Desertification Mapping Of West Asia
-A GIS and Remote Sensing Application
6. Desertification assessment and Mapping
6.1 Introduction
The problem of desertification in arid and semi arid areas has a long history through the past centuries. It has always been an overlap of long-term changes in climate and human activities. With the rising population and, therefore, growth in consumption of the very limited resources, scope and intensity of interface with ecosystems by human activities grow rapidly, leading to sever degradation of vegetation, soil and water resources, which compose the natural resources of human existence. In fact United Nations by its organizations defined the desertification as the land degradation of dry, semi dry and dry humid lands caused by human activities and natural causes. This study emphasizes on desertification processes, causes and indicators occurring in west Asia, then desertification mapping, assessment and monitoring.
6.2 General Criteria of desertification assessment
Criteria for desertification assessment must take in consideration the local conditions of the study area. These conditions should be investigated through field observations, with the help of remote sensing tools. The results of such investigation can be interpreted according to the type and degree of desertification. The types of desertification considered in this study are as follows: Vegetation cover degradation, wind erosion, water erosion and soil salinization
Criteria for the assessment and small scale mapping of desertification were developed for each land use type. The following categories of land use were identified: forest/woodland, rangeland and meadow, dry agriculture and irrigation agriculture. Because absolute figures of these characteristics vary from region to region, the criteria are given in relative figures (for example, percentage of the area covered with moving sand dunes or vegetation) (Kharin, 1999). Examples of criteria used for desertification assessment are: Plant community, percentage of climax species, decrease of total plant cover, loss of forage on rangeland, percentage of area covered with sand dunes, type of water erosion in percentage, Removal of topsoil horizon of arable land, Blow-outs of the area, loss of yield of main crop in percentage, Salinity of ground water, salinization, Salinity of irrigation water, Salt accumulation. For developing the criteria, desertification maps and experimental data collected in study area were used, as well as literature sources. We should be clear that these criteria are still more theoretical and the application is not always possible. To some degree of accuracy we could used some of these criteria, and some times we used criteria, which are related and have linear relationship with these criteria mentioned on above.
6.3 Desertification Assessment and Mapping
6.3.1 Methodology
The assessment of desertification, which we fixed in four forms: vegetation degradation, water erosion, wind erosion and salinization, needs the combination of several factors or criteria. This could be achieved only by the use of GIS tools. Generally we have a scatter information about criteria of desertification, which are distributed over the study area, and to extend these information over whole the study area we developed linear models with one or two variables. So we have generally for each criterion an image layer in which its gray level values represent different degrees of importance of the criterion. This methodology makes the combination of criteria within the GIS framework logic and easy. The final result of each desertification map has to forms one with range of 0-255 values and another one with four degrees of desertification: slight, moderate, severe and finally very severe degree of desertification, this make it easier to interpret the results. The method of mapping will be explained for each type of desertification separately.
6.3.2 Vegetation degradation mapping
For the vegetation degradation mapping we consider the two main indicators or factors: decrease of forage productivity and decrease of the percent of vegetation cover. The analysis of vegetation degradation was achieved through the following steps:
Step1: Decrease of forage productivity:
As we mentioned above there is a general decrease of production of dry matter for livestock. To convert what we have of information about the decrease of forage, we consider that this decreasing could be linked to two variables: the intensity of livestock and the vegetation index (NDVI). We defined the intensity of livestock as the number of heads per hectare of area (rangeland area). We calculated this intensity for the year of 1961 and 1999, and so we obtained two layers map of livestock intensity, then the difference of two periods give an indicator of the increasing of intensity of livestock. Here we assume that this situation reflects the intensity of overgrazing, a phenomena, which is considered to be the most important cause of vegetation degradation and consequently of desertification.
By a multiple regression calculation we obtained the following model:
F = 68.45 + 1.744 * L - 333.46 * NDVI
F: Decrease of forage in percentage
L: increase of livestock intensity
NDVI: Normalized difference vegetation index (NOAA AVHRR data 1km resolution, average of the period April 1992 through march 1993)
Using this model we calculated the first indicator image of vegetation degradation. This indicator was divided into four classes according to the criteria of desertification assessment.
Step2: Decreasing of vegetation cover percentage
The percentage of vegetation cover decrease is highly related to the Normalized Difference Vegetation Index (NDVI). In this study we used a model developed by Purevdorj (1998) to calculate the vegetation cover percent from the NDVI:
Vegetation Cover (%) = -4.337 - 3.733 * NDVI + 161.968 * NDVI2
The calculation was achieved for two periods on 1980s and 1990s. For the period of 1980s, we used an average of five years data of NOAA AVHRR 8km resolution, and for the second period the 1km NOAA AVHRR of the period of April 1992 through March 1993. Then the decrease on vegetation cover percent image was obtained by the difference of the two periods. The output image was scaled to values between 0 and 100, then divided into four classes according to criteria.
Step3: Integration of indicators
The decrease of forage productivity image and the decrease on vegetation cover percent image were combined to produce by overlapping the vegetation degradation image. This result contained 16 combinations possible, which were reduced to four classes by a decision rule from slight degree of degradation to very severe degree of degradation see Table 2 and Figure 2.
