3.1 Land Cover Assessment: 1992-93

3.1.1 Why NOAA-AVHRR Data?

A problem common in development planning is the inadequacy of information on the current landcover and available resource base. Without accurate information, policy makers often fail to make correct decisions. The frequent inadequacy of landcover information may be due to lack of trained personnel, equipment, or funds to collect information. Rapid changes in the resource base such as high rates of deforestation caused by increased population pressure may be one of the other reasons of deficient database.

One tool in providing current reliable land cover data is Remote Sensing: the acquisition of information without direct contact.

Although many types of remotely-sensed data of earth’s surface have been collected, NOAA AVHRR data have been used most frequently for the macro scale land cover assessments. This is because of its spectral bands that are well suited especially for the detection of important vegetative attributes. The contrast between the first two channels can be conveniently calculated to a measure of vegetation. Vegetation Index is one such ratio that has been shown to be highly correlated with vegetation parameters (plant species, leaf area, soil background reflectance and shadow) and hence, is of considerable value for vegetation discrimination.

3.2. Spectral Characterization and Land Cover Mapping

3.2.1 Classification Scheme

After interpreting the NOAA AVHRR data covering Pakistan, the following SIX major . Land cover categories were identified:

Forest areas include dry/moist temperate forests, sub-tropical forests, foothill and plateau scrub forests, tropical thorn forests, irrigated plantations, riverain forests and the coastal forests. Rangelands in fact, start from the alpine pastures in the extreme north (> 4,000 m) down to the Indus Plains (< 100 m) and Balochistan Plateaus.

An extensive area in the south-east and south of Pakistan is covered with Rolling Sand Planes and Dunes. This desert is called Cholistan in the Panjab and becomes Thar(parkar) in Sindh.

There was not much difficulty in separating most of the landcover features except for the Riverain Forests in the Panjab and Sindh, Mangrove Forests in Sindh and some degraded range/brushlands in north-western Balochistan (near the Pak-Afghan border). All these three categories of vegetation were giving the same reflectance values. Since there is an obvious degradation and reduction in the Riverain forests and Mangroves, it becomes all the more important to have a reasonably good estimate of the present state of affairs in these areas.

3.2.2 Land Use Distribution Assessment

Six major biomass types encompassing a variety of eco-systems (alpine to tropical) may be recognized (from the NOAA Imagery) for Pakistan including Northern Areas. Broadly speaking, they are lands under agriculture, rangelands (perennial and seasonal), coniferous-scrub-and other forest areas, riverain forests and finally, the mangrove forests). A variety of forests featuring varying degrees of disturbances, imbalances and degradations dominate most of the forest areas. Subsistence agriculture, fuelwood collection and grazing practices seem to be the dominant features of degradation (of natural vegetation) both quantitatively and qualitatively. Land area under snow and ice, water bodies and wastelands are some other features as derived from the NOAA imagery. Table 12.

Map 15 presents all the nine major land cover categories of Pakistan.

3.3 Natural Vegetation: a Comparison

Like many other countries, different figures are available in Pakistan for different land cover types occurring in the country. The following table (Table 12) has been added to show how the data obtained from the different sources in Pakistan differ from the NOAA classification.