Measuring Carbon Emissions from Deforestation at Donggala Regency, Central Sulawesi Province, Indonesia

Forest is a natural resource that is very important and beneficial for the livelihood either directly or indirectly. Forest has a variety of ecological functions. One of forest functions is to maintain the amount of stored carbon. The forest area changes into non-forest area resulted in reducing forest functions as a provider of environmental services. This study aims: 1) to determine the deforestation during the period of 2000-2011, 2) to make model of the landcover change using logistic regression model, 3) to measurecarbon emissions and valuation based on impact of deforestation. The materials used in this study are : a) Indonesian Topographic Map at Scale 1: 50,000, Geospatial Information Agency (BIG), b) landcover map (year of 2000 and 2011), scale 1 : 250,000, produced by director general of forestry planning, ministry of environment and forestry, 3) environmental variables (dependent variable) such as : distance from roads, distance from streams, elevation and slope. The spatial analysis is done by land change modeler which is module in Idrisi Terrset. Meanwhile calculations of carbon storage and economic value which are done by ecosystem service modelers also as a Idrisi Terrset. The results show that the rate of deforestation during the period of 2000-2011 at Donggala as high as13,448.07 ha or about 1,222.55 ha per year. The impact of the forest cover changes resulted in the decrease of carbon storage up to 3.66 million tons or equivalent to 13.42 million tons of carbon emissions. Economic losses caused carbon emission in period 2000-2011 up to US$ 38,188,465 (net present value, NPV)


Introduction
Forest is natural resources that are very important and useful for life and living either directly or indirectly. Direct benefits from the existence of the forest are timber, non-timber products and wildlife. While the indirect benefits are environmental services, such as watersheds, aesthetic function, an oxygen supplier and carbon sink.
Forest destruction, climate change and global warming reduce the indirect benefit of forest because forest is the largest carbon sink and play very important role in global carbon cycle and can hold carbon at least 10 times greater than other vegetations prairie grass, crops and tundra [1].
In 2008, the United Nations launched REDD (United Nations Collaborative Programme on Reducing Emissions from Deforestation and Forest Degradation in Developing Countries) provides a mechanism to mitigate climate change by sequestering forest carbon. REDD also promotes the secondary ecosystem service benefits associated withthis forest conservation, including protection of biodiversity and water quality [2]. The average annual deforestation in Indonesia for the period 2000-2012 was 671,420 hectares, accounts for 525,516 ha of deforestation in mineral land and 145,904 ha of deforestation in peat land. During this period,9 percent of deforestation occurred in Sulawesi Sulawesi [3]. Meanwhile, refers [4]  Application of remote sensing and geographical information system was used to estimate land cover changes from multi temporally information. Integration from carbon factor was taken from secondary data (previous studies) and land cover changes data highly expected to give information about carbon stocks changes [6].
Managing landscapes for carbon storage and sequestration requires information about how much and where carbon is stored, how much carbon is sequestered or lost over time, and how shifts in land use affect the amount of carbon stored and sequestered over time. Valuation is applied to sequestration, not storage, because market prices relate only to carbon sequestration. Discount rates are multipliers that typically reduce the value of carbon sequestration over time [7] How much carbon changes lost in previous studies have not been presented in spatial data (map) and based only on statistical figures, this study is based on spatial data and numbers so it will be easier to understand the form of distribution. The study aimed: 1) to determine the deforestation during the period of 2000-2011, 2) to make modeling the forest cover change using logistic regression model, 3) to measure carbon emissions and valuation base on impact of deforestation.

Materials and Method
Tool use in this study are ArcGis and TerrSet Software. Data Used were Indonesian Topographic Map, Scale 1: 50,000, Geospatial Information Agency (BIG); Land Use Land Cover Map of Donggala Regency, Scale 1: 250,000 (2000 and 2011), Directorate General of Forestry Planning Ministry of Enviromental and Forestry (Figure 1). Included in the data are four enviromental variable (dependent variable) such as distance from streams, distance from streams, elevaton and slopes, Indonesian Topographic Map, Scale 1: 50,000, Geospatial Information Agency (Figure 2). Meanwhile economic data area carbon price (US$ 15) and discount rates (7 %). A table of land use cover (LULC) classes, containing data on carbon stored in each of the four fundamental pools for each LULC class. This study only calculate carbon above (carbon density in aboveground mass (Mg/Ha) ( Table 1).

Landcover 2000
Landcover 2011 Landcover 2022 (Projected)  Pure dry agriculture (AUA) 8 7 Mixed dry agriculture (MxUA) 10 8 Shrub (Sr)  15  9 Paddy Field (Rc) 5 10 Fish pond/aquaculture (Po) 0 11 Transmigration areas (Tr) 10 12 Settlement areas (Se) 1 13 Mining areas (Mn) 0 14 Bare ground (Br) 0 15 Open Swamp (WB) 0 Source: Refer 8 Logistic regression model (LRM) was used to model and analyze the lancover change in IDRISI TerrSet. The objective of the present study was to assess the importance of the explanatory variables on landcover change from 1990 to 2000 and predicting the probability of change by 2011. The binary presence or absence is the dependent variable for the periods 1990-2000. The predicted landcover of 2011 was validated using ROC / AUC (Relative Operating Characteristic/Area Under Curve) module of IDRISI TerrSet. The ROC module is comparing a suitability image depicting the likelihood of that class occurring (the input image) and a boolean image showing where that class actually exists (the reference image). The ROC curve is the true positive fraction vs false positive fraction and the AUC is a measure of overall performance [8].

Carbon storage/ Sequestration Model and Valuation
Calculation of emissions using the stock changes approach (stock difference) which was measured at two different time points using two factors, namely: activity data and emission factors. The main data of carbon stocks changes was derived from the data of the land cover change [6].
Carbon strorage/ Sequestration model needs the map and data tables, including the economic data. Land use/land cover (LULC) map must be A raster format, with a LULC code for each cell. The dataset should be projected in meters (UTM 50 S). The year (2000, 2011 and 2022) depicted by the LULC map, for use in calculating sequestration and economic values [7].
The carbon prices observed in these instruments vary significantly, from less than US$1/tCO2e to $US130/tCO2e. Majority of emissions (85 percent) are priced at less than US$10/tCO2e (Kossoy, et.al, 2016). In this study use assumptions the social cost of carbon (SCC) US$15 per metric ton of C and market discount rate (r = 7% per year). The value of carbon sequestration over time for a given parcel x is:

Landcover Change
Landcover change analysis was done for Donggala compared time series data from 2000 until 2011. Table 2 show the changes of land cover Donggala Regency in from 1990 to 2022*.    The changes of forest cover was caused by the deforestation, either planned or not. Planned deforestation is usually in the form of changes planned by the government for the benefit of forest land for plantations, agricultural or residential development, which is carried out lawfully in accordance with the legislation. Unplanned deforestation is a deforestation through illegal activities. The forest degradation can be caused by illegal or unauthorized activities, such as harvesting and illegal logging.

Carbon storage/ Sequestration model and valuation
The output from model simulation is raster maps, produced by carbon stroage and sequestration modul are 1) Map of current carbon storage, Map of future carbon storage, Carbon sequestration map, Map of economic value of carbon ( Figure 5) and 2) a summary of storage and sequestration and net present values of sequestration (Table 4).  The reduction of CO2 in the air by the plants was called seprocess ( C ssequestration ).This C sequestration process occurs for the survival of plants which need sunlight, carbon dioxide gas (CO2) is absorbed from the air and water as well as nutrients absorbed from the soil. Through the photosynthesis process, the CO2 in the air is absorbed by plants and converted into carbohydrates, afterwards they are distributed throughout the body of the plants and eventually are dumped throughout the plant body. Thus, measuring the amount of C stored in the body of living plants (biomass) in a field can describe the amount of CO2 in the atmosphere absorbed by plants [9].
The map of the Net Present Value (NPV), is the economic value of total sequestration in Donggala Regency area. It could be observed from 2000 to 2011 (and 2011 to 2022 *) there was a value of US$ -187.48 to 25. 269 per metric ton per pixel in areas. Meanwhile in period 2011-2022 * there was a value of US$ 0 to -177,66.The output map are the changes in land cover and its effects in sequestration. It is possible to see that the raster has a large variety of classes and that more areas are close to becoming emmiters while others have already made the transition [10].
Base on Table 4, the Net Present Value (NPV) over time of total sequestration (between the current and the future ) in Donggala Regency is US$ -38.19 million caused by the carbon emissions, base on prioce of carbon is US$ 15 and discount market is 7 %. Negative values indicate carbon lost to the atmosphere (emissions of CO2). Costs required to restore the initial conditions, so that the forest come back to normal and can absorb CO2. The impact of land cover changes caused decreased carbon stocks, especially in the forest area. In 2000forest area is 389,896 ha, equivalent to the carbon storage of 72.83 million ton. In 2011, decline in forest area of by 13.448 ha, resulted in a decrease of carbon 3.66 million tons. The net present values of sequestration (between the current and the future) US$ -38.19 million caused by the carbon emissions, based on price of carbon is US$ 15 and discount market is 7 %.

Acknowledgments
We are grateful to Centers for Research, Promotion and Cooperation, Geospasial Information Agency (BIG) for the data and financial support.