Methodology

1. Boundaries of the basin

Using a digital model of land and topographic map scale 1:50 000 of the INEGI, the division of the Valle de Bravo basin was defined. The surface considered is 61 561 hectares.

2. Satellite images.

Three Landsat satellite images were used on different dates with the following characteristics (table 1 and figure2)

MSS Multispectral Scanner, ETM+ Enhanced Thematic Mapper plus

Figure 2. Landsat Image MSS 1973

Satellite images corresponding to the years 1973 and 1996 come from the Eros Data Center; the information was provided by the remote sensing laboratory of the Mexican Water Technology Institute; the image from the year 2000 was provided by the Ministry of the Environment, Natural Resources and Fisheries (SEMARNAP) by the Periodic National Forestry Inventory.

To select the area of interest for the work, the basin was located and the boundaries were defined, and a cut of the images was made.

3. Definition of the classification of the vegetation and use of soil

To define the classes of vegetation and use of soil for classifying the images, the classes proposed by the National Institute of Statistics, Geography and Data Processing, 1973, were taken as the basis. The classes defined for this study are the following:

l Seasonal agriculture

l Irrigation agriculture

l Grassland

l Oak forest

l Pine-oak forest (in this class, also considered was the oak-pine forest, inasmuch as during the field work certain zones were detected with the presence of said vegetation, although in a lesser quantity)

l Pine forest

l Boreal forest (in this class, the combination of boreal and pine forest was considered)

l Cloud forest

l Bodies of water

l Urban zones (in this class was included a zone where a development is currently being constructed)

4. Field work

The field work was conducted on September 22 and 23, 2000. Two agronomists participated; they currently work in the project called "Conservation and Rehabilitation of the Valle de Bravo Basin", directed by the Mexican Water Technology Institute and Chapingo University, as well as three geographers and one biologist from the National Commission for the Knowledge and Use of Biodiversity (CONABIO). Various areas of interest were covered, in an attempt to corroborate the boundaries and identify the use of soil and predominant vegetation, obtaining 24 verification points. The field information was essential for classifying the images. The area covered can be observed in figure 3.

5. Classification

The methods used for classifying the images were: supervised classification, indexes of vegetation and visual interpretation.

For the vegetation and use of the soil, a supervised classification was used; the training field were taken from the field work, from verification points contained in the Periodic National Forestry Inventory of 1994 and maps of the INEGI.

The urban zone was defined by a visual interpretation of the image.

Due to the fact that the field work was conducted this year (2000), the image of ETM+ was classified first, due to the fact that it has the greatest time proximity, and subsequently the 1973 and 1986 images, using as a basis the classified image of 2000.

In order to improve the classification, additional information was used to redefine certain classes, such as the case of the Boreal forest; for defining the boundaries; account was taken of the altitude variable, using a digital model of the land. The Cloud forest was redefined with the use of soil and vegetation map.

Home page | Michoacán Case | Guerrero Case | Valle de Bravo Case | Naturale desasters casoe Hurricane Pauline

For doubts, questions or comments
Geographical Information Systems Sub-director