Agricultural University of Athens
Σύστημα υποστήριξης λήψεως απόφασης για τη διαχείριση δασικών οικοσυστημάτων βασισμένο στα γεωγραφικά πληροφοριακά συστήματα και στα συστήματα γνώσης
Abstract
dc:descriptionForested Ecosystems provide a variety of goods as it is, for example, firewood, technical timber, fruits, roots of plants and recreation. They also, provide important indirect utilities, such as the protection from floods and the absorption of gaseous pollutants. Despite their particular value, Forested Ecosystems face up a lot of threats of an anthropogenic origin. Wildfires possess an important position, among these threats, especially in countries like Greece, in which hot and dry climatic conditions prevail. The reduction of the destruction rate of Forested Ecosystems can be achieved by a suitable multipurpose management planning. Nevertheless, the design and application of such a management faces obstacles, mainly due to its complexity and to the small experience from its application. In this Thesis, a Decision Support System is proposed, that aims at facilitating Forest Management Planning. Specifically, the System can provide consultation for the multiple purpose management of lowland pine forests and the reduction of wildfire destruction danger. To design and develop the proposed system, research work was undertaken that can be summarized as follows: - A general conceptual architecture of Decision Support Systems for Forest Ecosystems Management Planning was designed. This architecture is open and flexible to future implementations. - A series of suitable treatments were incorporated in the planning of multiple-purpose management, that aim at the reduction of the destruction danger from wildfires of the lowland Pine Forests and of their neighboring regions. - A Wildfire Destruction Danger Index was proposed, based on natural parameters that provide short and long term estimates, for small or large Forested areas. - A Soil Erosion Index was proposed, for the determination of the soil erosion danger. Its main characteristic is its simplicity as well as the fact that its calculation is based on data that can be gathered easily. - A Fire Simulator was developed, based on known models for wildfire simulation, whose main characteristic is the capability of detailed calculations for vegetation treatments. - A Decision Support System was developed, which is based on the general architecture that is proposed in this Thesis. The System can provide advice for the reduction of wildfire destruction danger, within the framework of the conventional multipurpose management planning. It uses three Rule-based Subsystems and the methodology of Fuzzy Logic for the handling of uncertainty. For the management and treatment of spatial data, it makes use of a Geographic Information System. The System, that was developed, has been tested in the Forest Prokopi-Drazi in Northern Evia. Its results have been compared with those of an approved Forest Management Study. The comparison has shown that the System advice, relevant to the conventional management, coincide to a very satisfactory degree with the proposals of the Forest Management Study. Moreover, the system provides advice for the preventive handling of forest vegetation, which contributes decisively to the reduction of the fire destruction danger.
Degree
thesis:*- Grantor dc:publisher
- Agricultural University of Athens
- Year dc:date
- 2008
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kaloudis, S.
- Καλούδης, Σπυρίδων
Subjects
dc:subject × 16- Συστήματα υποστήριξης αποφάσεων
- Γεωγραφικά Συστήματα Πληροφοριών (Γ.Σ.Π.)
- Οικοσυστήματα, Δασικά
- Πυρκαγιές, Δείκτης κινδύνου καταστροφών
- Πευκοδάση, Πεδινά
- Πυρκαγίες, Προσομοιωτής
- Decision support systems
- GIS
- Forest ecosystems
- Wildfire destruction danger index
- Pine forest, Lowland
- Fire simulator
- Γεωπονικές Επιστήμες και Κτηνιατρική
- Άλλες Γεωπονικές Επιστήμες
- Agricultural and Veterinary Sciences
- Other Agricultural Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/17191