Pipeline Routing Using Geospatial Information System Analysis

Mahmoud Reza Delavar 1 and Fereydoon Naghibi 2

1-Assistance Professor, Dept. of Surveying and Geomatic Eng., Eng. Faculty, University of Tehran, Tehran, IRAN Email: mdelavar@chamran.ut.ac.ir 2-M.Sc., Dept. of Surveying and Geomatic Eng., Eng. Faculty, University of Tehran, Tehran, IRAN Email: Fereydoon_naghibi@yahoo.com

Abstract

A prototype least cost pipeline routing was performed using various data and GIS analysis. Ahvaz-Marun oil pipeline in south west of IRAN was chosen for development of the prototype. The Ahvaz-Marun pipeline was about 34-km and carries Ethan gas from Marun Petrochemical Company (MPC) to Ahvaz. A model was developed incorporating pipeline length, topography, geology, land use, stream, wetland, road and a railroad crossings to identify a least cost pathway. Geospatial Information System (GIS) analysis was used for spatial modeling and overlay. Costs associated with terrain conditions, geology and landuse were given from actual costs that are used in the section of pipeline design of Iranian National Petroleum. The length and cost associated with existing pipeline that made by traditional approaches were compared those of the least cost pathway through GIS best path analysis. The existing pipeline path is 34-km long, and the least cost pathway is 35- km long. Although longer in length, the least cost pathway is 29% cheaper than the existing pipeline path. The results of this analysis demonstrate the benefits of integrating data within a GIS environment which acted as a spatial decision support system for pipeline routing.

1 - Introduction
GIS is a science and technology which combines different data from various sources for route design processes through spatial analysis. Also, it is used to unify project processes including environmental characterization and project team decision making [7]. A number of research have already been performed in pipeline route design using GIS which include optimal routing for pipeline [1,9],

selection of best route for expansion pipeline [6] and gas pipeline route selection using high resolution remote sensing images[5]. This article describes one of the efforts towards investigating innovative approaches to pipeline routing. The present study was initiated to demonstrate the use of various data from different sources and geospatial information system (GIS) analysis for developing a least cost pathway for pipeline placement using Ahvaz-Marun pipeline as an example. The study area selected is in Khuzestan Province in south west of IRAN. The site lies in the southern Zagros Mountains near the Persian Gulf and is an area of the low relief along the proposed pipeline route (Figure 1). Section two discusses data acquisition and input. In sections three and four, pipeline routing criteria and least cost path analysis will be expressed.

2 Data Acquisition
Maps and field work are required for pipeline routing, pipeline design and construction. For this route, topographic maps at a scale of 1:25000 produced by Iranian National Cartographic Center (NCC) were used. The available largest scale geologic map of the area is at scales of 1:250000 and 1:100000. The area of interest for this analysis is shown in Figure 2. The least cost pathway analysis, using various data and GIS analysis, was intended to confirm the best pipeline route within this site.

Coordinate

In Lambert

Ahvaz

Marun

Km Figure 1: Study area in geologic map at 1:250000 scale, Khuzestan Province in south west
of IRAN

2-1-Input of Topographic and Geologic Data

Topographic and geologic data of the Ahvaz-Marun pipeline area were prepared in a GIS ready format and used as input to the GIS database. The locations of roads, railways, wetland, forests and drainage features were derived from the topographic map layer. The map that produced by NCC is the base for national topographic database (NTDB) and has a number of features for instance location of roads, railways ,wetland ,forests , drainage features, elevation points,) (Figure 2). In this project, digital elevation model (DEM) was produced from the elevation data (resolution of DEM is 50m x 50m). The slope map, derived from the DEM, is shown in Figure 3. It was used as input to the least cost pathway analysis. Darker tones indicate greater slopes.

River

Forest

Building

Road Agricultures
Wetland

Coordinate In UTM

Figure 2: Land cover layer at 1:25000 scale

Geological formations in area under study are shown in Figure 4.These boundaries between geologic units were extracted in digital form from the geological map and incorporated into the GIS database. These geologic units were divided as very hard rock, hard rock, soft rock and very soft rock based on the descriptions in the geological map legend.

Ahvaz

Slope Map (in degree)

Marun
Coordinate In UTM

Figure 3: Slope map (in degree)

Coordinate In Lambert
Ahvaz

Marun

Km

Figure 4: Formations derived from the geological map

Also, Faults were considered in the analysis (least cost pathway) because crossing pipeline over them is dangerous and must be avoided (Figure 5).

Ahvaz

Marun

Km

Figure 5: Faults location derived from the geological map 3 Pipeline Routing Criteria
The factors influencing pipeline route selection are technical and engineering requirements, environmental considerations and population density. However, these factors are chosen to balance engineering and construction costs against environmental costs and future liability [6]. The engineering and technical considerations used in this research include pipeline length, topography, surface geology, river and wetland crossings, road and railroad crossings and the proximity to large population centers. High relief terrain would result in higher construction costs and increase the need for pump stations [5]. Cost factors used in the least cost path analysis were calculated from existing pipeline Iranian Petroleum Company project and its normalized baseline cost. Using cost of an existing pipeline project, percentages over the baseline costs were calculated for construction in rock, clearing of brush and tree, crossing of rivers, railroads, and passing through agricultural land and wetlands. Estimates were made of

the slope ranges that are associated with four terrain categories including flat, rolling, sharp choppy and rough that are commonly used by pipeline estimators. The cost of pump station however has not been considered in this analysis. As generally, it is not desirable to route pipelines through urban and industrial areas; these areas extracted from the topographic map were assigned high costs above the baseline value. The topographic, geologic and land use data were used to develop a least cost pathway for pipeline placement. The least cost analysis was performed by assigning cost factors associated with the crossing of slopes, streams, wetlands, roads, railroads, rock, agricultural land, urban and industrial areas; developing a cumulative cost surface; and then calculating a path of least resistance across that surface. The locations of stream, road, and railroad crossings were digitized from the topographic map. The areas where rock was likely to be encountered were defined from the geologic map. A landuse map, produced from NTDB, was used to identify agricultural land and urban areas. Pipeline construction costs associated with terrain conditions, geology and landuse were calculated from actual pipeline construction projects. Very high values were assigned to urban and industrial areas, crossing of road, railway and river features and to areas outside the defined boundary of the layers.

4 - Least Cost Pathway Analysis

The pipeline project had no constrained point for passing except in Ahvaz (destination) and Marun (source). The objective of the cost pathway analysis was to compare the cost of an existing pipeline route to a least cost pathway between the two points. The analysis was performed using 50-m resolution cells (for raster layers).In the other words, in each GIS layer, a value was calculated and assigned for each 50 m x 50 m cell in the study area. The analysis was accomplished by entering map data into a GIS. GIS analysis is used for spatial modeling and data overlay. The GIS provided a framework for developing and overlaying all the input layers and carrying out spatial analysis. ArcView and Arc/Info software were used for display and the spatial analysis, respectively. Least cost pathway analysis is a modification of algorithms that traditionally are used in GIS for drainage basin analysis, with the path

constrained to flow through specific nodes to specified endpoints. The first step in least cost pathway analysis is the generation of a single weighted surface layer, based on input layers generated from a table of weights that contribute to the cost of traversing between two points (cells)[2, 3, 4 and 10]. The combined weighted surface is analogous to a topographic surface, in that it has peaks (areas of relatively high cost) and valleys (areas of relatively low cost). The cost surface is shown in Figure 6. The darkest tones show the areas with highest costs and the lightest tones indicate areas with lowest costs. The highest costs are in urban areas and in large bodies of water and roads. Moderate costs are in forest and wetland with high slope. The lowest costs were in areas with bare ground, dry grass, less dense native vegetation and agriculture. From the weighted surface, an accumulative cost surface was generated. Least cost pathways could follow the resulting surface. Accumulative cost surface utilizes a single point of origin and accumulates the sum of cells as one moves from the origin. The accumulative cost surface consists of the sum of cells between any location on the surface and the point of origin. Sums accumulate as they radiate out from the point of origin. The accumulative cost surface does not show which cell to return to or how to get there. A separate surface has to be generated to returns a surface with a value ranging from 0 to 8 that can be used to reconstruct the route to the origin. Each value (0 to 8) identifies which neighboring cell to move into to get back to the origin. Once the accumulative cost and direction surfaces are created, least-cost path route can be derived from any designated destination cell or the end point [2, 3, 4, and 10]. There are a total of 1084 cells along the existing pipeline route and 1034 cell along the least cost pathway (Figure 7). Table 1 shows that the existing pipeline route traversed a larger number of urban, roads and wetlands cells than the least cost pathway.

Ahvaz Marun

Km

Figure 6: Cost surface

Ahvaz

Marun

Km

Figure 7 : Comparison between proposed and existing route

For each 50 m x 50 m in resolution element, there exist as many cells as there are data layers. In each data layer, specific costs above the baseline cost are associated with each feature listed in Table 1.
Table 1: Number of cells that existing and least cost pathway are crossing various features

Feature Urban and industrial Crossing of roads Crossing of rivers Wetlands Forest Very hard rock and hard rock Slope < 20

Existing Route 8 23 0 46 0 203 823

Least Cost Pathway 5 9 0 12 0 175 952

5 Conclusions
The results of the least cost analysis are shown in Figure 7 and Table 1.Incremental costs resulting from terrain, geology, and land use were accumulated for these routes (the existing and proposed pipelines) along the cost surface. The existing pipeline path was 34 km long and the least cost pathway was 35 km long. Although the least cost pathway was longer and the analysis indicated that it would be 29% cheaper to construct than existing pipeline path. These results indicate that the proposed pipeline route is the most cost-effective. Most of the cost difference between the existing pipeline route and the least cost analysis can be attributed to be greater cost associated with the larger number of urban, road and river crossing cells. Methods also need to be developed to eliminate sharp angles from the least cost pathway. Actual costs should include costs incurred by construction and environmental considerations. However, having built a database that includes topography, geology and land use from satellite imagery and available maps for an area of interest, additional data can be incorporated to refine the model.

However, these techniques must he used in conjunction with the many years of field experience of pipeline industry personnel and refined on a case by case basis to obtain the maximum benefits.

References
[1] Anonymous, 1993,"GIS lead to more efficient route planning", Oil Gas J., Apr. 26, 1993, pp. 81. [2] Berry, J.K., 1993, "Cartographic Modeling: The Analytical Capabilities of GIS", Oxford University. [3]DeMers, M., 2002,"GIS Modeling in Raster",New Mexico University, John Wily & Sons Inc. . [4] Douglas, D.H., 1994, "Least cost path in GIS Using accumulated cost surface and slope lines", Cartographica, Vol. 31, No. 3, pp. 37-51. [5] Hicken, J. and Krumbach,Y., 1998, "Use of high resolution remote sensing for route selection", Environmental Remote Sensing Centre, University of Wisconsin-Madison,Series ARC-UWM-004-97. [6] Montemurro, D. and Gale, T., 1996,"GIS-based process helps TransCanada select best route for expansion line", Calgary, June 22, Oil & Gas Journal pp. 63-71. [7] Naghibi, F., 2002, "Application GIS in Petroleum Industry", M.Sc. Research Seminar, Dept. of Surveying and Geomatic Eng., Eng. Faculty, University of Tehran. [8] Shafiee, S., 2000, "Evaluation of Usabilities of Raster-Based GIS for Optimum Path Determination", MSc. Thesis, Dept. of Surveying and Geomatic Eng., Eng. Faculty University of Tehran. [9] Sarkka, P. and Esko, L., 1999,"Optimal routing of pipeline", Helsinki, University of Technology, GIM, 6 Feb, pp. 6-9. [10] Stefanakis, E. and Kavouras, M., 1995, "Determination of the Optimum Path on the Earth Surface", European Conference on Spatial Information Theory: A Theoretical Basis for GIS, Springer-Verlag, pp. 241-257.