Construction of New Forest Roads in Malaysia Using a GIS-Based Decision Support System

This paper describes how a Geographical Information System (GIS)-based Decision Support System (DSS) was applied in selecting the most compatible block or compartment to construct a new forest road for a sustainable timber harvesting purposes. The study area was located in Gunong Stong Permanent Forest Reserve, Kelantan, Malaysia between latitude 500’ N and 525’ N and longitude 10120’ E and 10205’ E. Four criteria have been considered in this paper which is timber volume (m), slope (degree), ground condition and distance from the primary and secondary forest road that previously exists (m). ArcView and ArcMap softwares were used to evaluate all the criteria using remote sensing and field data. Results indicated that the best block that fulfill all the criteria chosen for the new forest road construction is Block Nos. 9 and 11.


Introduction
Virtually everyone makes hundreds of decisions each day.Good decision making means that the person are informed and have relevant and appropriate information on which to base the choice.Information has been referred as facts, numbers, historical data, graphics, pictures, and sounds (Sauter, 1997).Decision support system (DSS) technology and applications was built to support the decision making.DSS is being used in many fields including forestry.GIS is one of the decision support system tools that help the forest manager or silviculturist to do better decision making.GIS help the forester to do planning and preparation of a block or compartment based on their objectives or targets.GIS has the capability to map, record data, planning and involve in special projects.
Construction of new forest road is one of the main issues that arise in forestry management.Many criteria aspects have to be considered before a new road is constructed.In this study, four criteria have been identified which is timber volume (m 3 ), slope (degree), ground condition and distance from the road that previously exists (m).Scoring model is being used as a weighting for each classes of criteria (Williams, 2000) in this study.ArcMap and ArcCatalog is the GIS software that has been selected to decide the best block that fulfills all criteria.

Study area
State of Kelantan, Malaysia at the northern part of Peninsular Malaysia, in a northern-east direction, between latitude 5 o 00' N and 5 o 25' N and longitude 101 o 20' E and 102 o 05' E had been selected as the study area (Figure 1).Compartment 1 until compartment 16 has been selected as the demonstrative area (Figure 2).The terrain of the study area was classified as very steep slope (greater than 10 o ).The climate in the study area is a hot humid tropical climate.Temperature and rainfall has been categorized by three distinct features: uniform temperature, high humidity and heavy rainfall.The temperature rarely fell below 20.5 o C nor rose beyond 36 o C. The mean annual rainfall was seldom less than 2000 mm.The heaviest rainfall occurred during the November, December and January.Dry season was been marked from February until April.<Figure 1.> <Figure 2>

Identification and evaluation of criteria
The determination of criteria is very important as guidance for collecting data that is needed to help the effectiveness of decision making process.Four criteria have been identified: timber volume (m 3 ), slope (degree), ground condition and distance from the road that had been exist (m).All the selected criteria were classified.Table 1 through Table 4 shows the classification of the criteria.<Table 1: The volume of the harvest block classes> <Table 2: Criteria for evaluating slope> <Table 3: Criteria for evaluating ground conditions> <Table 4: Criteria for evaluating distance from existing secondary roads> Next, each classification of criteria was given a score.The maximum score given was one hundred and the minimum was zero (Table 5).This evaluation was very important for the calculation process in ArcMap software to decide the best compartment to construct the new forest road.<Table 5: Evaluation criteria for the harvest block selection>

Data collection
Data collection process is very important for data validation and integration.Data that was used in this study are described in Cabral, 2000.Table 6 shows the data for demonstration area .<Table 6: Information of demonstration area>

System implementation
ArcMap and ArcCatalog software had been chosen as decision support tools to find the best block or compartment to construct new logging roads.Both ArcMap and ArcCatalog software are integrated.In ArcMap, information concerning the demonstration area will be stored in a database (Figures 3 and 4).Next, scores will be given for each data that has been stored (Figure 5).The calculation of score value was done to get the result of this study using Raster Calculator function that exists in ArcMap.The formula used for calculation is: ArcCatalog has been used to save all the layer files from ArcMap.Combination of all layers will also be saved in ArcCatalog (Figure 5).

Results and discussion
Figure 6 shows the classes of volume of harvest for the blocks.The highest score, 60 shows the highest volume that can be harvested in a block.Meanwhile, 40 the lowest score showed the lowest volume that can be harvested.The highest volume that can be harvested in the demonstration area ranges from 40 m 3 /ha to 50 m 3 /ha whereas 20 m 3 /ha to 30 m 3 /ha is the lowest volume.Table 7 showed results of a selected demonstration area for the estimation of timber volume.

<Figure 6> <Table 7: Attributes of volume for harvest block classes>
The second criteria that had been considered in this study are slope.The demonstration area has all classes of slope from 0 o until more than 45 o .From the illustration in Figure 7, Block Nos. 8, 9 and 10 has the lowest slope.Block Nos. 3, 4 and 16 have the highest slope.If the slope is high, the location is considered not suitable to construct new forest roads.Table 8 shows the attributes of slope for each block.

<Figure 7> <Table 8: Attributes of slope>
The next criterion is the ground condition.The best ground condition is in block number 3, 4 and 16 (score = 100).The soil texture of these three blocks is sandy loam and fresh for soil moisture.The worst ground condition is in Block Nos.
8, 9 and 10 (score = 10) in which the soil texture is silty clay loam and the soil moisture is very wet.A good ground condition will make the construction of the road much better.Figure 8 shows the score of ground condition and Table 9 shows the attributes.<Figure 8> <Table 9: Attributes of ground condition> The last criterion taken into consideration is the distance of the block from an existing forest road.Distance is considered because it affects the cost for construction of the road.Block No. 13, 14, 15 and 16 has the highest score because those blocks have existing forest roads.Block No. 7 has the highest potential to construct new forest roads.This is followed by Block No. 8, 10 and 12 (score = 80).The lowest score is in Block Nos. 1, 3, 6 and 9 (score = 20) where the distance is very high from the existing forest road.Scores of distance from existing forest roads are shown in Figure 9 and the attributes shown in Table 10.<Figure 9> <Table 10: Attributes of distance from existing forest road> Figure 10 shows the result of this study which is the combination of all criteria (timber volume, slope, ground condition and distance from the road that previously existed) that had been considered.Block Nos. 13, 14, 15 and 16 have existing forest road.The block that has the highest total score is block number 9 and 11.Block Nos. 7 and 8 have total score equal to 250.This is followed by Block No. 5 (total score = 240), 10 (total score = 230), 6 (total score = 220), 4 (total score = 200), 3 (total score = 180) and 1 (total score = 160).The worst block to construct a new forest road is Block No. 2 (total score = 150).

Conclusion
GIS has a lot advantages in supporting the decision making process in sustainable forestry management especially for construction of new forest roads in minimal impact timber harvesting.The forester can achieve their objectives by using any combination layers in ArcMap and save the result using ArcCatalog for future reference and implementation.

Table 1 .
The volume of the harvest block classes

Table 5 .
Evaluation criteria for the harvest block selection

Table 6 .
Information of demonstration area

Table 7 .
Attributes of volume for harvest block classes

Table 8 .
Attributes of slope

Table 9 .
Attributes of ground condition

Table 10 .
Attributes of distance from existing forest road