Content uploaded by Arif Oguz Altunel
Author content
All content in this area was uploaded by Arif Oguz Altunel on May 07, 2020
Content may be subject to copyright.
5th International Conference on Geographic Information Systems (ICGIS-2008)
2-5 July, 2008 / Fatih University / Istanbul, Turkey
A CASE STUDY: WATERSHED SCALE FOREST ROAD ANALYSIS
USING GIS TECHNIQUES
Abdurrahim AYDIN1, Dr. Arif Oguz ALTUNEL2
ABSTRACT
Roads are crucial to sustainable forest management. Forest roads are needed for implementing any kind of forest
management practices, for acting as fire barriers during forest fires, to give way to higher pastures for grazing, to
provide access for recreational activities and also, in Turkey and many parts of the world, to provide a link
between forest dwellers, small villages, and towns settled inside, in the vicinity or around the forested regions and
the rest of the country. On the other hand, they require, during planning and construction afterwards, a great deal
of attention because they are proven to cause soil compaction due to heavy loads and traffic especially during
harvesting. Soil compaction leads to soil erosion, which is the main reason for water pollution and the destruction
of aquatic ecosystems. They also cause forest defragmentation and unforeseeable settlements. That is why a forest
road network must be well planned and carefully developed to ensure the sustainable management of forest
resources. In this study, watershed scale forest road analysis was conducted in Kayran creek watershed of Kuyucak
district of Aydin province in western Turkey. Forest lands and existing forest roads in the watershed were specified
and a digital data set was developed from the forest management plans and forest road maps using GIS
techniques. General road density, real road density, relative road density, and exploitation ratio were determined as
decision variables in forest road analysis. The results indicated that developing forest road networks in watershed
scale by using GIS techniques will significantly reduce time and improve economic efficiency in planning forest
roads.
Keywords: Forest roads, Road density, Kayran, GIS.
1. INTRODUCTION
Forest roads are built to provide easy access to forests. In general, transporting logs by logging trucks
through the low volume forest roads is less expensive than transporting them over the forest terrain,
however; the costs of building forest roads trade off against this reduction in transportation costs.
Besides, excessive sediment delivered to streams from a road section can have a dramatic effect on
water supplies, aquatic life, and wildlife populations (Akay, 2003). Therefore, the forest road network
should be planned to minimize the total cost of timber extraction, logging, and road construction, as
well as to decrease the environmental impacts (Sundberg and Silverside 1988).
Applying adequate forest management techniques and using modern technologies in forest operations
are very important for sustainable forestry. In long term, forest roads have a great influence on all forest
management activities. A forest road network is developed for many purposes (1) primarily
transportation of forest products, especially timber species, (2) easier and correct interventions on
forest cultivation, (3) wide application of mechanization in all forestry works, (4) transportation of
workers and delivery of material to distant sites, (5) better and easier organization of work, (6)
protection from fire and diseases, (7) easier measurement, management and supervision of the forests
and (8) access for the local people living inside forested areas (Bayoğlu 1996, Picman and Pentek 1996).
However, forest roads may have other vital functions such as rural development in countries like Turkey
where forest villages represent the majority of the rural communities living in mountain regions.
Forest roads are also used in other economic sectors including agriculture, mining, tourism, health
services, etc. Thus, if we look at with a broader perspective, there is a manifold of objectives in building
forst roads, although, it is the most expensive part of investment in forestry (Picman and Pentek 1996).
In locating forest road networks, optimal road spacing method has been generally used by the road
managers, considering various factors including spatial diversity of forest stands and terrain conditions,
which highly efect total cost of transportation and construction.
1 Ministry of Environment and Forestry, The Western Blacksea Forestry Research Institute, Bolu-Turkey
abdurrahim.aydin@boluarastirma.gov.tr, aoaltunel@hotmail.com.
11
Optimal road spacing is a value that provides a guide for locating roads and a target in overall expenses,
but does not suggest where the roads should actually be placed. Nevertheless, road location is not only
controled by optimal spacing but also by various restrictions dictated by the forest terrain. It is therefore
necessary to develop a road locating method that is capable of accounting the spatial variation in forest
environments and resources (Tan, 1999). General Forest Directorate of Turkey requires a forest road
spacing of 500 m in all forest stands.
Adequate road density is also crutial in developing a forest road plan. Road density is defined as the
number of linear metres of road per hectare (Erdaş 1997, Bayoğlu 1997, Ryan at all.2004). For any given
forest site, there is an optimum road density that minimizes the combined cost of construction,
maintenance and timber extraction (Ryan at all.,2004).
Relative forest road density, regardless of the extraction methods, should be at least 20 m per hektare
(Erdaş, 1997). This means a 250m strip from both sides of a particular road segment, which makes 500
m wide logging units. Within this 500m, extraction can be carried out uphill or downhill, depending
upon the topological conditions of the terrain.
However, if there is equal to or more than 250m3/ha, between 100-250m3/ha, or less than 100m3/ha
growing stock then the road density will respectively be 20m/ha, 10m/ha, or less than 10m/ha
(Bayoğlu, 1997). Futher, in degraded stands, the road density should be around 5 to 6m/ha (Erdaş,
1997).
Many tasks that were difficult and time-consuming to perform by traditional techniques began to be
dealt with faster, more reliable, and more easy ways in computer environment (Yılmaz 2002).
Geographic Information Systems (GIS) techniques have been increasingly used in developing forest
road networks to reduce time and road costs (Erdaş and Gümüş, 2000, Coulter et al. 2001, Akay 2003,
Akay et al. 2004, Aruga et al. 2005 and Altunel, 2006). GIS is a computer-based system that provides
the following four sets of capabilities to handle georeferenced data (1) input, (2) data management (data
storage and retrieval), (3) manipulation and analysis, and (4) output (Aronoff 1993). A GIS integrated
road system can dramatically reduce the construction and the maintenance cost of forest roads (Skally
2003). The first studies of GIS implementing forest road planning began to appear in the early 1990’s.
(Bayoğlu at all.1995).
2. MATERIAL AND METHODS
2.1. Study Area
Study area, Kayran creek watershed, is located in the Northern part of Aydin province in Southwestern
Turkey, and lies between the coordinates 28°29'15"-28°33'50" East and 38°03'34"-38°53'55"North, in
Aegean Region, western Anatolia, Turkey. The area of the watershed is 66.02km² and it forms a long
narrow shape (Figure 1).
The topography of the watershed is broken and ground elevation increases from south to north with
the maximum elevation of 1469 m. Mean slope is 38.5% and slope classes of watershed are; 5.0% 0-
6%, 7.7% 6-12%, 8,1% 12-20%, 14.1% 20-30%. Ground slope is over 12% is corresponds to 87.3%
of the study area. In 46.19% of the watershed, altitude stays between 800 m and 1200 m with the mean
alltitute of 796 m.
In the study area, there is a meteorological station (38°00' N, 28°33'E, H=600m) with short term
records (1971-2000). Annual precipitation in the watershed is about 744.7 mm.The wettest month is
January, while the driest month is August. Maximum daily rainfall is computed to be 102.5 mm in
December. In study area, Mediterranean climate dominates with hot and dry summers and mild and
rainy winters.
12
5th International Conference on Geographic Information Systems (ICGIS-2008)
2-5 July, 2008 / Fatih University / Istanbul, Turkey
Figure 1: Location of Kayran Creek Watershed
Study area lies in the Mediterranean vegetation region that consists of xerophytes (Atalay, 1983).
Vegetation cover in the watershed is made up of two zones: the first zone that climbs up to 1000 m is
Pyrenean pine (Pinus brutia Ten.). Oak groups are dominant in the Pyrenean pine forests that are the
main tree species of the area. Second zone starting from 1000-1100 meters is Crimean pine (Pinus nigra
Arn.). But vegetation limits in the area are not natural. Natural vegetation cover was disturbed by
human being and cultural plantations climbed up to the 800-1000 meters. Because of overgrazing, soil
is bare, topsoil eroded and surface of rangelands covered with stones and non-fodder plants. Some
rangelands were used for agriculture. Besides, villagers, near and within the forests, use open areas in
the forest as a rangeland. Some coppice forest areas, seen in 1962 dated air photos, had been turned
into fig and chestnut agriculture until 1983 (Anonymous, 1983, 2001).
2.2. Generating Digital Maps
Watershed delineation was performed on a 1/25,000 scale map, using nearest neighbour method with
a 1.34 m RMS error in Arcview 8.3. Additionally, forest management map and current road layout map
of Kayran watershed were digitized and georeferenced. In calculating the forest roads on the watershed
boundaries, half of the current total road length was considered, because this road can only be used in
one side to log a 250 m strip, inner part of the watershed in this case
2.3. Determining General, Relative and Real Road Density and Exploitation Ratio
2.3.1. General Road Density
General road density is determined by dividing the total road length, obtained by digitizing the
watershed road map, over the area of watershed, using the following equation;
GRD=TRL/A (1)
where;
13
GRD : general road density (mha-1)
TRL : total road length (m)
A :area of watershed; forested area, open area, settlement and agriculture practiced
area (ha)
2.3.2. Relative Road Density
Relative road density is determined by dividing the forest road length corresponding only to forested
area, obtained by digitizing the watershed road map, over the forest area, obtained by digitizing forest
management map, using Equation 2.
RRD=FRL/FA (2)
where;
RRD : relative road density (mha-1)
FRL : forest road length (m)
FA : forest area (ha)
2.3.3. Real Road Density
Real road density is determined by dividing the total road length, obtained by digitazing the watershed
road map, over the forest area, obtained by digitazing the forest management map, using the following
equation;
ReRD= TRL/FA (3)
where;
ReRD : real road density (mha-1)
TRL : total road length (m)
FA : forest area (ha)
2.3.4. Exploitable Forest Area and Exploitation Raito
Firstly, total road length in the watershed was determined by digitazing the road map to obtain
exploitable forest area. Then, 250 m buffers from both sides of the road was located. Finally, forest area
within and outside of the buffer line was determined. Exploitation ratio is the ratio of exploitable forest
area over total forest area;
ER= EA/FA*100 (4)
where;
ER : exploitation ratio (%;dimensionless)
EA : exploitated area (ha)
FA : forest area (ha)
3. RESULTS AND DISCUSSION
Based on the national forest road design guidelines, road density in coppice area, degraded coppice area,
and degraded forest area should be around 6m/ha, 5m/ha, and less than10 m/ha respectively. Since
growing stock in this watershed is between 100-250 m3/ha, the road density was estimated to be around
10 m/ha (Figure 2-a,b).
14
5th International Conference on Geographic Information Systems (ICGIS-2008)
2-5 July, 2008 / Fatih University / Istanbul, Turkey
Maps required for the analysis were generated from the data provided by the Nazilli Forest Directorate.
Various other information concerning the study area conducted on figure 2 and indicated in table 1.
Based on GIS analysis conducted on figure 3, decision variables including general, relative and real
forest road densities, and exploitation ratio in the watershed were determined and then presented in
table 2. Finally, the present situation in Kayran watershed was indicated in table 2.
Table 1:
Specifications of W
atershed
Watershed
specifications
Watershed
Area
(ha)
Forest
in
watershed
(ha)
Non
-
forest
(ha)
Coppice
(ha)
Degraded
coppice
(ha)
Degraded
forest
(ha)
6606.22
4501.97
2104.25
1493.81
298.18
1493.81
Watershed
specifications
Good
forest
(ha)
Exploitated
forest
(ha)
Non
-
exploitated
forest
(ha)
Forest road
Length
(ha)
Non
-
forest
road
Length
(ha)
Total road
length
(ha)
1294
.
05
2442.46
2059.51
72613.18
41655.75
114268.93
Table 2:
Determining
of General, Relative a
nd Real Road Density
Analysis
results
General road densi
ty
(mha-1)
Relative road
density
(mha-1)
Real road
density
(mha-1)
Road density
for coppice
(mha-1)
17,30
16,13
25,38
13,61
Analysis
results
Road density
Degraded coppice
(mha-1)
Road density for
degraded forest
(mha-1)
Road density
for good forest
(mha-1)
Exploitation ratio
(%)
22,62
13,89
17,70
54,25
(a)
(b)
Figure 2: Land-Use (a) and Road Map (b) of Kayran Creek Watershed.
15
(a)
(b)
Figure 3: Exploitated (a) Non-Exploitated Areas (b) In Kayran Creek Watershed.
The results indicated that, the road density in overall watershed is unnecessarly high, especaially in
degraded coppice (22.62m/ha).
Exploitated forest area was calculated as 2442,46 ha, which corresponds to 54.25% of the total forest
area (Figure 3-a). Non-exploitated area was calculated as 2059,51 ha which corresponds to 45,75% of
the total forest area (Figure 3-b).
This study has shown that, in this particular watershed, when roads were located for logging purposes,
no attention was paid on adequate road planning procedure. Therefore, the amount of roads laid down
in unit area was very high. Although the road density was so high, over 45% of the forest area remained
unexploitable.
4. CONCLUSIONS
Road construction and maintenance lead the most costly practices in forest management. To improve
the cost efficiency, computer-aided systems combined with high speed computers and advanced GIS
techniques should be used in planning stages of the road location. Although using GIS in planning
forest roads and logging operations began in the early 1990’s, there has been still little examples of GIS
applications in the field of forestry in Turkey. This study emphasises that GIS can help to understand
and analyze a proposed road system to reach and yield a desirable exploitation ratio well before it is
materialized. Furthermore, the best and most suitable routes can be determined using GIS techniques,
and then an optimum forest road network can be put into effect.
In this particular study, results indicated that the road density was less than what it should have been for
intensive forest management practices. Additionally, even though the road density can be acceptable for
logging, there are still large unexploitable patches. It can be concluded that the major reason for that is
inadequately planned forest road networks.. In the area, it is even possible to reach various particular
locations from two or more road segments, strengthening the conclusion that no sound planning was
used in this road layout. It is expected that conducting similar analyses in other forest regions of the
country will show similar results. Implementation of GIS in the planning stage of forest network
analyses will save a lot of capital for countries like Turkey where there are still more urgent
problem to be solved.
16
5th International Conference on Geographic Information Systems (ICGIS-2008)
2-5 July, 2008 / Fatih University / Istanbul, Turkey
5. REFERENCES
Akay, A.E. (2003). Minimizing Total Cost of Construction, Maintenance, and Transportation Costs with Computer-
Aided Forest Road Design. Ph.D. Thesis, Oregon State University, Corvallis, Oregon. 229 p.
Akay A. E., Karas, I. R., J. Sessions, Yuksel, A., Bozali, N. & Gundogan, R., (2004). Using high-
resolution digital elevation model for computer-aided forest road design, The Proceedings for
XXth International Society for Photogrammetry and Remote Sensing (ISPRS) Congress, July 2004, Istanbul,
Turkey.
Altunel, A. O. (2006). GIS Integrated Forest Road Planning, Fen Bilimleri Enstitüsü, Istanbul University,
Istanbul, Unpublished PhD. Dissertation, 145 p.
Anonymous, (1983). Buyuk Menderes River Basin, Northern Region of Nazilli Province Erosion
Control, Afforestation and Rangeland Improvement Planning Report, General Directorate of Forest,
Aydin, Turkey .
Anonymous, (2001). Kuyucak forest district management plan. General Directorate of Forestry, Ankara.
Aronoff, S. (1993). Geographic Information Systems: A Management Perspective. WDL Publications 3rd Edititon,
Canada.
Aruga, K., Sessions, J. & Akay, A. E. (2005). Application of an airborne laser scanner to forest road
design with accurate earthwork volumes. J. For. Res. 10 (2): 113-123.
Atalay, I. (1983). Introduction to the Vegetation Geography of Turkey, Ege University Social Sciences Faculty
Publication Number: 19, Izmir, Turkey .
Bayoğlu, S. (1996). Planning Of Forest Transportation. Istanbul University, Institute of Sciences publication,
pub. no: 8, Istanbul.
Bayoğlu, S. (1997). Forest Transport Facilities and Vehicles (Forest Roads). The Istanbul University, Forestry
Faculty Publications. Istanbul.
Coulter, E., Chung,W., Akay, A. & Sessions, J. (2001). Optimization of Forest Road Layout Using a
High Resolution Digital Terrain Model Generated from LIDAR Data, First International Precision
Forestry Symposium, University of Washington, College of Forest Resources, Seattle, Washington,
USA.
Erdaş, O. (1997). Forest Roads. Volume-I. Karadeniz Technical University Forestry Faculty Publications,
pub no:25, Trabzon.
Erdaş,O. & Gümüş, S. (2000). The Use of Geographical Information Systems in Selecting Forest Road
Routes.Turkish Journal of Agriculture and Forestry 24 (2000) 611-619.
Picman, D. & Pentek, T. (1996). The Influence of Forest Roads Building and Maintenance Costs on
Their Optimum Density Low-Lying Forests of Crotia. Proceedings of the Seminar on Environmentally
Sound Forest Roads and Wood Transport. 17-22 june, Sinaja, Romania.
Ryan, T., Philips, H., Ramsay, J. & Dempsey, J. (2004). Forest Road Manual. Guidelines For The Design,
Construction And Management Of Forest Roads. COFORD, Dublin.
Skally, C. (2003). Forest Road Geographic Information System (GIS) Data Collection Process and Summary of Road
Data. Minnesota Forest Resources Council, Documnet #LT-0302, MN.
Sundberg, U. & Silversides, C. R., (1988). Operational Efficiency in Forestry. Vol. 1: Analysis. Kluwer
Academic Publishers. 219 p.
Tan, J. (1999). Locating Forest Roads by a Spatial and Heuristic Procedure Using Microcomputers.
Journal of Forest Engineering. Vol 10 no.2 July 1999.
Yılmaz, O. Y. (2002). Temporal Geographic Information Systems. Review of The Istanbul University, Forestry
Faculty, Series B, Volume 52, Number 2. Istanbul.
17
