3.0 APPLICATION
Ground water vulnerability assessment has been modelled as a multifactorial decision-making process
based on the analysis of the irrigation (distribution over the year and volumes of water used), soil
permeability, and soil infiltration.
An IGIS has been implemented to assist in the analysis of these phenomena. Its includes a list of
layers derived from cartography, thematic maps, satellite images, aerial photos, alphanumerical data.
Table 1 lists the factors considered by the experts in modeling ground water vulnerability, specifying the
features involved in the factors considered, and the IGIS layers used to quantify them.
The land use features have been derived from a land cover map obtained by means of a non-trivial
classification sub task including the classification of Landsat TM images and the definition of suitable
mapping functions. More precisely, the presence of different land covers has been investigated using a set
of 512x512 pixel multitemporal Landsat TM images recorded in 1991. The Landsat TM multitemporal
images (scene 194/28-29. of 15/04/91, 18 06/91 and 08/08/91) were classified using the maximum
likelihood classification method, producing a land cover map with six land cover classes:
class 1
rice crops
class 2
poplar groves
class 3
bare soil
class 4
grass
class 5
autumn-winter crops
class 6
spring-summer crops
The land use features, in terms of images, have been derived from the land cover classes on the basis
of the mappings shown in table 2.
The elements of the "type of irrigation " image are assigned one three possible values corresponding to
three condition (types) of irrigation. Type 1 - "not irrigated area" - refers to land cover classes 2, 3, 4. d:
type 2 - "irrigated area" - refers to land cover class 6 and type 3 - "floaded area" refers to class 1.
The elements of the "soil adjustment" image are assigned one of three possible values corresponding to
three types of soil preparation. Type 1 - "obstruction" refers to land cover class 1; type 2 - "no tillage"
refers to land cover class 2.3.4 and
type 3 - "deep tillage" refers to classes 5 and 6.
The elements of the "vegetation covering maintenance" image are classified as one of four possible
values corresponding to four conditions (types): type 1 - "annual vegetal covering maintenance" refers to
land cover classes 2 and 4: type 2 "dry season vegetation maintenance" refers to land cover classes 1.5:
and type 3 "wet season vegetation maintenance refers to class 6. and type 4 - "bare soil" refers to class 3.
The images for the topographical features, "elevation" and "slope", have been derived from the digital
terrain model generated from cartographic maps. 1:25.000 scale.
The depression zones are potentially more vulnerable because chemical elements are more likely to
accumulate there, for effect of gravity.
The slope governs the relationship between runoff and infiltration: zones with low slope are more
vulnerable.
The image used to quantify the geological feature "permeability" has been derived integrating a soil map
(scale 1:25.000) with information about physical and chemical characteristics of the soil. Water table