|
|
|
GIS for planning environmentally sustainable activities in Kulathupuzha reserve forest, Kerala, India
A major factor contributing towards degradation is the anthropogenic activities from settlers on the forest fringe. Settlers include the indigenous people, pattayam holders (non-indigenous people awarded land ownership by the government) and encroachers. Indigenous people meet all their needs from the surrounding forests and for their livelihood forage into the forest to collect forest produce which are sold in local markets either directly or through intermediaries. Settlers engage in such activities as cattle rearing, farming, clearing, etc., which in most cases are destructive to the surrounding natural forests. At times, especially in the height of summer, forest fires are not uncommon and the cause can be, very often than not, traced to their activities. Such reckless activities when combined with questionable health of many plantations result in a picture, not so colourful, of the state of one of the finest tropical forests in the country.
As a first step towards an earnest effort to monitor and strengthen the forests of Kerala for sustainable development, a combination of remote sensing and geographical information systems is being put in place for Kulathupuzha. The new system will aid in developing micro-level plans for the management of forests. It is also envisaged to bring together small area geodemographics and GIS to leverage the existing programmes and institutions. The methodology identified is briefly described below.
Remote sensing and GIS
In earlier days practical difficulty in monitoring forested areas were many and, even if one were to do so successfully it may take a while before the data gets compiled to make any meaningful decision. This was exploited by unscrupulous traders in gathering forest produce and other items from the forest in an unsustainable manner. But today, the integrated approach utilising remote sensing and GIS has the capability not only to view and gather information over a relatively large area but also to assist in taking better decisions, both in a short time.. This approach is adopted in Kulathupuzha successfully for improved data collection on ground realities using remotely sensed data from IRS-1C satellite.
Imageries from IRS-1C are read for signs of degradation within the forest domain. Zones of degradation are marked from the imagery based on canopy density. Their relationship is widely recognised to be a direct one. Higher the density lesser the degradation and vice versa (ie., lower the density closer it is towards degradation). Crown density can, thus, be directly translated to degree of degradation. One of the aspects exploited very successfully in participatory management is this relationship in an imagery whereby it is easy for the stock holders to visualise the magnitude of degradation (de Boer and Roche, 1998). The three fold zonation followed in this study is more or less similar to the standards followed by the Forest Survey of India, viz., those above 40% is taken as undisturbed, less than 10% as fully disturbed (ie., degraded) and intermediate values as moderately disturbed. Caution, however, is exercised in applying this criteria uniformly as there are areas within natural forests where the crown density could be less but of reasons unrelated to those described above. It could possibly be due to temporal variation (ie., period in which the imagery was acquired), locations of natural reed breaks or even high altitude grasslands. To avoid such misjudgements the forest types were classified into 6 resource types - grasslands, reed breaks and plantations / estates, besides the 3 forest types mentioned earlier. The forest areas are thus, in the first place, classified into 5 types while reading satellite data before attempting to categorise them into different zones. Classification is performed by identifying representative sample plots for each type in the field and their locations precisely noted using a GPS. GPS readings are, subsequently, transferred onto the imagery and their spectral signatures studied, to identified similar types elsewhere in the area.
The objective of the whole exercise being to develop management plans on a micro-level the chosen basic unit is the drainage basin, the natural delimiter. Size or resolution of each unit has been fixed at third order basin or watershed (naming convention as proposed by Horton, 1945). Forested areas categorised and demarcated on a watershed basis are vectorised and transferred to a GIS. Vectorised data on transfer carries along as attributes of its geometry the forest type and its inferred state.
The data, input into a GIS, is stored in a geodatabase (ESRI, 1999). For each watershed in the geodatabase there is a separate feature dataset. Database connection for each feature dataset holds additional data collected from the field. This database contains information on soil, slope, rainfall, regeneration, etc. Within GIS the information is collectively analysed and the degree to which each watershed is environmentally affected is mapped. For example, if forest cover is less than 10%, top soil is poor and a slope higher than moderate (say, above 20 degrees) then the area will be marked as degraded.
Environmental degradation in each watershed is tackled on the basis of established silvicultural practices and is a function of forest type, extent of degradation and the physical conditions existing in the watershed. Selection criteria for each forest type and degree of degradation is built into spatial queries and when executed in a GIS would produce a map clearly identifying which areas within each watershed needs what type of remedial measure.
|
|
|