Introduction
Geographic information systems (GIS) have found widespread applications in many diverse fields as the number and range of such applications continue to increase with GIS extending to other fields. Biotechnology; a field in agriculture is one such area. The principal purpose of biotechnology is to improve productivity thus offering food security in areas that have faced food shortages in the past. With incorporation of GIS, meeting biotechnology goals has been realizable.
For instance, the South African government is realizing better crop production due to utilization of GIS coupled with biotechnology. With help of GIS, authorities are coming up with more authentic and precise crop estimates to counter uncertainty in the South African grain industry.
This move helps in achieving biotechnology goals; that is, improved crop production hence, food security. Producer Independent Crop Estimate System (PICES) is currently in use in South Africa and is used primarily to estimate area covered by grain crops. Nevertheless, there are several challenges facing integration of GIS in this agricultural sector as explicated later in this paper.
Why GIS in Biotechnology
Elementarily, understanding geography and establishing its relationship with other fields like agriculture, enables people to make wise decisions and in this case, wise food security decisions. According to Wyland (2009, p. 4), “the ability of GIS to analyze and visualize agricultural environments and work flows has proved to be very beneficial to those involved in the farming industry.”
In biotechnology, GIS is helping to foster production, manage land expeditiously, and cut unnecessary costs in production. It is true that farmers cannot control farming natural inputs like soil, land, and rainfall among others; however, understanding these inputs through analysis by GIS systems, would work to their advantage in achieving their objectives in farming.
For instance, through GIS, farmers can determine crop yield approximates, canvass soil amendments, detect soil erosion and postulate remediation strategies. Moreover, GIS is used to study other farming practices like pest control, fertilizer application, and crop diseases among others.
The Most Appropriate Data Model
The most appropriate data model in this agricultural application is the vector data models/structures (Soller 1999). “This mapping approach characterizes the vertical variations of physical properties in each 3-D map unit” (Soller & Berg 2003). This model addresses the issue of determining land under grain crops better. The three dimensional mapping element of this model qualifies it as the most appropriate tool.
Moreover, this model is the best tool for mapping geographical space; it presents spatial locations precisely and relates entities implicitly (Doe 1997). According to Johnson, (1998, p. 16), this model has “Points associated with single set of coordinates with lines connected to sequence of coordinate pairs and area sequence of interconnected lines whose 1st & last coordinate points are the same.” Therefore, given the objective of using GIS in this field, vector model stands out as the best.
Estimating land under crop cover has to be more accurate and reliable if farmers are to root out the nightmare of food insecurity. Above all, the vector model represents what the map really looks like including all the dimensions (Tomlinson 2005). The data obtained in this model can be stored in separate files with a link connecting them thus enhancing information sharing and preventing complete information loss at the same time (Soller & Lindquist 2000).
Moreover, this model comes in diverse varieties like, spaghetti, network model, dime files, and digital line graph (DLG) among others (Berry 1993). Therefore, based on the above key points, the vector data model is the most suitable one for this exercise.
Applicable Data Sets and Data Sources
As aforementioned, vector data model comes in varieties and this means that there are varieties of data sets available. In this agricultural field, one would consider using data sets like, ArcAtlas; Our Earth, which contains “global geographic and attribute data at three scales…with 1:20,000,000 for Africa” (ESRI Data & Maps 2002).
Others include digital charts of the world, and ESRI data and maps. These data sets enable users to break down geographical information into manageable sizes that can fit into computerized models (Lee & Kretzschmar 2003). The 1: 20, 000, 000 scales have been used successfully in South Africa to estimate land under grain crops as aforementioned. Applicable data sources in this case include information from farmers and crop field boundaries that have been digitalized from satellite imagery.
To complete this project of determining grain cropland cover in South Africa, the government is coming up with model maps representing this land. These model maps are the only products that this exercise has been able to come up with so far. To make these maps, several steps are involved as explicated next.
Steps Used in Obtaining the Maps
There are five steps used in developing the required maps that represent grain cropland cover in the estimation process. As aforementioned, the South African government is using PICES in this exercise.
The first step is getting satellite imagery. According to Fourie (2009, p. 9), the government of South Africa via Ministry of Agriculture provides the satellite imagery where, “SPOT Image Spot 5 satellite imagery with a 2.5-meter resolution is used as the base layer for digitizing.” After obtaining this satellite imagery, it is used to digitize the crop field boundaries.
This exercise uses ArchMap, which lies under ArchAtlas. For clear-cut boundaries and images, 1:10,000 scales are used then the elaborate metadata captured in ArcCatalog from one province to the other. Through this method, the nine South African provinces are already digitalized.
After obtaining clean and precise digitalized crop field boundaries, point frames are designed coupled with random sample point selection (Smith, Goodchild & Longley 2007). The importance of random sample point selection is to ensure representation of possible cropped fields as subjects of field survey.
A square point grid (45 m by 45m), is constructed for the total area under each of the nine provinces. All the grid points falling outside the field boundaries are eliminated from the test population for they are improbable to locate any crop. The digitalized fields are then divided depending on the possibility of getting a crop.
In these divisions, there are low, medium, and high divisions depending on the probability of getting a crop. It therefore follows that, the low classes have low probability, the medium have moderate probability while the high classes have the highest probability (Thurston, Poiker & Moore 2003).
The purpose of this classification is to improve sampling efficiency. Therefore, majority of sample points are drawn from areas with high probability. Wise (2002, p. 6) points out that the purpose of this process is to “obtain the most useful data within the budget constraints and keep the coefficient of variance (CV) as low as possible.” CV is the “ratio of standard deviation to the mean; it is used when comparing datasets with different units or widely differing means” (Doaks 1997, p. 16).
From each stratum, grid points are selected, taken to a Microsoft SQL server database, and grouped consistently from north to south and east to west to guarantee optimum distribution of sample points across a given geographical area (Elangovan 2006). Thereafter, a stochastic starting point is selected and points picked at even intervals based on the number of points needed in each stratum.
The fourth step is carrying aerial survey of sample points, which obtains crop data. This means that, the crops planted in each sample point are determined and the process involves use of a light aircraft. “A tablet PC, connected to a GPS running on an ArcPad is used to capture this data” (Fourie 1999, p. 1).
This survey involves collecting information on the crops growing on each sample noting whether the crops grow under irrigation or on dry lands. Moreover, more information on specific natural conditions of the land is included in this survey for future references.
Finally, the field data “captured and stored in shapefile format” (Burrough 1998, 69). This information is then transferred to a central server and then to SQL server database (Soller, Berg & Wahl 2000). From here, the information can be used in statistical analysis to estimate cropland under grain cover.
Limitations Facing GIS in Biotechnology
There are numerous challenges facing proper utilization of GIS in this agricultural field. The preferred GIS model is complex and utilizes complicated strategies. For instance, “Combining several polygon networks by intersection and overlay is difficult and it uses considerable computer power” (Worboys & Duckham 2004).
Moreover, this process is time consuming and tedious and “simulation modelling of processes of spatial interaction over paths not defined by explicit topology is more difficult because each spatial entity has a different shape and form” (Maguire, Goodchild & Rhind 1997, p. 97). Additionally, this process faces other computational problems and finally it is expensive and may be not applicable in areas with poor economical background.
Other issues concerning application of GIS in biotechnology include the ethical concerns surrounding use of biotechnology. Many people have not come into terms with use of biotechnology and in as much as GIS is there for improvement of food production, it may be impossible to implement it in areas where biotechnology is not welcomed. On the other hand, GIS has played key role in promoting biotechnology and food security. It has become easier to study arable lands for improvement of production.
Conclusion
In contemporary times, GIS is becoming a common place across diverse fields. In biotechnology and agriculture in general, GIS has come with numerous advantages. With GIS, the South African government has successfully estimated cropland planted with grains to predict how these lands can sustain grain requirements in the country.
Biotechnology is then used to foster productivity of the surveyed lands. On its own, biotechnology would not sustain and assure food security in South Africa because there could not be estimates of what is required. The vector model of GIS has been used widely in these studies given its three dimensional nature of analyzing geographical settings.
Several processes are involved starting with obtaining satellite imagery, through digitalizing crop field boundaries and designing point frame to aerial survey and statistical analysis. However, this process is expensive and time consuming given the complex nature of vector data structures and the required resources.
References List
Berry, J 1993, “Beyond Mapping: Concepts, Algorithms, and Issues in GIS”, Fort Collins, CO: GIS World Books.
Burrough, A & Mcdonnell, R 1998, “Principles of Geographical Information Systems”, Oxford; Oxford University Press.
Doaks, J 1997, “Another Introduction to GIS, New York: Bogus Press.
Doe, J 1997, “An Introduction to GIS,” New York: Bogus Press.
Elangovan, K 2006 “GIS: Fundamentals, Applications and Implementations”, New Delhi New India Publishing Agency.
ESRI Data & Maps 2002, “GSI Models”, CA: Environmental Systems Research Institute.
Fourie, A 1999, “Better Crop Estimates In South Africa”, ESRI Press, 9-14.
Johnson, B 1998, “Digital Geologic Map Data Model. V. 4.3: AASG/USGS Data Model Working Group Report”. Web.
Lee, J & Kretzschmar, W 1993, “Spatial Analysis of Linguistic Data with GIS Functions”, International Journal of Geographical Information Science, 7(6); 541-560.
Maguire, D, Goodchild, F & Rhind, W 1997, “Geographic Information Systems: Principles, And Applications”, New York; Longman Publishers.
Smith, M, Goodchild, F & Longley, A 2007, “Geospatial Analysis: A Comprehensive Guide to Principles, Techniques and Software Tools”, International Journal of Geographical Information Science, 8(6); 64-78.
Soller, D 1999, “Proposed Guidelines for Inclusion of Digital Map Products in The National Geologic Map Database”, U.S. Geological Survey Open-File Report 99 (386); 35–38
Soller, D & Berg, T 2003, “The National Geologic Map Database Project: Overview And Progress”, U.S. Geological Survey Open-File, Report 03–471.
Soller, D, Berg, T & Wahl, R 2000, “Developing the National Geologic Map Database, Phase 3—An Online, “Living” Database Of Map Information”. Web.
Soller, D & Lindquist, T 2000 “Development And Public Review Of The Draft “Digital Cartographic Standard for Geologic Map Symbolization”. Web.
Tomlinson, R 2005, “Thinking About GIS: Geographic Information System Planning For Managers”, ESRI Press.
Thurston, J, Poiker, T & Moore, J 2003, “Integrated Geospatial Technologies: A Guide To GPS, GIS, And Data Logging”, Hoboken, New Jersey: Wiley.
Wise, S 2002, “GIS Basics”, London: Taylor & Francis Worboys, M &Duckham, M 2004, “GIS: A Computing Perspective”, Boca Raton: CRC Press.
Wyland, J 1999, “Agribusiness Grows With Crop-Specific Maps”, ESRI Press.