Prediction Of Climate Effect On Agriculture Using K-Means And Apriori Algorithm

Project and Seminar Material for Data Science

Prediction Of Climate Effect On Agriculture Using K-Means And Apriori Algorithm


Abstract


This study was designed to assess prediction of climate effect for agricultural production by providing valCRBle support for decision-making strategies in Cross River Basin (CRB). A rainfall data recorded over 24 well-distributed rain-gauge stations confined within CRB covering time period of 1980 to 2014 were used for regionalization of homogeneous rainfall pattern. Principal component analysis was used to classify the study area into different homogenous rainfall regions with seasonal rainfall characteristics and spatial continuity. The first four leading rotated principal components (RPCs) explained Belg rainfall variability from 25% to 80% and for kiremt ranges from 18% to53% of variance, respectively. According to the finding, three climatological sub-divisions were found in Cross River Basin. On the other hand from 1970 to 2014 years of daily rainfall data of eight selected stations from CRB were analyzed to identify variation of rainfall characteristics such as onset and cessation dates, probability of dry spell frequency, seasonal rainfall amount and its temporal as well as spatial distribution. Besides, the risk of dry spells with varying number of days was computed using first-order Markov chain model. The result indicates that zone three has a relatively higher dry spell risk as compared to the other zones. April is start of small rainy season while September is the cessation of main rainy season. The small rainy season has higher dry spell risk and onset variability than main rainy season. The major seasonal predictors for the study area were found to be the Nino regions and Pacific Ocean dipole phase (IOD). Due to variability of ENSO phases, rainfall amounts in the major rainy season were high in La Niña but low in El Niño years. The By using weighted average rainfall from homogenous zones and zonal crop yield data of major crops obtained for the periods of 1995 to 2013,the rainfall- crop relationship result generated from this study revealed that, excess rainfall would reduce crop yield in zone one while relatively better for zone three. Hence, use of prediction of climate effect is enable to maximize agricultural rain feed crop productivity while minimizing the crop risk associated with seasonal rainfall. Crop yield risk analysis compared to the ENSO-based approach using cumulative density function refers that, during normal ENSO phase has resulted in preferable yield value and less crop risk by first stochastic dominance and second stochastic dominance senses. Hence, Normal phase is found to have the best risk efficient set identified for each three Zones crop productions planning. Therefore, users in the study area can take comparisons based on alternate ENSO forecasts for further insight to crop risk planning and management strategies. This research therefore had given attention to the response farming activities with scientific prediction of climate effect information respect to ENSO phase. It is customary to note that prediction of climate effect provides the best opportunity for farmers to adjusting time of sowing date, cultivar selection in the rain feed crop production.


Table of Content


Chapter One:

Introduction

  • 1.1 Background of the Study
  • 1.2 Statement of the Problem
  • 1.3 Purpose of The Study
  • 1.4 Objectives of the Study
  • 1.5 Research Questions
  • 1.6 Significance of the Study
  • 1.7 Scope of the Study
  • 1.8 Limitation of the study
  • 1.9 Definition of Terms
  • 1.10 Organization of the Study

Chapter Two:

Review of Related Literature

  • 2.1 Conceptual Review
  • 2.2 Theoretical Review

Chapter Three:

Research Methodology

  • 3.1. Description of the Study Area
  • 3.2. Types and Sources of Data and Methods of Data Collection
  • 3.3. Methods of Data Analysis
  • 3.4. Methods of Prediction of climate effect
  • 3.5. Linear Relationship between Crop Yield with Seasonal Rainfall
  • 3.6. Cumulative Density Functions (CDFs) and Stochastic Dominance (SD) Analyses

Chapter Four:

Data Analysis and Result Presentation

  • 4.1 Data Presentation

Chapter Five:

Summary, Conclusion and Recommendations

  • 5.1. Summary and Conclusions
  • 5.2. Recommendations
  • REFERENCES
  • APPENDIX 

Chapter One


Introduction

1.1 Background of the Study

WMO (2005) stated that, prediction of climate effect systems should be considered for crop risk management and risk assessment approach. Prediction of climate effect is climatic pattern of forthcoming seasons and in particular the extent to which the expected climate differs from the climatological conditions. The distribution of possible outcomes is expected to match what has been observed in the past. However, the climate is often perturbed in predictable ways by various factors, which make these perturbations to the distribution of possible outcomes which are the basis for and target of seasonal forecasting (Stockdale et al., 2010). An important source of seasonal predictability comes from the El Niño-Southern Oscillation (ENSO), a quasi-regular variation in the atmosphere and ocean in the tropical eastern Pacific (Goddardet al., 2003).

The seasonal climate information that can be provided to reduce climate related crop risk or to assist in planning and decisions making in tropics is well known that rainfall is one of the most important climatic element for rain feed crop production (Baigorria, 2008). From an analysis of recent rainfall conditions, rain feed agricultural crop production has direct relationship with inter-seasonal to seasonal rainfall and it is the sectors most vulnerable to seasonal climate variability. The impact is even stronger in Africa, where rain feed agriculture is important for the daily subsistence, and where adaptive capacity is low (Nuhu et al., 2007; Cooper et al., 2006). Particularly, the seasonal climate variability is a major source of crop production risks (Harwood et al., 1999). Hence, prediction of climate effect has the potential benefits on the decision making process in agricultural crop production planning (Stern and Easterling, 1999; Jones et al., 2000; Hansen, 2002). In accordance with this, seasonal climate forecasting (SCF) can increase the confidence of the decision-makers and be perceived as having insurance value by a risk-averse manager (Washington et al., 2004). When this information must be made available and accessible at the right time and place and demands on crop risk management skill and adjusts possibly many interrelated decisions (Shin et al., 2009).

The seasonal climate forecast should capture the high-frequency modes of weather/climate variability (e.g., wet/dry spell sequences) properly to use it in a crop model for a reliable crop yield risk analysis, the yield risk forecast is based on location and ENSO-based seasonal climate (Fraisse et al., 2006). The frequency of rainfall within a season exhibits higher predictability than the seasonal total rainfall (Robertson et al., 2008). In Nigeria the degree of yield variability over time is changed not only by the amount of seasonal rainfall, but also by the pattern and frequency of the rainfall cycle (Adugna, 2005). Hence, Farmers are more concerned about within-season characteristics (onset, cessation, and likelihood and severity of mid-season dry spells) of rain than total seasonal rainfall, which will help farmers for plan properly in terms of timing of planting to avoids the crops reaching the critical stages at times when there are high probability of dry spells (Elijah et al., 2007). Reliable prediction of rainfall characteristics, especially onset date of the rain, is needed to determine a less risky planting date or planting method, or sowing of less risky types/varieties of crops in responsive farming (Stewart, 1991).

Annual variation of the onset, intensity, duration and cessation of the rainfall has negative impacts on socio-economic, agricultural crop and environmental development (Elijah et al., 2007). Particularly in areas where rainfall variability is frequently befallen during sowing and crop growing period, seasonal rainfall prediction is substantial for rain-fed agricultural activities and natural risk monitoring. Hence, the amount of water available for crops strongly depends on the rainy season’s onset, length, and cessation, temporal and spatial distribution that can directly indicate the climatic suitability of the crop and its chances of success or failure in a season. Therefore, local level based prediction of climate effect is very important for crop production and decreasing crop risk that are associated with seasonal climate.

The Cross River River rises from the high plateau of central Nigeria near Ginchi town, west of Cross River extends Koka dam and based on physical and socio-economic factors the CRB is part of lands where all digital elevation map are located about 1500 m above sea level (Taddese et al., 1998). The rainfall type of CRB is bimodal type one, which are two rainy season and one dry season, namely belg and kiremt are rainy season, while bega is dry season (NMSA, 1996). Based on FAO (1998) the dominant soil types are Cambisols and Vertisols. The chief agricultural crops of Cross River River Basin are cereal crops, pulse and oil seeds. Cereal crops are Teff, Barley, Wheat, Maize and Sorghum. Out of Oil seed Nueg is commonly grown and other crops accustomed in the region are Pulses such as broad Beans, Horse beans,

Chickpeas, Peas, lentils and Guaya, which are widely grown in rotation with grain crops and Oil seeds (CSA, 2014). Enset, a banana-like plant, is grown on the high land of almost all types of fertile soils of the basin.
There is a significant gap between the information needed to support farm decision making which are routinely available given by forecasting center at the study area. By using homogenous rainfall zones data and agricultural crop production data of major cereal crops of CRB, to attempts to show patterns of rainfall and provides insight into the preparation of an early warning system in the CRB. Therefore the concept of “response farming” can be apply according to seasonal climate outlooks to assists farmers to use scientific prediction of climate effect for improving their traditional farming practices.

In different pocket areas of CRB, rainfall related crop risks like prolonged dry spells, wet spells, untimely rainfall, rainfall variability, flash flood in plain areas and outbreak of crop pests are the challenges which affect crop production. Delay in sowing period due to rainfall variability can cause severe reduction in yield or entirely expose to loss of crop production. This is mainly because of inexistence of optional crops to sow on their farm land after the sowing period of Teff has been elapsed. Therefore the derived rainfall regimes and local level prediction of climate effect would have useful in the planning and management of rainfall dependent agricultural crop planning activities in the sub basin. This thesis research involves activities according to seasonal rainfall performance and reducing climate variability impacts on crop risk to improve crop yield production. Hence, this study has a general objective of characterizing and predicting seasonal rainfall for supporting rain-fed crop production risk management. Specifically, the research study was conducted with the following objectives:


1.2 Statement of the Problem

Higher temperatures and declining rainfall patterns, as well as increasing frequency of extreme climate events (such as droughts and floods), are the expected future climate in the tropics (IPCC, 2007; Mitchell & Tanner, 2006; IPCC, 2001). In southern Africa, for example, rainfall patterns show a declining trend of summer rainfall (about 20%) from 1950-1999 and a high frequency of droughts, predicted to intensify in the 21st century (Mitchell & Tanner, 2006). Predictions for 2050 by the US National Center for Atmospheric Research show that the declining trend in rainfall is set to continue and the region is expected to be 10−20 percent drier than the previous 50 years (Mitchell & Tanner, 2006). These predicted changes in climate are expected to have differential impacts on agricultural productivity, food security and other sectors, across spatial and temporal scales. In the tropics and Africa in particular, changes in climate are expected to be detrimental to agricultural livelihood (IPCC, 2007; IAC, 2004; Dixon, Gulliver & Gibbon, 2001; IPCC, 2001). Recent studies suggest that agricultural crop productivity in Africa will be adversely affected by any warming above current levels (Kurukulasuriya et al., 2006; Kurukulasuriya & Mendelsohn, 2007a; Seo & Mendelsohn, 2007a). Local ecosystems provide the main source of livelihood for many of the world’s poor. Most of the rural poor in sub-Saharan Africa rely for their livelihood and food security on highly climate-sensitive rain-fed subsistence or small-scale farming, pastoral herding and direct harvesting of natural services of ecosystems such as forests and wetlands (Mitchell & Tanner, 2006; Leary et al., 2005; Roach, 2005; IPCC, 2001; Kandlinkar & Risbey, 2000). The productivity of this livelihood base is highly vulnerable to climate-related stresses, such as changes in temperature, precipitation (both amount and variability), and increased frequency of droughts and floods. The vulnerability of the majority of the poor in Africa to climate-related stresses is worsened by widespread poverty, HIV/AIDS, lack of access to resources (e.g. land and water) and management capabilities, wealth, 2 technology, education, ineffective institutional arrangements, and lack of social safety nets (Leary et al., 2005; Nyong, 2005; APN, 2002; IPCC, 2001). Studies based on the Global Environmental Facility (GEF) African Climate Project estimated the economic impacts of climate change on African agriculture (Dinar, Hassan, Mendelsohn & Benhin, 2008). These studies however, analysed impacts on dryland crops, irrigated crops and livestock separately. This is a significant limitation, since factors affecting the choice between crop and livestock production or their combination (mixed systems), cannot be separated. The selection must be an endogenous decision made by agricultural producers in response to varying climates and other circumstances. The decision of what to produce and how to produce it is accordingly an important adaptation mechanism in the face of changing climate and other ecological and economic circumstances. This is of special importance to Africa, where the majority of poor smallscale farmers practice mixed crop−livestock agriculture and few depend on crops or livestock alone. One main objective of this study is therefore to measure the aggregate impact of climate change on income from all agricultural production systems (crop, livestock and mixed) in Africa, and to predict future impacts under various climate scenarios. In addition, the study also measures and compares impacts on specialised crop and livestock farms.

The results are contrasted with findings of other regional studies using the same data but generating different climate response functions for crop and livestock farming separately (Kurukulasuriya et al., 2006; Kurukulasuriya & Mendelsohn, 2007a; Seo & Mendelsohn, 2007a). Climate is changing and mitigation efforts to reduce the sources or enhance the sinks of greenhouse gases will take time and may also be very expensive (Stern, 2006). Empirical studies measuring the economic impacts of climate change on agriculture in Africa (Kurukulasuriya & Mendelsohn, 2007a; Seo & Mendelsohn, 2007a; Mano & Nhemachena, 2007; Benhin, 2006; Kabubo-Mariara & Karanja, 2007) showed that such impacts can be significantly reduced through adaptation. Adaptation is therefore critical 3 and of major concern in developing countries which are most vulnerable, particularly Africa. While African farmers have low capacity to adapt to such risks, they have survived and have coped with climate change in various ways over time and under changing circumstances (Kurukulasuriya & Rosenthal, 2003). The second objective of this study is to analyse adaptation measures used by African farmers and determinants of their choices. Better understanding of how farmers have coped with and adapted to climate change is essential for designing incentives to enhance private adaptation. This is also true for public adaptation as better understanding will help governments to design programmes to help farmers adapt. Supporting the coping strategies of local farmers through appropriate public policy, investment and collective actions has the potential to facilitate increased adoption of adaptation measures. Such adoption will reduce the negative consequences of predicted future climate changes, with great benefits to vulnerable rural communities in Africa. Our analysis is different from other adaptation studies in that we consider climate prediction on agriculture, compared to the analysis conducted by Maddison (2007) on the same sample of African farmers, which is based on farmers’ perceived adaptations.


1.3 Purpose of the Study

The purpose of the study is to assess prediction of climate effect on agriculture using k-means and apriori algorithm


1.4 Objectives of the Study

  1. To identify homogenous seasonal rainfall regime
  2. To characterize seasonal climate pattern at local level;
  3. To examine local level seasonal rainfall predictions for crop yield risk management;
  4. To identify seasonal crop planning decision supporting strategies in the study area.

1.5 Research Questions

  1. What are the homogenous seasonal rainfall regime?
  2. What are characteristics of seasonal climate pattern at local level?
  3. What are the local level seasonal rainfall predictions for crop yield risk management;
  4. What are the seasonal crop planning decision supporting strategies in the study area?

1.6 Significance of the Study

The problem associated with climate change is a major concern to farmers across the globe and the society at large.

The findings of this study will be beneficial to farmers, geographers, government and nongovernmental organizations. It will help enlighten farmers on the influence of climate prediction on crop yield. This will also help for a crop production and living conditions for farmers.

This study will equip the geographers with information that will help them educate farmers on climate change issues. Government will also use the information from this study to guide farmers as regarding their crop planting rotations. This will thus bring to an end their idea of low yield seen across the globe.

1.7 Scope of the Study
The research covered activities in cross river basin specifically on the prediction of climate effect on agriculture using rain fall and crop yield as variables


1.8 Limitation of the Study

There is no study undertaken by any researcher that is perfect. As such this study may not be perfect and subject to the efficiency of the data analysed.


1.9 Definition of Terms

Climate change refers to long-term shifts in temperatures and weather patterns. These shifts may be natural, such as through variations in the solar cycle. But since the 1800s, human activities have been the main driver of climate change, primarily due to burning fossil fuels like coal, oil and gas.

Climate prediction are inherently probabilistic statements about the future climate conditions on timescales ranging from seasons to decades or longer, and on spatial scales ranging from local to regional and global.


1.10 Organization of the Study

This study is divided into five chapters. Chapter one is introduction which consists of the background to the study, statement of problem, research questions, research hypotheses, objectives of the study, the significance of the study, the scope and limitations of the study and finally the organization of the study. Chapter two deals with the literature review which consists of the conceptual literature, theoretical literature, empirical literature, theoretical framework. Chapter three gives the research methodology including research design, population of study, sample size, sampling technique, method of data collection, instrument of data analysis, method of data analysis, validity/reliability of instrument. Chapter four is presentation and analysis of data, discussion of findings. Chapter five gives the summary, conclusion and recommendations.


Chapter Five


Summary, Conclusions and Recommendations

5.1. Summary and Conclusions

This study was carried out at Cross River Basin of southern Nigeria, which has high potential for varieties of crop production. Basically, the study examined characterizing seasonal climate and predicting for better decision supportive strategies for crop production and planning. To determine local climatic patterns and establish better climate forecast method, the principal component analysis has been applied, from which the CRB climate was classified into three homogenous rainfall zones. The onset of the growing season is commonly in the month of April with highly variable and dry spell risk, while cessation is in the month of September with less variance have been found. The number of rainy days of long year observation has no statistically significant change in kiremt season. The local surface temperature analysis has shown that, there is statistically significant change of increasing annual trends of temperature in the study area.

According to the result output established between zonal rainfall index and global sea surface temperature correlations, which are mostly manifested as ENSO phenomena and IOD phase. It has been found that, SST anomalies of the Nino regions and IOD phase have relatively strong correlated with all zonal rainfall indices during belg season, while Nino regions, cross equatorial flow and the associated monsoon systems are the mostly identified as rain bearing system during kiremt season. IOD phase has more significant signal to belg rainy season rather than Kiremt season. Hence, seasonal climate pattern in CRB is affected by the El Ni˜ no Southern Oscillation (ENSO) phases and there is a close relationship between the increase and decrease of rainfall depending upon the warm or cold phases of the phenomenon. Therefore the variability of rainfall patterns during the major and small rainy seasons as well as annual totals rainfall at CRB were strongly linked to ENSO phases. The rainfall patterns during the major seasons shown that, there is different risk level during different ENSO phases at CRB. For instance, the results revealed that rainfall amounts of major rainy season became very high in La Niña but low in El Niño years. Conversely, during small rainy season, comparatively high rainfall amounts were observed when El Niño was apparently established. There is enough rainfall amounts prevail during non-ENSO years over most parts of Cross River Basin.

According to the tools of crop risk management resulted on crop yield risk analysis, the stochastic dominance described the weight of risk on crop yield in comparing alternative cropping systems on the risks producers of different ENSO phase. Teff, maize, wheat and sorghum cereal crops risk analysis showed that, crop yield has good performance and less risk during non-ENSO years than El Niño and La Niña phases.

Therefore, this study revealed that there are a lot of techniques to reduce the challenges on the effective use of seasonal climate information for various cropping practices. This has been manifested as a lack of knowledge in climate information; lack of specific local climate information and lack of knowledge about climate variability impacts and the associated decision responses. Users can get a clear indication of forecast reliability and accuracy developed in skillful model prediction and the timely provision. Hence, decision makers can take comparisons based on alternate ENSO forecasts for further insight to crop risk planning and management strategy. Therefore using prediction of climate effect that focus on managing seasonal climate variability is an important means of preparing for the near future of climate change.


5.2. Recommendations

The progress of relevant observing systems, including many of the ocean and land components of the Global Climate Observing System (GCOS), is crucial for improving seasonal prediction. Thus, statistical multiple linear regression models(SMLR) and forecast systems across a range of time scales are also good practice in addition to pre-seasonal ocean SST and surface wind flow which are valCRBle information in order to identify the analogue years from past years history. However, due to presence of climate change and variability, climatological (long term) observations alone are no longer sufficient for seasonal rainfall prediction for crop production planning in the future. Therefore, due to the presence of climate change and variability, seasonal predictions based on GCOS, SMLR and pre-seasonal ocean SST and analogue year approaches alone are not sufficient. Hence, different sophisticated computer models like super computer and dynamic atmospheric circulation models needs to be considered while predicting seasonal rainfall.

A better prediction of climate effect service for decision-making must be delivered by the National Meteorological Agency not only at national level but also at stations level for local users. Such information are farming practices that allow timely preparation of the land and planting of long and short cycle dry spell resistant crops and effective use of rainfall received during the April to May season are more essential. In the belg season due to existence of dry spell risk and onset variance as well as it’s unevenly and erratic distribution over time makes difficult to decide annual regular crop planting period starting from April. Therefore, Kiremt rainy season is more advisable strategic plan for regular rain-fed crop production planning in Cross River Basin.

A stochastic dominance analysis using cumulative density function (CDF) of crop yield based on different ENSO phase has identified the treatments that would be preferred by individuals within a range of preferences. The outcome of CDF of yield has a preferred on normal ENSO phase than La Niña and El Niño. Therefore, users in the study area can take comparisons based on alternate ENSO forecasts for further perception to crop risk planning and management strategies. Hence, using CDF of crop yield analysis is recommended as tools for crop yield risk management.


Project Material Download

6,000 Naira

The Complete Material Will Be Sent to You in Just 2 Steps

Quick & Simple…


Step One Purchase

Make Payment (Through Transfer) of ₦6,500 to Any of the Account Below

Access Bank PlcAcc No: 0811003731
Samphina Academy
Current Account
Zenith BankAcc No: 1225513212
Samphina Academy
Current Account

Or CLICK HERE To Pay With Debit Card


FOR STUDENTS OUTSIDE NIGERIA
CLICK HERE To Purchase Material ($25)
FOR GHANIAN STUDENTS
Make Payment of 200 GHS to 0553978005 | Douglas Cloud Osabutey | MTN MoMo

Step Two Purchase

Send the Following Details on WhatsApp ( 08143831497) After Payment

  1. Payment Details

  2. TOPIC: Prediction Of Climate Effect On Agriculture Using K-Means And Apriori Algorithm

The Complete Material Will Be Sent To You On WhatsApp After Receiving Your Details
T & C Apply


  Contact Our Help Desk


Need a Different Topic? Perform a Quick Search



List of Related Works

Click on Any Topic to Preview the Content

samphina.academy

Samphina Academy

Samphina Academy is an Online Educational Resource Center that is aimed at providing students with quality information and materials to aid them in succeeding in their academic pursuit.