complex Urban Land Cover in Comparison with Multispectral Data: Venice City Case Study (Italy)

This study aims at comparing the capability of different sensors to detect land cover materials within an historical urban center. The main objective is to evaluate the added value of hyperspectral sensors in mapping a complex urban context. In this study we used: (a) the ALI and Hyperion satellite data, (b) the LANDSAT ETM+ satellite data, (c) MIVIS airborne data and (d) the high spatial resolution IKONOS imagery as reference. The Venice city center shows a complex urban land cover and therefore was chosen for testing the spectral and spatial characteristics of different sensors in mapping the urban tissue. For this purpose, an object-oriented approach and different common classification methods were used. Moreover, spectra of the main anthropogenic surfaces (i.e. roofing and paving materials) were collected during the field campaigns conducted on the study area. They were exploited for applying band-depth and sub-pixel analyses to subsets of Hyperion and MIVIS hyperspectral imagery. The results show that satellite data with a 30m spatial resolution (ALI, LANDSAT ETM+ and HYPERION) are able to identify only the main urban land cover materials.

Aerosol optical retrieval and surface reflectance from airborne remote sensing data over land.

Quantitative analysis of atmospheric optical properties and surface reflectance can be performed by applying radiative transfer theory in the Atmosphere-Earth coupled system, for the atmospheric correction of hyperspectral remote sensing data. This paper describes a new physically-based algorithm to retrieve the aerosol optical thickness at 550nm (τ550) and the surface reflectance (ρ) from airborne acquired data in the atmospheric window of the Visible and Near-Infrared (VNIR) range. The algorithm is realized in two modules. Module A retrieves τ550 with a minimization algorithm, then Module B retrieves the surface reflectance ρ for each pixel of the image. The method was tested on five remote sensing images acquired by an airborne sensor under different geometric conditions to evaluate the reliability of the method. The results, τ550 and ρ, retrieved from each image were validated with field data contemporaneously acquired by a sun-sky radiometer and a spectroradiometer, respectively. Good correlation index, r, and low root mean square deviations, RMSD, were obtained for the τ550 retrieved by Module A (r2 = 0.75, RMSD = 0.08) and the ρ retrieved by Module B (r2 ≤ 0.9, RMSD ≤ 0.003). Overall, the results are encouraging, indicating that the method is reliable for optical atmospheric studies and the atmospheric correction of airborne hyperspectral images. The method does not require additional at-ground measurements about at-ground reflectance of the reference pixel and aerosol optical thickness.

Use of airborne hyperspectral images to assess the spatial distribution of oil spilled during the Trecate blow-out (Northern Italy)

On February 1994 a large area close to Trecate was affected by an oil blow-out from an AGIP rig located within the Ticino Regional Park. One month later an airborne survey has been carried out in the framework of the CNR Lara Project, by utilizing the Daedalus AA5000 MIVIS spectrometer with 102 channels from visible to thermal infrared. Different authors stress, for oil slicks discrimination, the utility of laser and microwaves based techniques, but the high spatial and spectral MIVIS resolutions can improve the detection of the relative coverage by spilled oil. This task has been performed by applying hyperspectral unmixing methods to the MIVIS calibrated data, obtaining an oil fractional image with respect to other chosen end-members. The analysis has shown a good agreement between the results of the unconstrained unmixing technique applied to MIVIS data and the ground truths, offering a tool useful to quantify in a synoptic overview the effects of oil spills over land, by relating the ppm of oil with the oil hyperspectral information gathered by MIVIS.

Optimal spectral domain selection for maximizing archaeological signatures: Italy case studies.

Different landscape elements, including archaeological remains, can be automatically classified when their spectral characteristics are different, but major difficulties occur when extracting and classifying archaeological spectral features, as archaeological remains do not have unique shape or spectral characteristics. The spectral anomaly characteristics due to buried remains depend strongly on vegetation cover and/or soil types, which can make feature extraction more complicated. For crop areas, such as the test sites selected for this study, soil and moisture changes within near-surface archaeological deposits can influence surface vegetation patterns creating spectral anomalies of various kinds. In this context, this paper analyzes the usefulness of hyperspectral imagery, in the 0.4 to 12.8 μm spectral region, to identify the optimal spectral range for archaeological prospection as a function of the dominant land cover. MIVIS airborne hyperspectral imagery acquired in five different archaeological areas located in Italy has been used. Within these archaeological areas, 97 test sites with homogenous land cover and characterized by a statistically significant number of pixels related to the buried remains have been selected. The archaeological detection potential for all MIVIS bands has been assessed by applying a Separability Index on each spectral anomaly-background system of the test sites. A scatterplot analysis of the SI values vs. the dominant land cover fractional abundances, as retrieved by spectral mixture analysis, was performed to derive the optimal spectral ranges maximizing the archaeological detection. This work demonstrates that whenever we know the dominant land cover fractional abundances in archaeological sites, we can a priori select the optimal spectral range to improve the efficiency of archaeological observations performed by remote sensing data.

A Methodology to Assess the Accuracy with which Remote Data Characterize a Specific Surface, as a Function of Full Width at Half Maximum (FWHM): Application to Three Italian Coastal Waters

This methodology assesses the accuracy with which remote data characterizes a surface, as a function of Full Width at Half Maximum (FWHM). The purpose is to identify the best remote data that improves the characterization of a surface, evaluating the number of bands in the spectral range. The first step creates an accurate dataset of remote simulated data, using in situ hyperspectral reflectances. The second step evaluates the capability of remote simulated data to characterize this surface. The spectral similarity measurements, which are obtained using classifiers, provide this capability. The third step examines the precision of this capability. The assumption is that in situ hyperspectral reflectances are considered the “real” reflectances. They are resized with the same spectral range of the remote data. The spectral similarity measurements which are obtained from “real” resized reflectances, are considered “real” measurements. Therefore, the quantity and magnitude of “errors” (i.e., differences between spectral similarity measurements obtained from “real” resized reflectances and from remote data) provide the accuracy as a function of FWHM. This methodology was applied to evaluate the accuracy with which CHRIS-mode1, CHRIS-mode2, Landsat5-TM, MIVIS and PRISMA data characterize three coastal waters. Their mean values of uncertainty are 1.59%, 3.79%, 7.75%, 3.15% and 1.18%, respectively.

Suitability of CASI and ATM airborne remote sensing data for archaeological subsurface structure detection under different land cover: the Arpi case study (Italy)

In this paper multi-sensor airborne remote sensing has been applied to the Arpi archaeological area of southern Italy to assess its suitability for detecting and locating subsurface archaeological structures and to delineate subsurface remains beyond the current limits of ground geophysical data. To this aim, the capability of CASI and ATM reflectances in the VIS–NIR spectral range and the ATM apparent thermal inertia for subsurface archaeological prospection have been assessed at different sites of the Arpi archaeological area. First, linear spectral mixture analysis has been applied to CASI and ATM images to retrieve the dominant land cover for the selected subsurface structures, and then, the spectral bands most effective for the archaeological buried structure detection as a function of the land cover characteristics have been evaluated. The results reveal that multi/hyperspectral airborne remote sensing data can represent an effective and rapid tool to detect subsurface structures within different land cover contexts. Therefore, the proposed methodology can be used to perform a preliminary analysis of those areas where large cultural heritage assets occur by prioritizing and localizing the sites where to apply archaeological prospection.

Hyperspectral remote sensing data to map hazardous materials in a rural and industrial district: The Podgorica dwellings case studies

In this paper, we present the results of a hyperspectral airborne and in situ campaign in Montenegro aimed at individuating and monitoring two hazardous materials. They are the residues of the bauxite processing, i.e. red mud, and the asbestos fibers applied in the building materials. We perform laboratory analyses of asbestoscement, red mud and soil samples collected in the study area for (a) recognizing the dominant minerals using XRay Diffraction and X-Ray Fluorescence; (b) identifying the optical characteristics of the samples using a portable field spectrometer; and (c) characterizing their spectral features and remote sensing detection requirements. A least-squares fitting procedure, on the basis of the significant red mud and asbestos-cement reflectance spectral features, was applied to airborne hyperspectral remote sensing data collected over the study area. Results show that hyperspectral remote sensing data can provide an efficient, fast and repeatable tool for mapping and monitoring the diffusion of pollutants providing the location of the hazardous areas to be checked.

Red mud soil contamination near an urban settlement analyzed by airborne hyperspectral remote sensing

The red mud dust risk involves the accumulative contamination of land and dwellings in the community with highly alkaline fine particulate containing heavy metals and other pollutants. This paper demonstrates that hyperspectral airborne remote sensing data can provide an effective, rapid and repeatable tool for mapping and monitoring the spread of red dust providing the location of the polluted areas to be checked. We perform field and laboratory analyses of red mud and soil samples collected in the study area and identify the optical characteristics of the samples to characterize the red mud spectral features. Next, we use hyperspectral airborne data covering an aluminium processing plant in Montenegro (EU). The joint use of MIVIS reflectance and emissivities data allowed us to individuate and map those sites on which the red dust is spread by the dominant winds, where a check for reclamation or a neutralization intervention is required.