Dimitrios A. Vaiopoulos

Combined use of remote sensing, GIS and GPS data for landslide mapping

The term landslide includes a wide range of ground movement, such as slides, falls, flows etc. mainly based on gravity with the aid of many conditioning and triggering factors. Particularly in the last two decades, there is an increasing international interest on the landslide susceptibility, hazard or risk assessments. In this paper a combined use of Remote Sensing, GIS and GPS data is presented. The area of study is Lefkas Island in the Ionian Sea, Greece. The Ionian Sea presents a very big seismicity and the Landslides phenomena are very often. During the last twenty-five years more than 20 big landslides have been recorded on Lefkas Island. The latest landslides have been recorded in August 2003 as a consequence of a 6.1R earthquake. Multispectral and multitemporal satellite data were used for landslide detection. More especially we have used Landsat MSS, Landsat TM & ETM data and Terra ASTER data. Because most of the landslides are mostly bare of vegetation they present high reflectance. As a result big landslides can be detect from satellite data. ASTER stereo-pairs have been used for the creation of two DTM’s. The images cover an area of 60X60 km. The images have been received in the near infrared (0,78-0,86μm) part of the spectrum with a spatial resolution of 15m. The first stereo-pair was acquired on January 2000 and the second one on August 2003 a few days after the earthquake and the landslides. The same GCP’s were used for the creation of the two DTM’s. In some cases the landslides can be detected as a difference of the two DTM’s. All the landslides have been mapped in-situ using a GPS receiver. Finally all the above data have been integrated in a GIS and detailed landslides maps have been produced. Keywords-component: Landslides, satellite data, DTM, GPS stero-pairs, GIS.

1-D and 2-D spatial frequency analysis of self-potential field anomalies, produced by a polarized sphere

This paper gives analytical expressions for the 1-D and 2-D frequency spectra of the self-potential field produced by a polarized sphere. In 1-D, the amplitude spectrum of the potential field leads to a criterion for determination of the depthh to the centre of the sphere. The polarization angle of the buried sphere can be calculated from the maximum point of the amplitude spectrum of the electric field. In 2-D, the depth to the centre of the polarized sphere can be calculated if the polarization is vertical.

EO-1 Hyperion and ALI bands simulation to Landat 7 ETM+ bands for mineral mapping in Milos Island

During the last two decades, airborne hyperspectral sensors such as the AVIRIS or DAIS have been proved very useful but quite expensive tool for the detection and mapping of earth surface minerals. On November 2000 the launch of the Earth Observing 1 (EO-1) satellite, which included Hyperion, the first spaceborne imaging spectrometer, provided a new low cost tool in remote sensing research. This study evaluates hyperspectral data from Hyperion, as well as multispectral data from the EO-1 Advanced Land Imager (ALI) and the Landsat 7 ETM+ for mineral mapping in Milos Island. The three sensors examined in this study have similar spatial resolution, totally different spectral resolution and radiometric quality characteristics. All the data were collected the same day within one-minute time. As a result the atmospheric conditions were exactly the same and that make the data ideal for comparison. The performance of the EO-1 Hyperion imaging spectrometer with the Advanced Land Imager (ALI) and the Landsat 7 ETM+ sensor was compared using a method that aggregated portions of the Hyperion 10 nm bands to simulate the broader multispectral bands of ALI and ETM+. The general process was to calculate a weighted sum of the Hyperion bands that covered each Landsat band. The weights used in the sum were derived, by comparing the spectral response of the hyperspectral bands with the respective multispectral band. Different band ratios like the TM3/TM1 sensitive on the iron oxide detection, or different combinations sensitive on mineral (TM5/7, TM5/4, TM3/1) or hydrothermal anomalies (TM5/7, TM3/1, TM4/3) detection were used for the comparison of the three data sets and the results are presented in this study.

Testing the performance of the MNDVI vegetation index

The Modified Normalizes Differences Vegetation Index is defined by: MNDVI = (c.NIR-Red)/(c.NIR+Red). c is a real number, which generally takes values between 0.1 and 10. NIR and Red are the reflectances at the Near Infrared and Red channels, respectively. In the present paper the performance of the MNDVI vegetation index is studied, using an ALOS image over a burnt forest area of Greece. For each produced MNDVI image, the statistical parameters of the histogram (standard deviation and entropy), the semivariogram and the frequency spectrum are calculated. It is observed that the entropy and standard deviation present a peak at a characteristic c value which depends on the statistical parameters of the NIR and Red channels. The semivariogram also changes with c and presents the most rapid increasing tendency with distance at the same characteristic c value. Therefore, changing c in the MNDVI produces images with different tonality contrasts and spatial variations, which may help the potential user to broaden the spectrum of the available vegetation index images and detect targets of interest.

EO-1 Hyperion and ALI bands simulation to Landat 7 ETM+ bands and comparison

Numerous satellite sensor systems have been launched during the last twenty years and satellite data are increasingly being used in regional or global vegetation monitoring. The observation of global vegetation from multiple satellites requires much effort to ensure continuity and compatibility due to differences in sensor characteristics and product generation algorithms. More recently the launch of hyperspectral sensor like Hyperion make the compatibility problem even more difficult as the very narrow hyperspectral bands need to be simulated to the broader multispectral bands before proceed to any further comparison. In this study we tried to compare multispectral (Landsat ETM+ and EO-1 Advanced Land Imager) data with hyperspectral (Hyperion) data for the vegetation cover mapping of Milos Island. All the data were collected the same day within one-minute time. As a result the atmospheric conditions were exactly the same and that make the data ideal for comparison. The performance of the EO-1 Hyperion imaging spectrometer with the Advanced Land Imager (ALI) and the Landsat 7 ETM+ sensor was compared using a method that aggregated portions of the Hyperion 10 nm bands to simulate the broader multispectral bands of ALI and ETM+. The general process was to calculate a weighted sum of the Hyperion bands that covered each Landsat band. The weights used in the sum were derived, by comparing the spectral response of the hyperspectral bands with the respective multispectral band. The Normalized Difference Vegetation Index was used for the comparison of the three data sets and the results are presented in this study.

Use of vegetation indexes and PCA method with remote sensing data for the classification of burnt areas according to how many times they have been burnt

The objective of this study was to process multitemporal satellite data in order to detect burnt areas and classify these areas according to how many times they have been burnt. The area of study is situated in Western Peloponnese near the site of Ancient Olympia. In 1986, 1998 and 2000 three big fires have burnt more than 500.000.000 m2 of forest and rural land in the broader area. In order to detect the vegetation changes and classify the burnt areas for the period 1984-1999 we used the following multitemporal satellite images: A Landsat 5 TM cloud free subscene, acquired on July 27 1984, A Landsat 5 TM cloud free subscene, acquired on September 18 1986, A Landsat 7 ETM cloud free subscene, acquired on July 28 1999, We applied the NDVI (Normalized Difference Vegetation Index) to all the satellite images. Then, we created two new images with two bands each. one using the vegetation indexes images of 1986 and 1984 and a second one using the vegetation indexes images of 1999 and 1986. Then, we applied the PCA method to the new images. After the fires of 1986, 1998 and 2000 local authorities have mapped the burnt areas using traditional methods. With joint use of the thematic maps and the above produced images of Principal Components we managed to classify the burnt areas according to how many times the have been burnt. The general conclusion is that we can use satellite data with the vegetation indexes PCA method for the accurate mapping of burnt areas and the vegetation monitoring. Burnt areas for more than twice cannot be regenerated on its own so the classification of the burnt areas according to how many times they have been burnt is very important in order to locate the areas that needs reforestation.

Ameliorating the spatial resolution of Hyperion hyperspectral data

In this study seven fusion techniques and more especially the Ehlers, Gram-Schmidt, High Pass Filter, Local Mean Matching (LMM), Local Mean and Variance Matching (LMVM), Pansharp and PCA, were used for the fusion of Hyperion hyperspectral data with ALI panchromatic data. Both sensors are onboard on EO-1 satellite and the data are collected simultaneously. The panchromatic data has a spatial resolution of 10m while the hyperspectral data has a spatial resolution of 30m. All the fusion techniques are designed for use with classical multispectral data. Thus, it is quite interesting to investigate the assessment of the common used fusion algorithms with the hyperspectral data. The area of study is the broader area of North Western Athens near to Thrakomakedones village.

The role of the correlation coefficient on the statistical behavior of a simple band ratio

The subject of the present paper is how are the statistical parameters of an image of a simple band ratio are influenced, in quantitative terms, by the correlation coefficient between the two bands. The possibility to modify the expression for the ratio of highly correlated bands, in order to enhance the tonality contrast between the target of interest and the adjacent pixels, is also studied. Using proper bivariate Gaussian distributions it is shown that the standard deviation and the entropy of the image decrease considerably, as well as the correlation coefficient increases. When the correlation coefficient is high, it is possible to use the square of the simple ratio, in order to improve the contrast of the image. The results and conclusions of this paper may be useful in mapping the vegetation cover or detecting lithological units and mineralization zones with the aid of remote sensing technology.

Monitoring of the environmental changes at Lefkas Island using remote sensing data and GIS techniques

Lefkas Island is situated in the Ionian Sea, Greece. The Ionian Sea presents a very big seismicity and the landslides phenomena are very often. As a result many landslides have been recorded on Lefkas Island during the last twenty-five years. Also the island suffers from summer forest fires and large rural areas have been burnt. In this paper a combined use of multitemporal and multisensor Remote Sensing data and GIS techniques for the environmental monitoring of Lefkas Island is presented. Multisensor and multitemporal satellite data were used for landslide detection and burnt area detection. The satellite data used cover the period from 1977 to 2003. More especially we have used: A Landsat MSS scene of 1977, A Lndsat TM scene of 1986, A Landsat TM scene of 1989, A Landsat TM scene of 2000, A Landsat ETM scene of 2000, An Aster Vnir scene of 2000, An Aster Vnir scene of 2003. All the images have been orthorectified and resampled to 30m pixel size. Then using different band ratios we have managed to locate the burnt areas and the areas damaged by landslides. All the results have been verified by in situ measurements using a GPS receiver. The classification results from the satellite data, the in situ measurements and all the necessary maps (topographic geological etc.) have been integrated in a GIS database.

A study of the performance of a new sharpening filter

In the present paper is studied the effect of a recently proposed filter on satellite images. In spatial frequency domain, the response of the filter is controlled by the parameters b and k, which are positive real numbers. 5 x 5 convolution masks at the image are constructed, which simulate the response of the filter at frequency domain, for various b and k values. In order to study the performance of the filter in qualitative and quantitative terms, these masks were applied on various satellite images, which were taken over urban centers, as well as broader regions with different land cover types and geomorphological features. For each filtered image, there were computed the standard deviation and the signal to noise ratio. The statistical analysis showed that for k between 0.5 and 0.7 and for b between 0.8 and 1.2, one can produce images with considerably enhanced tonalities between adjacent pixels. Experimentation with satellite images showed that convolution masks with these k and b parameters produce images where lineaments such as coastal lines, drainage or road network can be clearly seen. The potential user is encouraged to try on various k and b values, in order to obtain the optimum result for the area under study.