CIGR Ejournal Style and Format Guidelines ARID ZONE JOURNAL OF ENGINEERING, TECHNOLOGY & ENVIRONMENT AZOJETE - CIGR Section VI Special Issue: Innovation & Technologies for Sustainable Agricultural Production & Food Sufficiency AZOJETE, December, 2018. Vol. 14(SP.i4): 111-120 Published by the Faculty of Engineering, University of Maidiguri, Maidiguri, Nigeria. Print ISSN: 1596-2490, Electronic ISSN: 2545-5818 www.azojete.com.ng ____________________ *Corresponding Author Email: lawalbalo@yahoo.co.uk 111 ORIGINAL RESEARCH ARTICLE DEVELOPMENT OF CROP WATER PRODUCTION MODEL IN A RAINFED TROPICAL MAIZE CROP CULTIVATION L. A. Balogun*1, T. A. Ewemoje2, M. A. Aderemi3, A. R. Nasirudeen1, D. Lasisi1 and A. O. Hammed4 (1Department of Agricultural and Bio-Environmental Engineering, Oyo State College of Agriculture and Technology, Igboora; 2Department of Agricultural and Environmental Engineering, University of Ibadan, Ibadan. 3Department of Agricultural Technology, Federal College of Forestry, Ibadan. 4Department of Agricultural Engineering, Federal College of Agriculture, Ibadan.) ARTICLE INFORMATION Received October, 2018 Accepted December, 2018 Keywords: Rain-fed Maize Water Yield Secondary data Regression Nigeria ABSTRACT Crop production, which is one of the sources of food for human and animal sustenance is a function of the availability of adequate quantity of water. The rainfall being seasonal is the main source of water for agricultural production in Nigeria. Maize production is majorly through rainfed agriculture in Nigeria and the irregularity of which affect the yield. This research work was to develop crop-water production model in tropical rain fed maize crop cultivation using maize yield and rainfall data from Oyo state. One of the major problems of food production in attempting to determine the relative future roles of irrigated and rain-fed agriculture is the lack of sufficient ground and accurate tool on a localized basis. Hence, the essence of this research works. Using correlation and full quadratic regression analysis, the effects of some rainfall indices (monthly and annual rainfall, raindays, and rainfall onset and rainfall cessation) on maize yield in Oyo State were examined. The results of the correlation statistics showed that cessation has the strongest association (r= - 0.284) with maize yield. The analysis also showed that early maize and late maize suffer moisture deficiency in March and November respectively while excessive rainfall of June/July and September also have implication for maize yield. It was also observed that the rainfall characteristics jointly contributed 96.7% in explaining the variations in the yield of maize per hectare. Conclusively, a model was development for predicting maize yield in Oyo State. The study also recommends the use of state own yield so as to harmonize with state rainfall data, application of appropriate moisture conservation management practices that ensured efficient use of water and use of drought resistance crop species with shorter growing periods as adaptive measures to the changing rainfall pattern within the study area. ©2018 Faculty of Engineering, University of Maiduguri, Nigeria. All rights reserve 1.0 Introduction Crop production, which is one of the sources of food for human and animal sustenance is a function of the availability of adequate quantity of water. The water, natural is to be gotten directly from rainfall, but this source of water is also climatic dependent. This water is the main factor for any crop http://www.azojete.com.ng mailto:lawalbalo@yahoo.co.uk Balogun, et al. Development of crop water production model in a rainfed tropical maize crop cultivation. AZOJETE, 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818, www.azojete.com.ng 112 production on which other factors may depend. The rainfall being seasonal is the main source of water for agricultural production in Nigeria. This has restricted production of crop to only raining season during which the intensity of the events will still affect yield of crop production to some extent. In an interview with the director general of NIMET, Prof Anthony Anforom held on an NTA program ”INSIGHT” in August, 2015, the director made it known that as the date, the production of crop in Nigeria is limited to only rain-fed agriculture. That this has affected the crop production for the year, because of the late onset of rainfall and low rainfall as forecasted in the previous year. This, in general, limits the production of food in some areas that depended on rainfall for production and with limited amount of total rainfall per year. The vast potential of rain-fed agriculture needs to be unlocked through knowledge-based management of natural resources for increasing productivity and income to achieve food security in the developing world. Soil and water management play a very critical role in increasing agricultural productivity in rain-fed areas in the fragile SAT (Semi-arid Tropics) systems (Wani et al., 2009). Maize crop is one of the generally grown cereal crop in the world. Maize, the main source of food in the tropical region, is grown in different varieties. The grains are rich in vitamins A, C and E, carbohydrates, and essential minerals, and contain 9% protein. They are also rich in dietary fiber and calories, which is a good source of energy (IITA, 2015). Crop production as stated earlier is limited to rainfall in some developing country and maize production is not an exception. Maize production is majorly through rain-fed agriculture in Nigeria and the irregularity of which affect the yield. International Institute of Tropical Agriculture and other research institute scientists have developed high yielding, drought-resistant and disease-resistant varieties that are adaptable to sub-Saharan Africa's various agro-ecological zones but many of which have not been adopted by the local farmers. Therefore this research work aimed at developing crop-water production model in tropical rain fed maize crop cultivation, with focus on rainfall dataset for Oyo state between 1990 and 2013 periods and maize yield for the same period to determine rainfall indices for the years. One of the major problems of food production in attempting to determine the relative future roles of irrigated and rain-fed agriculture is the lack of sufficient ground and accurate tool on a localized basis. The variability and uncertainty of the climatic condition have drastic effect (low yield) on the crop production, the study of this variability and effect on the production will help for future planning to avert the effect of climatic change, hence the essence of this research work. (a) (b) Fig 1. (a) Photograph of maize plantation, and (b) Harvested maize http://www.azojete.com.ng Arid Zone Journal of Engineering, Technology and Environment, December, 2018; Vol. 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818; www.azojete.com.ng 113 2.0 Material and Methods 2.1 The Location of the Study Area Nigeria is located between latitudes 4° and 14°North of the equator and longitudes 3° and 14° East of Greenwich Meridian, and in West African region. The country is of area of 923768 sq kilometer, with tropical climatic condition and annual precipitation on the average ranging from 1800mm in the west to 4300mm in the east and 1300mm inland. The country experiences double rainfall maxima from March to July and September to October. Ibadan city where Nigeria Meteorological agency (NIMET) resides is located in coordinates of 7°23’47”N and 3°55’0” E and with mean total rainfall of 1420.06mm, falling in approximately 109 days and mean maximum temperature of 26.46°C (Wikipedia, 2015). 2.2 Sources of Climatic and Crop Yield Data In order to develop the crop-water production model for productivity of water for rain-fed maize crop, weather data comprising of rainfall indices like dates of onset and end of rainy season, annual amount, temperature, duration and rain days, and crop yield are needed. Data on daily rainfall (from which monthly and annual values were derived) was collected over 24 years (1990- 2013) from the official records of the Meteorological Centre of NIMET, Iseyin. The choice of this length of time is in line with the available data for the study, which is in accordance with numbers years for weather data in characterizing the climate of an area, as adopted by the World Meteorological Organization. The data used for this study were archival data on rainfall (in millimeters) and maize yield (in kilograms/hectare). The data on annual maize yield (kg) was collected from the official FAO database for Nigeria and for the same number of years 24 (1990 - 2013) 2.3 Derivation of Other Parameters From the rainfall data collected, the following parameters were determined: Date(s) of onset of the rainy season (in days); Date(s) of end of the rainy season (in days); Duration of the rainy season (in days);Annual number of rain days (in days); and, Annual rainfall amounts (in mm). There are several methods of computing onset, end and duration of the rains such as used by Ilesanmi (1972) and Benoit (1977). However, Walter’s formulation as modified by Olaniran (1983) was adopted in this study because of its high prediction reliability among other methods as used by Ifabiyi and Omoyosoye, (2011); Emmanuel and Fanan, (2013). The method was expressed as: (1) Where: DM = the number of days in the month containing the date of Onset/End; A = the accumulated total rainfall of the previous months; TM = the total rainfall for the month in which 51 mm or more is reached and 51 mm = the threshold of rainfall for both Onset/End month. Where such onset date was followed by dry spells of up to 14 days, the next rain day date that is not followed by such dry spells was chosen. For computing Cessation or End dates, the formula was applied in reverse order by cumulating the total rainfall backwards from December. Duration of the rainy season was derived by counting the number of rain days between the onset-date and the end- date of the rains in a given year. A rain day is a period of 24 hours (10:00 am - 10:00 am local time) ../../../user/Downloads/azojete143/www.azojete.com.ng Balogun, et al. Development of crop water production model in a rainfed tropical maize crop cultivation. AZOJETE, 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818, www.azojete.com.ng 114 with at least 0.3 mm of recorded rainfall amount. Annual rainfall total is the amount recorded for an entire year at a particular place (Emmanuel and Fannan, 2013). 2.4 Suitability of study area for maize production The monthly water consumptive use of maize within its growing season as given by Lema (1978) were compared with average monthly rainfall obtainable in the study area during the growing season to bring out the suitability of the study area for maize production. This is as shown in the Table 1. Table 1: Monthly water consumptive use of maize within the growing season month 1st 2nd 3th 4th 5th Total Consumptive use (mm) 100 90 96 75 65 426 2.5 Model Formulation The study employed the mean, standard deviation and coefficient of variability in analyzing the variations in the study variables. Partial correlation and multiple linear regression analysis were the statistical tools used to establish the relationship and effect of rainfall characteristics on maize yield. The regression model (as used by Emmanuel and Fannan, 2013) for the study was computed as: Where: Y = the value of the dependent variable (maize yield/ ha); b0= Y intercept b1, b2, b3, b4, ··· bn= regression coefficients (each b represents the amount of change in Y (Maize yield/ha) for one unit of change in the corresponding x-value when the other x values are held constant; x1, x2, x3, x4, ··· xn= the independent variables (i.e. rainfall onset, cessation, duration, annual totals and annual number of rain days respectively); and e = the error of estimate or residuals of the regression. Apart from the coefficients of the independent variables (rainfall characteristics), coefficient of multiple determinations (R2) was used to determine the percentage explanation achieved jointly by the rainfall characteristics. This method is preferred since it gives the best linear and unbiased estimates among other estimators. Several authors to effectively study the impact of climate on crop yield (Emmanuel and Fannan, 2013) have used this. 3. Results and Discussion 3.1 Variability in Rainfall Characteristics The descriptive statistics of the rainfall characteristics are shown in the tables below. Table 2 shows the limit statistics of rainfall indices of the area of study and this indicates earliest onset of rainfall as 27th January 1997, latest onset of rainfall as 11th April 2007. Mean onset of Rainfall as 18th March http://www.azojete.com.ng Arid Zone Journal of Engineering, Technology and Environment, December, 2018; Vol. 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818; www.azojete.com.ng 115 and earliest cessation of rainfall as 5th October 2001, while latest cessation of rainfall as 25th December 1999 and mean ceasation of rainfall as 28th October 1999. Table 3 shows the limit statistics of other rainfall indices of the study area. Table 3 shows minimum annual amount of rainfall, number of rain-days, annual duration of rainfall, and maize yield as 909 mm (in 2001), 78 days ( in 2006 ), 82 days ( in 2006) and 11181.31 Hg/ha (in 1992) respectively. From table 3, the maximum annual amount of rainfall, number of rain-days, annual duration of rainfall, and maize yield as 1596.4 mm (in 1991), 119 days ( in 1993 ), 122 days ( in 1999) and 21961.28 Hg/ha (in 2009) respectively. The mean annual amount of rainfall, number of rain-days, annual duration of rainfall, and maize yield as 1212.72 mm, 94 days, 99 days and 14612.25 Hg/ha respectively. The result indicated that there was a delay in the dates of onset accompanied by early cessation dates of the rainy season, shortening duration of the growing season and number of rain days, consequently declining maize yield per hectare even though high rainfall amount was received. It was also be observed that the lowest rain-days was 78 days in 2006 and lowest annual rain-days ( 82 days) in the same year which might due to the late onset ( 12th march) and earlier cessation (25th October) of rainfall in the year. However, maize was not the lowest due to high rainfall (1169 mm) in the year. It can also be observed that 1992 recorded the lowest yield of maize per hectare (11181.30 Hg/ha), probably due to the latest onset date (12th April), earliest cessation date (30th October), short duration (92 days), low number of rain-days (88 days) and annual rainfall amount (1047.5 mm) that occurred in the same year. The lowest annual amount of rainfall (909 mm) in 2001 and shorter annual duration (91 days) might be due to earliest cessation (5th October) of rainfall in the year. The highest amount of rain-days (119 days) and maximum annual duration (122 days) of rainfall and higher annual amount of rainfall (1447 mm) were recorded in 1993 which might be as result of latest cessation (25th December) in the same year but lower maize yield (11847.81 Hg/ha). Table 2. Limit statistics of onset and cessation of rainfall at the study area Indices Mean Earliest Day Year Latest Day Year Onset 79 27 1997 103 2007 Date/Days 18-Mar 27-Jan _ 11-Apr _ Ceasation 303 280 2001 361 1993 Date/Days 28-Oct 5-Oct _ 25-Dec _ Table 3. Limit statistics of rainfall indices and maize yield at the study area Indices Mean Minimum Year Maximum Year Annual Amount 1212.72 909 2001 1596.4 1991 Rain-days 94 78 2006 119 1993 Annual Duration 99 82 2006 122 1993/99 Maize Yield 14612.25 11181.31 1992 21961.28 2009 3.2 The Coefficient of Variability Table 4 shows the descriptive statistics of rainfall indices and maize yield. The coefficient of variability of the rainfall characteristics shows that cessation dates of the rainy season has the highest coefficient of variability (56.1%), followed by dates of onset ( 29.11%), annual rainfall ../../../user/Downloads/azojete143/www.azojete.com.ng Balogun, et al. Development of crop water production model in a rainfed tropical maize crop cultivation. AZOJETE, 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818, www.azojete.com.ng 116 amount (13.82%), duration of the rainy season (12.77%) and the least variability of 12.12% was found in annual rain-days of the rainy season. Table 4. Descriptive statistics of rainfall indices and maize yield at the study area Indices Mean Standard Deviation Coefficient of variance (%) Sample Variance Maize Yield(Hg/Ha) 15825.33 2730.03 17.25 7453088.84 Onset(Days) 79 23 29.11 539 Cessation(Days) 303 17 56.10 290 Annual Amount(Mm) 1212.72 167.69 13.82 28116.81 Rain-days(Days) 94 12 12.77 142 Annual Duration(Days) 99 12 12.12 140 On a general note however, the coefficient of variability of (17.25%) was recorded in maize yield per hectare (Table 4). This result means that the annual rainfall durations were more reliable and predictable whereas the dates of cessation were more unreliable and un- predictable in the study area. The highest coefficient of variability of Maize could be as a result of the joint effect of the variability in all the rainfall characteristics studied and other climatic and non-climatic factors not directly considered in this study. 3.3 Suitability of the study area for maize production The analysis of maize consumptive water use is also presented in Table 5. The mean rainfall amount obtainable in March according to Table 5 is 52.0mm. This amount fell short of the 100 mm that is required at the first set of planting. Thus, early maize plants in March may experience inadequate watering due to the nature of onset. Table 5.Mean monthly pattern of rainfall for Oyo State (1990 – 2013) Months Rainfall(mm) Jan 10 Feb 18.2 Mar 52 Apr 106.3 May 150.8 Jun 165.8 Jul 152.8 Aug 147.8 Sep 198.9 Oct 170.2 Nov 21.5 Dec 19.5 Mean(x) 101.2 http://www.azojete.com.ng Arid Zone Journal of Engineering, Technology and Environment, December, 2018; Vol. 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818; www.azojete.com.ng 117 A critical period during the planting season of maize is the month of November; as the minimum required (65mm) looks almost impossible to be met. In November the obtainable rainfall is only 21.5mm. This indicates a need of irrigation for the period for the purpose of preventing drought to the planted crop. The total consumptive use of 426mm of rainfall for maize production can be met in Oyo state as both rainfall obtainable for early maize and late maize were 627.7 mm and 710.7 mm respectively are more higher than the water consumption of maize production. However, it should be noted that the pattern of rainfall distribution is also relevant to the agronomy of any crop. Table 6. Monthly water consumptive use of maize within the growing season Month 1st 2nd 3th 4th 5th Total Consumptive use (mm) 100 90 96 75 65 426 Rainfall obtainable for early maize (mm) 52 (March) 106.3 (April) 150.8 (May) 165.8 (June) 152.8 (July) 627.7 Rainfall obtainable for late Maize (mm) 152.8 (July) 147.8 (August) 198.9 (September) 170.2 (October) 21.5 (November) 710.7 3.4 Relationship between Rainfall Characteristics and Maize Yield The correlations between rainfall characteristics and maize yield shows that dates of onset (0.263) had weak (significant) positive correlation with maize yield; dates of cessation (-0.284), annual rainfall amounts (-0.241),and rain days (-0.174) had weak negative correlations with maize yield; while annual rain days (−0.088) had very weak negative correlations with maize yield (Table 7). The dates of onset have greatest effect on the annual variation in yield. This implies that the area of study is prone to early onset of rainfall that has effect on the yield of early maize. Annual rain-days have the lowest influence on maize yield. This reveals that only the rain days during production season has effect on the yield of maize significantly. Table 7. Correlation coefficients analysis of rainfall indices and maize yield at the study area Chracteristics Yield Onset Cessation Rain-Days Annual Amount Annual Rain-Days (Hg/Ha) Date Date Day Mm Day Yield (Hg/Ha) 1 Onset .263 1 Cessation -.284 -.016 1 Rain-Days -.174 -.405 .481* 1 Annual Amount -.241 -.418 .408 .710** 1 Annual Rain-Days -.088 -.287 .415 .975** .660** 1 *Coefficient is significant at 0.10 confidence level (2-tailed), **Coefficient is significant at 0.05 confidence level (2-tailed). 3.5 Model for Predicting Maize Yield 3.5.1 Coefficient of prediction The model for predicting maize in relation with rainfall indices is presented in coefficients as shown in table 8. ../../../user/Downloads/azojete143/www.azojete.com.ng Balogun, et al. Development of crop water production model in a rainfed tropical maize crop cultivation. AZOJETE, 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818, www.azojete.com.ng 118 From the regression coefficients shown in Table 8, the regression equation or predictor model of the study was stated thus: Where; Y = Predicted yield of maize in the area. X1, X2, X3, X4, and X5 are Onset date, cessation date, rain-days duration, annual amount and annual rain-days respectively. From this model, it can be inferred, that, given a unit change in any of the rainfall characteristics while holding others constant, the highest variation in yield of maize in the area will be accounted for, by Rain days ( 19851.6 kg/ha), followed by interaction of Rain days ( 859.51 kg/ha), interaction of Annual rain days (756.40 kg/ha), Annual amount (503.65kg/ha), Cessation (356.20 kg/ha), interaction of Cessation and Annual rain days (258.64 kg/ha), interaction Onset and Rain days (167.61 kg/ha) and interaction Annual amount and Annual rain days (25.49kg/ha). Also followed by interaction of Onset and Cessation (11.99 kg/ha), interaction of Onset (0.469 kg/ha), interaction of Onset and Annual amount (0.07657 kg/ha), interaction of Annual amount (-0.161 kg/ha),interaction of Cessation and Annual amount (-1.202 kg/ha), interaction of Cessation (-7.256 kg/ha), interaction of Annual rain days and Annual amount ( -22.04 kg/ha), interaction of Onset and Annual rain days (- 155.42 kg/ha), interaction of Cessation and Rain days ( -221.88 kg/ha), interaction of Rain days and Annual rain days (-1613.6 kg/ha), Onset (-4309.2 kg/ha), and the least change in yield will be from annual Rain days (-35940.4 kg/ha). Table 8. Coefficient of prediction for full quadratic multiple regressions Coefficient P value Std Error -95% 95% t Stat VIF b0 656575 0.308 536247 -1050000.761 2363150.624 1.224 - b1 -4309.2 0.270 3196.2 -14481.1 5862.6 -1.348 29474.7 b2 356.20 0.918 3167.9 -9725.3 10437.7 0.112 13324.1 b3 19851.6 0.547 29320.2 -73458.5 113162 0.677 629446 b4 503.65 0.177 286.27 -407.38 1414.7 1.759 13607.9 b5 -35940.4 0.290 28040.0 -125176 53295.4 -1.282 558065 b6 0.469 0.933 5.153 -15.93 16.87 0.09104 1440.2 b7 11.99 0.481 14.94 -35.56 59.54 0.802 60003.0 b8 167.61 0.332 145.03 -293.95 629.17 1.156 466521 b9 0.07657 0.888 0.499 -1.511 1.664 0.154 1026.5 b10 -155.42 0.313 128.48 -564.29 253.45 -1.210 445507 b11 -7.256 0.399 7.399 -30.80 16.29 -0.981 29698.7 b12 -221.88 0.316 184.83 -810.09 366.34 -1.200 3888302.362 b13 -1.202 0.302 0.966 -4.277 1.873 -1.244 20737.7 b14 258.64 0.237 175.30 -299.26 816.54 1.475 3405930.095 http://www.azojete.com.ng Arid Zone Journal of Engineering, Technology and Environment, December, 2018; Vol. 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818; www.azojete.com.ng 119 b15 859.51 0.203 530.27 -828.03 2547.1 1.621 7949715.86 b16 25.49 0.126 12.11 -13.06 64.04 2.105 687535 b17 -1613.6 0.184 937.33 -4596.5 1369.4 -1.721 25375258.58 b18 -0.161 0.188 0.09466 -0.462 0.140 -1.699 9044.2 b19 -22.04 0.127 10.50 -55.45 11.36 -2.100 526325 b20 756.40 0.164 412.74 -557.12 2069.9 1.833 5132571.28 These showed that among the rainfall characteristics, number of Rain days (X3) is the most valuable variable for the variation in maize yield in the study area indicating that the yield of maize increases as number of rain-days increases. This is followed by interaction of Rain days (X12), meaning that there was higher yield of maize under years with high number of Rain days than those with low rain days. The Rain-days and Annual amount over rainfall period has been the major determinant of maize yield in the area. The contributions of Rain days and Annual amount are followed by those of Annual rain days, dates of cessation and date of onset but with insignificant coefficients. The coefficient of multiple determinations (R2) of 0.936 which was computed as 93.6% was observed. This means that 93.6% of the variations in the yield of maize per hectare for the past 25 years in the study area can be explained jointly by the variations in the five identified rainfall characteristics. The remaining 6.4% of the variations in the yield of maize can be attributed to other unexplained factors such as farming practices, soil properties, planting dates, weeds, fertilizer application seed varieties, pest and diseases, harvesting and other rainfall characteristics/climatic factors and technological involvement. 3.6 Validation of the Regression The Figure 1 shows the scatter diagram of predicted yield and the original yield. The figure shows a good correlation between predicted yield and the yield, hence, the ability of the regression equation to predict yield, when all included rainfall indices are available. Fig 1. Scatter diagram of predicted yield versus yield 4.0 Conclusions The study employed the mean, standard deviation and coefficient of variability in analyzing the variations in the study variables. Partial correlation and multiple linear regression analysis were the statistical tools used to establish the relationship and effect of rainfall characteristics on maize yield. This study established high variability in rainfall characteristics, which indicates high variability in ../../../user/Downloads/azojete143/www.azojete.com.ng Balogun, et al. Development of crop water production model in a rainfed tropical maize crop cultivation. AZOJETE, 14(sp.i4):111-120. ISSN 1596-2490; e-ISSN 2545-5818, www.azojete.com.ng 120 Maize yield per hectare. Additionally, the study showed the suitability of the area for maize production. The study shows that maize experiences moisture deficiency at the early stages (especially early maize) and at the last month of the growing season (late maize) in the study area. In the months of June, July and September rainfall is in excess of the total amount required and that there is relative response of variability in onset/cessation date to variability in rain days. This results in variability in amount of rainfall per rain period. The results also reveal that number of rain days and annual rainfall amount have the strongest influence on maize yield per hectare in the study area. Conclusively, a predictive model was established for maize yield in the area. References Benoit, P. 1977. “The Start of the Growing Season in Northern Nigeria,” Agricultural Meteorology, Vol. 18, No. 2, 1977, Pp. 91-99. doi:10.1016/0002-1571(77)90042-5 Emmanuel, MA. and Fanan, U. 2013. Effect of Variability in Rainfall Characteristics on Maize Yield in Gboko, Nigeria, Journal of Environmental Protection, 2013, 4, 881 887 http://dx.doi.org/10.4236/jep.2013.49103 Published Online September 2013 (http://www.scirp.org/journal/jep) FAO. 2002. Agriculture: towards 2015/30. Technical Interim Report. (http://www.fao.org/es/esd/at2015/toce.htm). FAOSTAT. 2006. Food and Agriculture Organization, Rome, Italy. FAO, 2013. Food and Agricultural Organization Statistics Database online. http//fao.org/faostatagateway/golobrowse/Q/Qc/E. Retrieved on 24th December, 2013 Ifabiyi, IP. and Omoyosoye, O. 2011. Rainfall Characteristics and Maize Yield in Kwara State, Nigeria .Indian Journal of Fundamental and Applied Life Sciences ISSN: 2231 6345 (Online) An Online International Journal Available at http://www.cibtech.org/jls.htm 2011 Vol. 1 (3) July-September, pp.60-65 IITA 2015. Maize http://www.iita.org/maize Iken, JE. and Amusa, NA. 2004. Maize Research and Production In Nigeria. African Journal of Biotechnology Vol. 3 (6), pp. 302-307. Available online at http://www.academicjournals.org/AJB. ISSN 1684–5315 © 2004 Academic Journals IITA 2009. Annual Report on Maize, www.iita.org/cms/details/maize-project-details.aspx. Retrieved on 27th July 2015. Ilesanmi, OO. 1972. “An Empirical Formulation of the Onset, Advance and Retreat of Rainfall in Nigeria,” Journal of Tropical Climatology, Vol. 34, 1972, pp. 17-24. Molden, D. (ed.) 2007. Water for Food, Water for Life: a Comprehensive Assessment of Water Management in Agriculture. Earthscan, London and International Water Management Institute (IWMI), Colombo, Sri Lanka, pp. 1–39. Olaniran, OJ. 1983. “The Start and End of the Growing Season in the Niger River Basin Development Authority Area of Nigeria,” Malaysian Journal of Tropical Geography, Vol. 9, 1983, pp. 49-58. Van Eijnathen LM. 1965. Towards the Improvement of Maize in Nigeria. Ph.D. Thesis. Wageningen, The Netherlands. Wani, SP., Sreedevi, TK., Rockström, J. and Ramakrishna, YS. 2009. Rain fed Agriculture Past Trends and Future Prospects. International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) © CAB International 2009 vol. 1 World Rural Observations 2012. 4(1) http://www.sciencepub.net/rural http://www.scirp.org/journal/jep http://www.fao.org/es/esd/at2015/toce http://www.iita.org/maize http://www.academicjournals.org/AJB http://www.iita.org/cms/details/maize-project-details.aspx http://www.sciencepub.net/rural http://www.azojete.com.ng ARTICLE INFORMATION 1.0Introduction