Bioresource Inventory using Remote Sensing for Regional Energy Planning
Ramachandra, T.V., Vijaya Prasad, B.K. and Samapika Padhy
Centre for Ecological Sciences
Indian Institute of Science
Bangalore 560 012, India
Address for Communication:
Dr. Ramachandra, T.V.
Energy Research Group [CES]
Centre for Ecological Sciences
Indian Institute of Science
Bangalore 560 012, India
E Mail:
Telefax: 91 - 080 - 3315428 / 3342085 / 3341683 [CES-TVR]
Telephone: 91 - 080 - 3340985 / 309 2506 / 3340985
Bioresource Inventory using Remote Sensing for Regional Energy Planning
Ramachandra, T.V., Vijaya Prasad, B.K. and Samapika Padhy
Centre for Ecological Sciences
Indian Institute of Science
Bangalore 560 012, India
Abstract:
Energy is the fundamental tool to attain minimum quality of life. It is also an indicator of the economic status of a region. One of the main concerns today is to meet the ever-growing energy demand in an environmentally sustainable manner. Bioenergy continues to contribute significantly to the total energy consumption. It plays a critical role in sectors such as domestic and rural industries. In this context it is necessary that the regional planning exercises formulate policies to develop sustainable bioenergy systems consistent with the objectives of ecodevelopment and environmental conservation. However, lack of adequate relevant information of different bioenergy resources in a regional planning framework, hampers effort to develop alternatives to achieve multiple goals set by environmental objectives and the energy demand on the resource. Hence, for sustainable development, there is a need to determine the inventory of demand and resource situation in a region. In this regard a study has been carried out in Kolar district to explore an environmentally sustainable bioenergy strategy by developing bioresource supply and consumption database. Our study shows that the bioenergy constitute 82-85% of the district’s total energy consumption. This paper provides insight to the bioresource inventory at district level, using spatial and temporal tools and implications on policy structures.
1.0Introduction:
1
Bioenergy continue to contribute significantly to the total energy consumption. In sectors such as domestic and rural industries sector, they play a critical role. In this context it is necessary that the regional planning exercises formulate policies to develop sustainable bioenergy systems consistent with the objectives of ecodevelopment and environmental conservation. However, lack of adequate relevant information on a different bioenergy resources in a regional planning framework hamper efforts to develop alternatives to achieve multiple goals set by environmental objectives and the energy demand on the resource. This paper provides insight to the bioresource inventory using spatial and temporal tools and implications on policy structures are examined.
Detailed planning would be required from National, to State, to District, to Taluk and village levels. The inappropriate selection and site matching of species or management strategies can have adverse effects and lead to degradation and abandonment of land. However, the correct selection of plant species can allow the economic production of energy crops in areas previously capable of only low plant productivities. Simultaneously multiple benefits may accrue to the environment. Such selection strategies allow synergistic increases in food crop yield and decreased fertiliser applications while providing the local source of energy and employment.
Bioresources play a dominant role in the energy balance of various states in India, and shortages of bioenergy exist in many regions. One of the important databases needed for management of forests , regional energy planning, etc. is the total quantity of biomass per unit area referred to as biomass density. This also aides in estimating emissions of carbon dioxide resulting from changes in the vegetation cover. Bioresource inventories have shown to be valuable sources of data for estimating biomass density, but inventories for any states in India are few in number and poor in their quality. This lack of reliable data has been overcome by use of a promising approaches that produces geographically referenced, estimates by modelling in a geographic information system. (GIS) and remotely sensed data. This approach used to produce geographically referenced, spatial distributions of potential and actual above ground biomass density in Kolar district.
Multi temporal satellite imagery (LISS III) with GIS would help in developing agro ecological zones in the district - as the first stage of a multi level sampling frame for estimating available bioresources. Use of multi stage sampling design enables a regional inventory of bio resources inventory quickly and effectively. Automated selection of sampling units from digitally classified satellite imagery are proved very efficient, and the methodology for deriving sampling expansion factors makes the result highly robust with respect land cover classification accuracy. The regional agro ecological zones provides a consistent sampling frame to identify the bioresource status (surplus / deficit). The methodology could be successfully applied to other regions / states.
1
The resource base in Kolar under each sector such as forests, agriculture, horticulture and animal residues are analysed spatially.
2.0 Objectives
1. To estimate the bioresources available in the arid (Kolar) zone of Karnataka State, India. The resource base in a region under each sector such as forests, agriculture, horticulture and animal residues would be analysed spatially. the estimates of woody biomass derived from field data includes -
a) Estimates of biomass productivity, above ground biomass - forest inventory,
b) Bioresidues available from agriculture and horticulture sectors,
2. Annual increment and regeneration potential of the bioresources within the region and
3.To assess the bioresource status in the Kolar district
3.0 Study Area: Kolar District
The Kolar District is located in the southern plains region of the Karnataka State, India. It lies between 770 21' to 780 35' east longitude and 120 46' to 130 58' north latitude and extends over an area of 8,225 sq.kms. The population was 22.17 lakhs in 1991 (as per census report). For administrative purposes the District has been divided into 11 Taluks. There are 15 towns and 3,325 inhabited villages in the District.
Kolar belongs to the semi arid zone of Karnataka. In the semi arid zone apart from the year to year fluctuations in the total seasonal rainfall, there are also large variations in the time of commencement of rainfall adequate for sowing as well as in the distribution of drought periods within the crop growing season. Kolar district depends upon the distribution of rainfall during the southwest and northeast monsoon seasons. Out of about 280 thousand hectares of land under cultivation 35% is under well and tank irrigations. There are about 951 big tanks and 2934 small tanks in the district. The average population density of the district is 2.09 persons/hect (rural) and 2.69 persons/hect (rural+urban). The population density ranges from 1.44 (Bagepalli), 1.69 (Gudibanda), 1.70 (Srinivaspura) to the maximum of 2.55 (Kolar). While, the population density in taluks lies within this range - Bangarapet (2.52), Malur (2.38), Gauribidanur (2.36), Sidlaghatta (2.16), Chintamani (2.10), Mulbagal (2.04), Chikkaballapur (1.92). Fig 1a, depicts talukwise population and livestock densities.
1
4.0 Methodology
An integrated approach involving compilation of data from government agencies and institutions, application of spatial and temporal analyses using remote sensing data, Geographic Information System (GIS technique) and conventional field survey (ground truthing) have been adopted in this study (Fig 1b). The vegetation map for Kolar district (talukwise) were prepared based on the interpretation of IRS-1C satellite imageries, using the visual interpretation keys such as tone, colour, texture, pattern, association, size, shape, topography and drainage. This involves
Data acquisition, loading, composite generation and georeferencing
Training sites, Ground truth collection (field data collection).
Signature generation for classification
Demarcation of boundaries and transfer of administrative features
Extraction of statistics
Taluk and district boundaries, and road network were digitised from SOI toposheets.
Georeferencing has been done by extracting GCPS from topographical map and using GPS. Multispectral classifications was carried out using soft classifiers (based on Bayesian probability theory).
5.0 Biomass Resources Assessment
Non-availability of accurate, reliable and up-to-date data for various biomass resources is the main reason hindering an accurate assessment of the bioresources of a region. Such data need to be obtained and updated periodically. Surveying, sampling and analytical procedures is used for the collection of this data.
5.1Wood Biomass
-Average biomass content per unit area in different forests, forest reserves, plantations, woodland transitions in various climatic zones;
-Average biomass being removed and added in above areas on annual basis in various climatic zones;
-Wood supply and demand for rural and urban areas for all States of the Federation for various applications of wood;
-Energy value of various types of wood available and its present mode of utilization as fuel.
1
5.2 Forrage Grasses and Shrubs
-Average biomass content per unit area in wooded shrub grasslands, shrublands, grassland/shrubland transitions and grasslands, annual harvestable and renewable percentages and methods, energy values, and present utilization.
5.3Residues and Wastes
Crop Residues and Wastes
-Total annual crop production of food and cash crops and their hectarage for various climatic zones.
-Average crop residues and wastes per unit area for each of food and cash crop in different climatic zones, their proportion to net grain yields, harvestable percentages and methods, energy values, and present utilization.
Animal Wastes
-Total number of animals and their categories in different climatic zones/States.
-Average amount and type of feed, and waste production per day for each type of animal, waste collectable percentages and methods, energy values, and present utilization.
The resource base in a region under each sector such as forests, agriculture, horticulture and animal residues is analysed spatially.
5.4 GIS in Bioresources Assessment and Monitoring
The acquisition of basic inventory data is fundamental in the regional energy planning endeavour. Data include vegetation type, soil type, species type, class/stand structure, canopy details, density and the boundaries of management units. Data collection techniques range from selecting sample plots (quadrats or transects) for ground surveys to using topographic maps, and emerging Global Positioning Systems (GPS), alongwith Remotely Sensed data and Geographic Information System (GIS) make direct and substantial contributions. Geographic Information System can contribute to assessment of bioresoorce availability, demand and offers the potential to predict future needs.
Spatial data input, editing, maps creation, overlaying, reclassification and suitability analyses characterize the inventorying, monitoring and decision making process. Resource assessment include
1
1. inventorying bioresources available for fuel, food and fodder from various categories of land cover,
2. related data such as topography, soils, roads and hydrology and
3. assessment of bioresource productivity (from forests, agriculture, horticulture, etc.).
In addition to remote sensing, spatial positioning technologies have begun to influence surveying techniques and, thus resource inventories. Global Positioning System (GPS) technology is based on a set of orbiting satellites (a total of 24), which provides three dimensional positional fixes with an accuracy within tens of meters. With four or more satellites in view, a GPS receiver can interpret the carefully timed satellite signals to determine geometrically the latitude, longitude and altitude at the operators position. GIS applications of GPS include georeferencing of satellite imagery and navigating to sample sites for ground truth exercises particularly relevant for forest and plantation inventories.
5.5 IRS-1C LISS -III Data for Bioresource Assessment
IRS-1C with 23.5 m spatial resolution provide data outputs adequate or comparable to the scale of 1:50,000. Eleven taluks in Kolar district has been selected using March 1998 data and was analysed using soft classifiers (based on Bayesian probability theory). Bioresource is estimated using yield data for each vegetation type in the inventory: Forests, Agriculture, Horticulture, Shrub land, etc. Yields were multiplied by spatial coverage (area) for each land use category. Talukwise residues from livestock were computed from population and dung yield data for each type of animal.
5.6 Interpretation of Remotely Sensed Data for Land Use / Land Cover
Sensor records response based on many characteristics of land surface, including natural and artificial cover. Usually the elements of tone, texture, pattern, shape, size, site and association are used to derive information about land use / cover mapping.
5.7Sampling Frame
1
An important initial objective is to develop the agro-ecological zonation to provide a valid basis for extrapolating the results of the supply survey to the regional level. The woody biomass and agricultural residues surveys requires a regional zonation which reflected the range of natural vegetation as well as agricultural land use. A suitable zonation is also required to provide a valid sampling frame to spatially link the results of the supply and demand surveys.
The use of satellite imagery enables actual land cover classes to be mapped at a regional level, which is a more preferable approach to developing a valid and robust sampling frame. There are well-established methodologies for developing land cover zonation at national scales by using multi-temporal imagery to distinguish between patterns of vegetation activity with time. The imageries, with Geographic Information System and combined with ancillary data on rainfall, topography, climate, and the extent of irrigated farmland to produce a zonation of land cover types for the whole Region.
Sampling units was selected within each IRS scene for field measurement of woody biomass and crop residues. Acquisition dates depends on cloud free period for both woody vegetation and crop sampling. This required a trade-off between the optimal season for classifying woody vegetation (June-July)_and for classifying crops at their stage of maximum greenness (March-April for the spring or rabi harvest and September for the autumn or kharif harvest).
5.8 Selection of Sampling units for Measurement of Woody Biomass
Sampling units for field measurements of biomass fuels would fit within each scene. Both imagery, topographic maps, GPS were used the field work.
Digital classifications of vegetation cover from the LISS be used as a second-level sampling frame for drawing field samples for the woody biomass survey. For a selection of primary sampling units for field work, to ensure robustness with respect to any variation in classification accuracy between images. The following approaches were used.
1.The agro-ecological zonation provide a sound sampling frame at the regional level. The zonation provides a basis for introducing consistency between scenes, in that the proportion of vegetation cover classes falling within each zone would be typical of that zone.
2.For the woody biomass survey, ground truthing was carried out to confirm whether vegetation cover classes derived from the imagery contained significant woody biomass resources. This included cultivated farmland.
1
6.0 Literature Review - Present Role of Bioenergy:
Bio energy is one of the primary sources of fuel in our country. Recent study by us on energy utilisation in Karnataka considering all types of energy sources and sectorwise consumption reveals that traditional fuels such as firewood (7.440 million tonnes of oil equivalent - 43.62%), agro residues (1.510 million tonnes of oil equivalent - 8.85%), biogas, cowdung (0.250 million tonnes of oil equivalent-1.47%) accounts for 53.20% of total energy consumption in Karnataka [Ramachandra, T.V. and Subramanian, D.K., 1997]. In rural areas the dependence on bio energy to meet the domestic requirements such as cooking and water heating purposes is as high as 80-85%. Fuelwood and agricultural residues are also widely used as fuel in rural industries such as cashew processing and other agro processing industries, brick kilns, and in commercial sectors such as hotels etc. Investigations of energy consumption in few selected villages of Kolar Taluk reveal that per capita fuel wood consumption for domestic purposes such as cooking, water heating etc., are in the range 1.3 to 2.5 kg/person/day [Pramod Dabrase, et.al. 1997.].
In developing countries such as India there is seen to be a large difference in the Energy Consumption Patterns in the Urban and Rural areas. According to a survey carried out in 1963-64 and 1973-74 (by National Sample Survey), the average per capita consumption of energy has not changed significantly during this period and it also indicated that the per capita consumption of energy in Rural areas is more than that in Urban areas, which is mainly due to the relatively low efficiency of traditional(energy) devices and the availability of free fuel. These surveys also indicated that the share of commercial and Non-commercial energy in the rural areas works out to 20% and 80% respectively, Corresponding figures for Urban areas are 49% and 51% [NCAER, 1985]
6.1 Remote Sensing Applications in Bioresource Inventory
1
Kimothi,M.M. et.al. (1997) carried out a study of horticultural plantations in Kumarsain Tahsil in Simla district of Himachal Pradesh using remote sensing IRS-LISS II satellite data, Survey of India Topographic maps, Forest working plan maps of Kumarsain range along with Ground truth data on location, site characteristics, growth stages and cultural practices of horticultural plantations. This study shows that the identification and discrimination of horticultural plantation needs multi-season satellite data. FCC of IRS LISS II revealed that during the April month maximum contrast between the horticultural area, forests, agriculture and other land use categories was observed. The overall interpretation accuracy assessed on 40 sample points was found to be 87% at 90% confidence limits.
Palaniyandi, M. and Nagarathinam, V. (1997) carried out land use / land cover mapping of Thiruvallur area of Chengai-MGR district in Tamil Nadu for 1986-90 using Landsat 5 TM, and IRS-1A LISS II, Sept.1986 and Survey of India Topographic maps. This study shows very little increase of 60 ha in agricultural land and a declination in forested areas from 6,593 ha in 1986 to 6,415 ha in 1990. Degraded/open forests areas have declined during the period of the analysis from 3,928 ha to 3,043 ha accommodated by an increase in forest blanks from 490 ha in 1986 to 1,297 ha in 1990.