0749-B1

Potential Biomass Density Estimation of the Makiling Forest Reserve Using GIS

Cristino L. Tiburan Jr.^[1], Myrna G. Carandang, Rex Victor O. Cruz and Nathaniel C. Bantayan

Abstract

Rainforests are known as carbon sinks, but due to the extensive degradation of resources all over the world, adverse effects of global climate change are recently being experienced. Hence, estimating forest biomass becomes indispensable. A Geographic Information System (GIS) was used to determine the potential biomass density of the Mount Makiling Forest Reserve (MFR). Five thematic maps were used, namely: 1) modified Weck's climate index map; 2) mean annual precipitation map; 3) elevation/topographic map; 4) slope map; and 5) soil texture map.

The total potential biomass of MFR was estimated at 1 690 326.50 Mg. Total actual biomass at three different periods was also estimated using forest inventory data. It was found that the total actual biomass for the years 1992, 1997 and 1999 were 1 640 372.72 Mg, 1 513 675.95 Mg and 1 542 506.79 Mg, respectively. These results implied that the potential of the reserve to produce biomass has not been optimized. This is also very significant in the sustainable management of MFR because it provides information that can be used for better planning and management operations.

Furthermore, the study found that over the years, some of the conversions of land uses in MFR were made in areas of high potential biomass density. Because of this, a decrease in the production of biomass in MFR was depicted in the 1990s. This was observed also in the degradation ratios (DR) computed. Among the three different periods, it was in 1992 (DR = 0.97) that degradation was found to be lowest compared to the other years. Along with this observation, actual biomass and potential biomass of various land covers in MFR were also compared. The study revealed that total potential biomass of secondary forest was lower than its total actual biomass. On the other hand, mossy forest, agroforestry, brushland, and non-forest areas have higher total potential biomass than their total actual biomass.

Introduction

The tropical rainforests are considered as the richest, oldest, most productive and most complex ecosystem on earth. It is estimated that as many as 30 million species of plants and animals, more than half of all life forms, live in tropical forests. Though they only cover less than 2% of the earth's surface, they are habitats of 50% to 70% of all life forms on our planet.

But through the years, tropical forests became the forefront of global crises due to extensive and massive forest clearings happening all over the world. Loss of biodiversity and global environmental changes are the two main effects of deforestation. It resulted to vast areas of damaged, marginal and degraded lands. Aside from these, forest fragmentation and habitat degradation further aggravated the impacts of deforestation to climatic change and hydrological cycle. A report by the International Geosphere - Biosphere Program estimates that in the last 300 years, there has been a net loss of approximately 6 million sq. km of forest area (an area about the size of Australia). In the Philippines, the average rate of deforestation is estimated to be 100,000 ha/year (Master Plan for Forestry Development 1990).

Forest biomass is very useful for comparing structural and functional attributes of forest ecosystems across a wide range of environmental conditions. It is also very relevant to issues related to global change especially to carbon accumulation in the atmosphere known as "global warming". Thus, many efforts have been extended on biomass estimation because it provides an estimate of the carbon pools in forest vegetation. It was assumed that about 45% of the biomass content is carbon (Lasco 2002). Furthermore, biomass estimates also provide the means of calculating the amount of carbon dioxide that can be removed from the atmosphere by regrowing forests or by plantations because they establish the rates of biomass production and the upper bounds for carbon sequestration (Brown 1997).

Over the years, many attempts have already been made to estimate forest biomass. Among these is the use of existing forest inventory data. But because global implications of resource loss became conspicuous and tremendous, a more advanced approach of biomass estimation is needed. This method will not only provide enhancement in the ability to estimate biomass but can also hasten assessment of tropical forests over large scale areas. Nowadays, information technology is gaining popularity and one of these is the Geographic Information System (GIS). Hence, this study aims to acknowledge the gap presently manifested under Philippine setting with regards to the application of geographic information system in estimating the potential biomass density of forest resources.

Methodology

Description of the Study Area

The Mt. Makiling Forest Reserve (MFR) has a total area of 4,244 ha and was established as a reserve in 1910. The highest point in the reserve is Peak 2 which is located at 14°08' north latitude and 121°11' east longitude. The reserve lies within 65 km south of Metro Manila and it covers the municipalities of Los Baños, Bay and Calamba in the province of Laguna and Sto. Tomas in the province of Batangas. Furthermore, MFR has six (6) major zones based on watershed boundaries.

MFR is also rich in biological diversity. It contains native and exotic species which are classified into 225 families, 949 genera, 2,638 species, 19 subspecies, 167 varieties and a number of cultivars of flowering plants and ferns. Likewise, there are about 44 mammals, 241 bird species, 69 reptiles and 21 amphibians present within the reserve. It also offers many interesting places for outdoor recreation and ecotourism spots like Mudspring, Flat Rocks, Makiling Botanic Gardens and panoramic views from the Peaks, National Art Center, Pook ni Maria Makiling and many more.

In terms of its demography, there are about 280 households. Most of these families settle on areas accessible to livelihood and where the terrain is not very steep. The occupants of the reserve are mostly migrants dating back in 1898-1899 with 19 original families. At present, however, there are about 1,225 total population of the area. The area is also composed of many land uses such as agroforestry, agricultural farm, plantations, old growth forests, leased areas, grasslands and the likes.

Potential Biomass Density Index

The model that was used in the study generated an index of the potential biomass density of MFR. This was patterned after the methodology used by Iverson et. al in 1994 in estimating the biomass density of the South and Southeast Asia. The model was additive in nature that performs a weighted overlay of map layers based on significant assumptions and specifications. There were five thematic maps used in the study and these are as follows: 1) the modified Weck's climate index (25 points); 2) mean annual precipitation map (25 points); 3) elevation/topographic map (13 points); 4) slope map (12 points); and 5) soil texture map (25 points). In total, a maximum number of 100 points can be generated.

Calibration of the Potential Biomass Density Index

After the overlay analysis of the five thematic maps, a model depicting the indices of the potential biomass density of MFR was produced. From this, real biomass density values were assigned to a particular range of index values. These values were derived from existing forest inventory data of MFR. To define the range of values in between, the upper and lower limits were set first. After the limits were established, linearity was assumed to biomass density between these various ranges.

Potential Biomass Density Map

The values derived from the calibration of the potential biomass density index were substituted to the weights initially given. These values varied from the highest range of biomass density the forest reserve can possibly provide and the lowest biomass alternative. This model now showed the potential biomass density (PBD) of MFR under conditions where no human or natural disturbances exist. To determine the potential biomass density in a given period, the existing PBD was masked through an overlay analysis with a land cover map corresponding to a particular year. This, however, depicted the change in forest area caused by deforestation only. The resulting maps were called PBD-92, PBD-97 and PBD-99.

Results and Discussion

Modified Weck's Climate Index

The modified Weck's climate index (mWCI) provided an avenue to assess the combined effects of different climatic parameters to the potential productivity of the area specifically biomass production. The index included mean annual precipitation, relative humidity, temperature and mean length of daylight. It was observed that within 10 years, there were instances where precipitation fell below 2,000 mm/year particularly during the El Niño phenomenon that hit the Philippines in 1994 and 1997. On the other hand, the annual average relative humidity was computed at 81.80%, mean annual temperature was 29.01°C and the mean length of daylight was 5.28 hours. The average growing season for MFR was identified to be 9 months.

Mean Annual Precipitation Map

An independent map layer for the mean annual precipitation of MFR was used. Two sets of data were collected in two different stations, the Agrometeorological station and the Makiling Central Nursery station. Since data from 1991-2000 was lacking in the latter station, estimation of the missing rainfall data was done. First, a t-test was conducted on the data present for both stations to determine whether they are significantly different from each other. It was found that t-computed (1.59) was less than the t-value (2.228) which means that they were not significantly different. Therefore the mean difference of rainfall for the two stations was used to determine the missing rainfall values in the Makiling Central Nursery. Initially, six classes of rainfall were identified but were reclassified to satisfy the assumptions made in assigning weights. Rainfall class 2401-2800 mm covered the highest area in the reserve (3,054.62 ha) while the lowest (44.28 ha) was at rainfall class 3201-3600 mm.

Elevation Map

The elevation of MFR ranged from 40 masl to 1,100 masl. Most areas of the reserve were found to have elevations ranging from 200 masl to 500 masl (53.05%). In addition, the highest area covered by a given elevation class was between 300-400 masl (18.41%) while the lowest observed was along the elevation class greater than 1,000 masl (0.83%).

Slope Map

The slope of MFR was characterized into five classes portraying areas from flat to steep or mountainous terrain. The area was generally described as rolling to hilly with some portions considered as rugged to mountainous. It was noted that 35.70% of MFR or an area of 1,551 ha, has a slope class of 15%-25% while the lowest slope class was recorded at slope greater than 40% which has an area of 44 ha only.

Soil Texture Map

In MFR, the soil textural class was considered to be the best soil suitable for carrying forest biomass because it originates from loamy soils (Iverson et al. 1994). It was marked that 40.83% or 1,774 ha of the area has silty clay loam that covered portions of the three municipalities in Laguna and the municipality of Sto. Tomas in Batangas.

Calibration of the Potential Biomass Density Index

An overlay analysis was performed to generate a potential biomass density map representing the sum of the weights given to individual map layer. The process generated 20 indices ranging from a total of 77 points to 100 points. The highest index covering an area of 771.35 ha had a sum of 84 points. The derived indices were reclassified into 13 index classes (Table 1 and Figure 1).

To assign real biomass density to these index classes, the upper and lower limits were determined using existing forest inventory. The upper limit was known to be concentrated on secondary forest and mahogany plantations with biomass density of 672.80 Mg/ha and 634.99 Mg/ha, respectively. On the contrary, the lower limit was observed in an agroforestry system (alley cropping) having a biomass density of 3.80 Mg/ha. Therefore, the upper limit class of biomass density was set at >600 Mg/ha and the lower limit at <50 Mg/ha.

Results showed that potential biomass density class of 250-300 Mg/ha covered the highest area accounting to 18.43%. Furthermore, MFR was considered to possess high potential for biomass production because the distribution of classes was concentrated between 350-400 Mg/ha to >600 Mg/ha.

Potential Biomass Density Map

The potential biomass density map was also referred as PBD map. However, a PBD map of MFR at a particular year described the potential biomass density of a given land cover within the range of biomass density classes. The result from this process gave an overview of the general trends of biomass density in the reserve over the years.

PBD-92 Map

In this map, evidence showed that portions of the secondary forest were once open areas and cultivated lands. It can also be explained that most part of the secondary forest originated from closed dense forest because 60% were situated on areas with potential biomass density of >350 Mg/ha. On the other hand, 75% of the agroforestry areas have potential biomass density of >200 Mg/ha which indicated that most of these areas have high potential in biomass production. Furthermore, the brushland and non-forest areas were mostly located on areas where potential biomass density class was at >350 Mg/ha.

PBD-97 Map

This map presented almost the same trend with that of the PBD-92. Most of the areas in the secondary forest, accounting to almost 2,000 ha, have high potential biomass density. This observation can be interpreted that these areas have emanated from closed forests. With regards to agroforestry areas, more than 70% have biomass density of >250 Mg/ha which significantly addressed the occurrence of cultivation areas. A great percentage of brushland and non-forest areas were situated along sites having high potential in biomass production.

PBD-99 Map

The PBD-99 map depicted the same trend with that of the two previous PBD maps. Most the areas of these land covers were situated on sites having high potential biomass. Brushland areas were viewed to be confined only on locations with biomass density of >550 Mg/ha. In reference with non-forest areas, it was noted that these sites were mainly located along biomass density of <450 Mg/ha. This apparent evidence of forest change resulted in the decrease of the total biomass of the area.

Total Biomass and Degradation Ratio

The total actual biomass of the three different periods yield lower results compared to the total potential biomass of the reserve. Based from this observation, it can be explained that existing vegetation covers of MFR have not fully maximized the potential of the reserve. The total actual biomass of PBD-92 was computed nearest to the total potential of the reserve with a difference of almost 50,000 Mg only. It was also found out that over the years, the total actual biomass decreased.

Figure 1. Potential biomass density map of the Mt. Makiling Forest Reserve.

Actual and potential biomass along various land covers were also compared (Figure 2). In the three different years, secondary forest manifested higher actual biomass computation than that of its potential biomass. It was found out that only 16% of the area covered by secondary forest have potential biomass density of >500 Mg/ha. This can be the reason why result of total potential biomass was lower than that of the total actual biomass. Another reason could be attributed to the mean aboveground biomass density used for determining the total actual biomass. Since the entire area of secondary forest was multiplied with the mean of the inventory data for this land cover (567.64 Mg/ha), the computed total actual biomass was higher from its total potential biomass. But on the other hand, the total potential biomass of mossy forest, agroforestry, brushland, and non-forest areas was higher than the total actual biomass. This trend was observed since most of the areas under these land covers were located on sites having high potential biomass density.

Figure 2. Potential and total biomass of Mt. Makiling Forest Reserve under various periods.

Although there was the indicative presence of differences between the total actual biomass and the total potential biomass across the various land covers, the generated result was regarded to be valid. This can be validated on the biomass computed for the mossy forest. Considering the three years, a mean difference of almost 4,500 Mg only was determined. This was the lowest mean difference also across all land covers. The result was very significant since only the mossy forest remained to be undisturbed over the past years unlike the other land covers that manifested prevalent disturbances and degradations.

Degradation ratio (DR) provided an understanding of the intensity of degradation due to deforestation in the reserve with reference to its potential biomass. In the derivation of DRs, it was found that PBD-92 (0.97) had the lowest rate of degradation while PBD-97 (0.90) manifested the highest disturbance rate made on the reserve.

Conclusion

The result of the study showed that potential biomass of the reserve was not yet fully optimized. Comparing the three different periods, it was observed that PBD-92 was the closest to the total potential biomass. In addition, it was found out that actual biomass of secondary forest was higher than the potential biomass since most of the areas covered by secondary forest yield lower potential biomass density. On the contrary, other land covers like mossy forest, agroforestry, brushland and non-forest areas have higher potential biomass. This was evident because most of the areas covered by these land covers have high potential biomass density.

Literature Cited

Apan, A. A. 1999. GIS Applications in Tropical Forestry. 132 pp.

Bantayan, N. C. 1996. Participatory Decision Support Systems: The Case of the Makiling Forest Reserve, Philippines. Ph.D. Dissertation. University of Melbourne, Australia.

Brown, S. and A. E. Lugo. 1991. Biomass of tropical forests of South and Southeast Asia. Can. J. For. Res. 21:111-117.

Brown, S., L. R. Iverson and A. E. Lugo. 1992. Biomass estimates for tropical forests. World Resources Review 4:399-384.

Brown, S. 1997. Estimating biomass and biomass change of tropical forests: A primer. FAO Forestry Paper No. 134. Food and Agriculture Organization. Rome. 55 pp.

Congalton, R. G. and K. Green. 1992. The ABC of GIS. Journal of Forestry.

Iverson, L. R., S. Brown, A. Prasad, H. Mitasova, A. J. R. Gillespie and A. E. Lugo. 1994. Use of GIS for estimating potential and actual biomass for continental South and Southeast Asia. Chapter 3. Effects on land use change on atmosphere CO₂ concentrations.

Lasco, R.D. 2002. Carbon Budgets of Tropical Forest Ecosystems in Southeast Asia. Paper delivered under the UPAA of Metro Toronto Professorial Chair, College of Forestry and Natural Resources, University of the Philippines Los Baños, College, Laguna. 21 January 2002.

Lasco, R. D. and F. B. Pulhin. 2002. Philippine Forest Ecosystems and Climate Change: Carbon Stocks, Rate of Sequestration and the Kyoto Protocol. Environmental Forestry Programme, College of Forestry and Natural Resources, University of the Philippines at Los Baños, Laguna.

Master Plan for Forestry Development. 1990. Department of Environment and Natural Resources.

Sanchez, P. A. 1976. Properties and management of soils in the tropics. John Wiley and Sons, New York.

Vallesteros, A. 2002. GIS-Aided Determination of Socio-Economic Variables Affecting Landuse Change in the Mt. Makiling Reserve, Philippines.