R. Nagarajan & Subrata Mahapatra
Center of Studies in Resources Engineering
Indian Institute of Technology, Bombay
Powai, Mumbai 400 076, INDIA
Tel: 91- 022 -576 7681 Fax; 91-022-5783190
Orbital satellite borne multi sensors generates enormous data on meteorology and land cover for analysis. It is useful in comparing the current situation with that of the past and issue warning on the forthcoming drought event if necessary. There is a need for an integrated information system having distributed database at local levels. This study highlights the land information based system developed for this purpose for drought prone Marathwada region, India.
The information requirement for decision making during drought is diverse and spatial/temporal in nature. Drought is defined as a deficiency of precipitation over an extended period of time resulting in water shortage and extensive damages to crops resulting in loss of yield. It begins with agriculture section (heavy dependence on soil moisture/water) as well as people dependent on source of water. It diminishes with agriculture sector, soil moisture/water and source of water. Its impact on society result from the interplay between a natural event (less precipitation than expected resulting from climatic variability) and demand people place on water supply. The demand for economic goods increases with increase in population and per capita consumption. Different subject perspective of drought, mitigation aspects, input data requirement and concept of land based information system is summarized in Figure 1.
Advancement in technological observation and data collection and storage and retrieval systems provide a) a measurement of abnormality of recent weather b) opportunity to place current status with the historical event perspective and c) spatial and temporal representation of historical events. Vulnerability of an area is estimated based on Biological (agriculture), Meteorological (rainfall), Hydrological (water resources) and Socio-economic (population) indicators assuming that it is a process and the said indicators reflect the severity. Hence there is an immediate need for a system to store information collected at local level on a compatible format (for vertical integration and easy either way flow) which could be used in the integration of regional, country and global level analysis/forecast
This system has been developed to store information of the past, compare it with the current data and assess the vulnerability of an area using orbital temporal remotely sensed and ancillary data in understanding the situation and decide on the mitigation planning.
Keeping the data requirement in mind, entire data sets were collected from the site. Multi date/season orbital satellite images were used in demarcation of cultivated area: fallow land, crop type and area, non-cultivable land: open area and rock outcrop areas, natural vegetation cover density: dense, moderate, low/sparse, surface water (spread/volume): reservoir/lake/pond, river, settlement: villages, infrastructure: road/railway network. Site based information were observed/monitored on meteorology- rainfall, temperature, humidity, evaporation, ground water: water table, aquifer characteristics; population: human/animal, soil moisture, infiltration characteristic of soil cover, information related to drought event and mitigation measures etc. Contour information and boundaries of forest and other restricted/protected areas were taken from thematic maps. Maps/plans were digitized, images/photographs were scanned, tables were created for point attribute data and text was created as shown in Figure 2. They were stored as Land Information base, with adequate known data security measures. Their ID number identified theme polygons. The identification mark enables the link between the spatial and attribute database. Ground photographs and images were identified by their ID address. Attribute data were logically linked by relational data base concept through associate data items. They were matched by their identification. This base could be shared between the local representatives and co-ordination units for further analysis.
A comprehensive knowledge on the prevalent parameters of the historical drought events would aid in understanding the status of current situation. Evidential reasoning approach, (Jankowski, 1995) which combines multiple pieces of evidence coming from different background knowledge and ancillary information, was used in designating the current event. The present day parametric conditions were compared with the range observed during the past. The probability of occurrence is classified as Very High, High, Moderate, Moderate to low and low. Retrieval of information could be accomplished by queries, relational operators such as union, difference, join, restriction etc. Map overlay techniques was used for change detection/assessment of crop/water surface area in estimating the crop water requirement or volume of water availability. Simple drought indices would indicate the status of onset of drought was calculated using the meteorological and other relevant data sets.
As the water availability plays an important role in drought events, natural drainage basin has been identified for analysis instead of revenue/administrative boundary. About 190 sq. km, area of Jhod nadi basin, a tributary of Godavari river in Nanded district of Maharashtra has been taken up for this study. Even though, this falls within 890-mm annual rainfall zone, frequent rainfall of 400 mm or below is reported from this area, in addition to drought events in the past. Maximum temperature of 450C and minimum of 100C is reported from here. Horizontally bedded Deccan basaltic rock is exposed here. The aquifer is confined to fractured and vesicular type. Shallow open dug wells support the drinking/partial agriculture water requirements until January/February. Drinking water schemes (water tanker) begins from March onwards. Spheroidal weathering leaves boulders on the surface cover infested with pebble/cobble. Rainfall determines the crop pattern and yield. Pulses, cereals, cotton etc are grown. Water intense crops are on the increase. Land use and land cover information derived from Indian Remote Sensing satellite data of 1989, 1991 and 1998, indicate the changes in the cultivated lands and non-cultivable lands. Total crop water requirement was calculated from the crop area. Distribution losses were also been considered in the water dependency analysis. Evaporation and other incidental loss (inclusive of percolation) were considered in the estimation of surface water availability.
Human and animal population density is on the ascend when compared between (89) 1951 and (219/sq.km) 1991 census. Water loss in the water distribution network is significant. Various drought indices (Meteorology) such as Standard Precipitation Index (SPI), Normalized Deviation of rainfall, Balm Molly Index, Aridity Index, Hydrological, Biological and Soico-economic Indicators were estimated. Based these indicators, the vulnerability of villages were designated on a map format (Figure 3) Further, quick movement of relief material to individual vulnerable villages could also be calculated.
This system provides an overall picture about drought and its information requirement for decision-makers. Further it also provides subject information for operational and strategic planning. It is very easy to use at field level by relief and administrative personnel. Further, the site based data entry could be done without affecting the normal functioning of the system.
The financial support received from Ministry of Water Resources, New Delhi, is acknowledged herewith. We acknowledge the help received from different Governmental and Non-governmental agencies.
Jankowski P (1995), Integrating geographical information system and multiple criteria decision make methods, International J. GIS, V9, pp.251-293.
Nagarajan R, (1998), Investigation of drought through remote sensing and land information system, Interim report submitted to, Indian National Committee on Hydrology, Ministry of Water Resources, New Delhi.
Note: Since the outputs of this study are available on color. Due to print restrictions they are not enclosed here.