Relevant Papers
|
| The use of Remote Sensing in the Evaluation of Natural Regeneration Potential, Erosion Risk and Desertification Risk, after Forest Fires. |
D.Rokos, P.Kolokoussis Laboratory of Remote Sensing, National Technical University of Athens |
Abstract Deforestation, caused by wild forest fires, is a leading factor contributing to soil degradation. The investigation of areas with low potential for natural regeneration and high risk for soil erosion and thus of serious desertification threat, is of critical importance for reliable and effective decision making for the protection of forest environment. The capability of Remote Sensing to investigate the natural regeneration and other relevant processes and to evaluate the status of the vegetation, after forest fires, is studied in this research. Four forest areas in the prefecture of Attica (Greece), which have been repeatedly burned during the recent years, have been selected as study areas. LANDSAT TM, SPOT and ERS-1 images were acquired and field operations were carried out, for all the study areas. A significant number of special transformations of the satellite images were carried out. Statistical analysis of the brightness values of the various bands and some selected transformations of the remotely sensed images were carried out, to investigate correlations with relevant field data. This statistical analysis guided the selection of the appropriate bands and transformations which were used during the classification and reclassification procedures. The final output of this research project was the production of natural regeneration potential, risk of soil erosion, desertification risk and vegetation maps. |
The full paper (Doc file) |
| Post Fire Mapping of Typical Mediterranean Forest Areas Using Landsat TM Images. A Knowledge Based Approach for the Prediction of Forest Sustainability. |
D.Rokos, P.Kolokoussis Laboratory of Remote Sensing, National Technical University of Athens |
Abstract |
Forest fires, especially during summertime, are a major problem for Mediterranean countries and post fire mapping and monitoring of the burnt forest areas are very important for the protection of forest ecosystems. |
| The main objectives of this research were the monitoring and evaluation of natural regeneration processes as well as the prediction of forest sustainability by integrating photointerpretation and remote sensing methods and techniques with the available geographical data in a GIS environment. A knowledge based approach, involving topographic, soil and human activities data, capable of predicting the natural regeneration potential and the soil erosion risk for typical Mediterranean areas after forest fires was investigated. |
| The whole work was based on the statistical analysis of satellite and ground truth data. Statistical analysis revealed the best procedures for identifying different vegetation species and estimating natural vegetation density. |
The quality of the results of the proposed forest sustainability prediction model depends on the availability and reliability of the necessary data. Aspect, slope, soil depth and soil permeability were considered as the main input to the proposed model. The aspect and slope layers were obtained from the DEMs of the areas under investigation, while soil permeability data were digitized from the existing maps. The possibility of indirect estimation of soil depth using Landsat TM satellite imagery has also been investigated. |
A quite accurate monitoring of the regeneration processes has been achieved. According to ground truth data, the accuracy of the vegetation type identification was 86% and the accuracy of the vegetation density estimation 88% for the specific research area. |
The predictions of natural regeneration potential were proved realistic while the predictions of soil erosion risk were accurate enough only for slightly and non grazed areas. |
The poster session presentation (RTF file) |
| Evaluation of the Natural Regeneration Potential and Desertification Risk after Forest Fires in the Prefecture of Attica by GIS methods and techniques. |
|
Abstract |
The aim of the research work presented in this paper is the development of a GIS methodology for the evaluation of natural regeneration potential and of soil erosion and desertification risk, resulting from forest fires. The following four study areas: Lavrio, Pendeli, Pateras mountain and Varnavas, in the prefecture of Attica, were selected, due to their frequent forest fire events. The design and implementation of such a Geographical Information System, which will be capable to estimate through deterministic procedures, using a logical If, Then, Else shell, if the specific burned areas show enough tendency for natural regeneration potential or soil erosion risk, by taking into account: a) the specific knowledge of the factors which stengthen the regeneration procedure and those that prevent the desertification, b) techniques deriving from GIS, c) satellite remotely sensed imagery analysis and processing and d) field data (for system training and evaluation of the results), are analytically discussed. System implementation includes: digitization, storage, retrieval, management, analysis, processing and display of spatial referenced and linked thematic information, for evaluating the natural regeneration potential, the risk of soil erosion and the desertification risk, of each area. In the initial phase, the GIS consists of the following data layers: Topographic maps, digital elevation models (with slope and aspect data), soil type, soil depth and land use data before and after fires and remote sensing data, for the areas under consideration. By this GIS, we spatially analyze and relate the above layers aiming at the creation of composite data overlays, based upon empirical correlation rules supplied by forestry experts. The finally produced maps, showing the natural regeneration potential and soil erosion and desertification risk, are compared with the appropriate data of the field operations, to create reliable presuppositions for a more sophisticated design of the operations aiming to the protection of forest environment in these areas. This research project was funded by EEC (Environment Research Programme (EEC), Contract #: EV5V-0025). |
The full paper in Greek (Doc file) |
| Structuring an Integrated Information System for Development and Environment of the Metsovo Province |
|
| Abstract |
The Laboratory of Remote Sensing of the National Technical University of Athens, has planned and is accomplishing the development of an Integrated (qualitative and quantitative) Information System, concerning the data of the Physical and Socio-Economic Reality of Metsovo Province, as well as of their multidimensional relations, interdependencies, interactions and their change trends in time. |
These dynamic data form the necessary reliable, accurate and integrated infrastructure for Development planning and implementation, as well as for observation and monitoring of Earth for the protection of the Environment. Thus, the System of the Laboratory of Remote Sensing will be the necessary and reliable basis for decision making on Development and Environment issues of the Metsovo Province. |
In this System the following data have already been entered in proper digital form:
|
|
The appropriate digital geometric and radiometric preprocessing of the above satellite remote sensing imagery, as well as some specific processing of them have already been carried out and the following value added images were produced: the most appropriate color composites for natural resources inventories of the region, the classification of land use and maps of the relevant changes in time, as well as the maps of slopes and aspects of the Metsovo Province. |
| Finally, the hardware and software of the System developed by the Laboratory of Remote Sensing of the NTUA gives the possibility of data integrations, taking advantage of the appropriate combinations and correlations of all the existing data, information layers, pictures, digital elaborations and maps etc., on a unique and accurate geodetical reference system, so that to secure the best possible integration of the analogical and digital remote sensing and multimedia methods and techniques in a Geographical Information System environment. |
|
 |
National Technical University of Athens |