A Case Study to develop and test GIS/SDSS methods to assess the production capacity of a Cocoa Site in Trinidad and Tobago

Abstract

Pritam Kumarsingh

A Case Study to develop and test GIS/SDSS methods to assess the production capacity of a Cocoa Site in Trinidad and Tobago

Trinidad and Tobago has achieved world-wide affirmation for its unique fine and flavour cocoa (Theobroma cacao). This classification of cocoa commands price premiums on the global market. However, the industry has declined over years. Investors and farmers are concerned about their rate of return, cost of operations and consistency of production. A potential remedy for reversal of this decline is understanding and managing the complex interrelationship among site specific characteristics and accurate analyses of their influence on production. Assessment of how site conditions impact production, enable stakeholders to reliably forecast revenue streams and make informed decisions on crop management and risk mitigation. Traditional evaluation methods are inadequate. Technology-based, Spatial Decision Support System (GIS/SDSS) supported by Geographic Information System provides a sufficiently resourced platform which enables efficient and effective decision making despite the complexity of analyzing the contributing factors. This research project utilizes soil properties and climatic data to develop and test GIS/SDSS methods to forecast production capacity of a traditional cocoa production site consisting of 4,458 hectares in Gran Couva, Trinidad and Tobago.
The GIS/SDSS methods incorporate computational methods from various expert sources to determine the level of influence of each factor on cocoa production. The research is based on the chemical and physical properties of the soil. These are point based, site specific data layers and include pH, Cation Exchange Capacity (CEC), texture and drainage. The Length of the Growing Period (LGP) is used as the climate factor for comparison of evapotranspiration rates with rainfall. These data points are converted to continuous surfaces through interpolation. A Multi Criteria Analysis (MCA) using a weighting mechanism is used to ascertain the level of influence of each factor. The assigned weights of each factor are tested for consistency using Analytic Hierarchy Process (AHP) and Sensitivity Analysis. Additionally, Fuzzy Values are derived for the attributes of each factor to determine realistic production values over the site. The productive capacity is computed using mathematical relationships categorized into geographic classifications where each class depicts the level of production using GIS as the platform.
The results show that site suitability for cocoa production is divided into moderate and high production classes. 45.8% of the site is classed as moderate production and 46.1% is high production and 8.2% of zero production which suggest that the site, given its current conditions can potentially produce 2,700Mt annually. The results also demonstrate the need for crop management to include adaptation to the effects of climate change in the future. As temperature increases and rainfall decreases, LGP is a critical component of production. This requires that stakeholders develop policy for mitigation as part of the overall crop management.

Keywords: Geography, GIS; MCA; Fuzzy; AHP; Cocoa; SDSS; Crop; Management; Climate; Soil; Agriculture; Production; LGP
Advisor: Cecilia Akselsson
Master degree project 30 credits in Geographical Information Sciences, 2021
Department of Physical Geography and Ecosystem Science, Lund University
Thesis nr 132

Popular Abstract

Pritam Kumarsingh

A GIS/SDSS approach to assess cocoa production
Cocoa production requires an understanding of the complex interrelationship of the climate and soil attributes of a site. A combined Geographic Information System and Spatial Decision Support System (GIS/SDSS) model can assist the farmers and investors in their decision making for crop management, mitigation and cost/ benefit analyses. This research develops and tests methods for this application to an existing cocoa site consisting or 4,458ha located in Gran Couva, Trinidad and Tobago.
Development of the GIS/SDSS is based on experts’ judgement which applies weights to each factor soil and climate factors using a Multi Criteria Analysis (MCA). These weights do have a level of subjectivity which is measured by using the Analytic Hierarchy Process (AHP) method and a sensitivity analysis to determine the consistency and accuracy among the weights before they are accepted. The next stage is to derive continuous surfaces through a process of interpolations of each datasets. This is a process which converts point data into a continuous surface giving values across the site based on the distance from the points. Fuzzy values are applied to the attributes on a scale between 0 and 1. The value gives the level of influence on production. 0 is the value of no production and 1 is optimum production. This allows the model to consider those geographical areas which may not be ideal for optimum production, but where some production is still possible as is the case in the real world. Climate data are combined to obtain the Length of the Growing Period (LGP) which is calculated from a Food and Agriculture Organization (FAO) standard equation. This compares evapotranspiration with rainfall. Land use such as roads, protected areas and rivers are removed from the surface since no production is possible in those areas and are assigned a 0 value. The last stage is to quantify and classify the production values using GIS computational methods into low, moderate and high production levels. The model also removed slopes exceeding 40% since these areas were considered as necessary for erosion control.
The results show that the site chosen has 45.8% moderate production and 46% high production with 8.2% are areas with no production. Estimating the yield for moderate value at 400kg/ha and high production at 1Mt/ha, the site estimated productive capacity is a total of 2,700 Mt. A 10% decrease in LGP shows that the unfavourable areas increases from 25.5% to 58.4% while favourable areas decrease from 32% to 5.7% and moderate areas from 42.5% to 35.9%.
An efficient approach to a complex issue
The factors affecting production can either be considered separately or collectively. The level of influence of the factors can give priority to the crop management policies to improve production. These can include location specific fertilization and irrigation. The data can be manipulated and simulated with different scenarios to determine outcomes which were traditionally onerous for farmers and investors. They can also use additional data to refine the model. Modeling can assess risks such as forest fires, pests and diseases which are also critical to cocoa production. The stakeholders in the cocoa industry can therefore assess the cost of operations and revenue generation in a more efficient manner and make informed decisions.


Keywords: Geography, GIS; MCA; Fuzzy; AHP; Cocoa; SDSS; Crop; Management; Climate; Soil; Agriculture; Production; LGP
Advisor: Cecilia Akselsson
Master degree project 30 credits in Geographical Information Sciences, 2021
Original Title: A Case Study to develop and test GIS/SDSS methods to assess the production capacity of a Cocoa Site in Trinidad and Tobago
Department of Physical Geography and Ecosystem Science, Lund University
Thesis nr 132

• author: Kumarsingh, Pritam
• supervisor: Cecilia Akselsson
• organization: Dept of Physical Geography and Ecosystem Science
• course: GISM01 20211
• year: 2021
• type: H2 - Master's Degree (Two Years)
• subject: Earth and Environmental Sciences
• keywords: geography, GIS, MCA, Fuzzy, AHP, cocoa, SDSS, crop, management, climate, soil, agriculture, production, LGP, Trinidad, Tobago
• publication/series: Master Thesis in Geographical Information Science
• report number: 132
• language: English
• id: 9043596
• date added to LUP: 2021-05-04 21:04:20
• date last changed: 2021-05-04 21:04:20