Internationl Research Journal of Agricultural Economics and Statistics Volume 4 | Issue 1 | March, 2013 | 18-24

Research Paper Impact of global warming on rainfall and wheat production of district in ,

D.T. DESHMUKH AND H.S. LUNGE

See end of the paper for ABSTRACT : India is an agricultural country and agriculture production very much depends on temperature and authors’ affiliations rainfall. Mostly agriculture in India is rainfed. Vidarbha is the eastern region of State. Nearly 89 per Correspondence to : cent of cultivated area of Vidarbha is under rain fed farming.Now a days global warming has become a great D.T. DESHMUKH challenge for the agrarian economy of India. This paper analyses the agriculture production of wheat, average Department of Statistics, maximum and minimum temperatures and total rainfall data for eighteen years obtained from IMD, Pune for Brijlal Biyani Science of Vidarbha. Regression and correlation analysis is obtained and their significance is tested. It College, AMRAVATI (M.S.) INDIA is observed that minimum temperature is increasing significantly for Amravati district where as rainfall and wheat Email : dtdeshmukh.1721 production revealed decreasing trend. Increased temperature and reduced rainfall affects wheat production in @gmail.com Amravati district. KEY WORDS : Agriculture, Climate Variables, Correlation, Regression, t-test Paper History : Received : 26.07.2012; HOW TO CITE THIS PAPER : Deshmukh, D.T. and Lunge, H.S. (2013). Impact of global warming on rainfall and wheat Revised : 12.01.2013; production of Amravati district in Vidarbha, India, Internat. Res. J. agric. Eco. & Stat., 4 (1) : 18-24. Accepted: 14.02.2013

INTRODUCTION In India, monsoons are getting more variable, less predictable and very extreme. It is projected that by the end of the 21st Agriculture plays a key role in overall economic and social century rainfall over India will increase by 15-40 per cent, and well being of India. Agriculture is an economic activity highly mean annual temperature will increase by 3-60C (NATCOM, dependent on climatic conditions. Rain fed agriculture and 2004). Importantly, April of the year 2010 was reported to the farmers are trapped in a phase of continuous crisis. Temperature warmest individual month ever. Eleven of the last twelve years and rainfall are key factors for agriculture production that will during 1995 to 2006 rank among the 12 warmest years in the affect yield of rainfed crops. India ranks first among the rainfed instrumental record of global surface temperature since 1850. agricultural countries of the world in terms of both extent and Analyses done by the Indian Meteorological Department value of produce. Rainfed agriculture is practiced in two-thirds (IMD) and the Indian Institute of Tropical Meteorology (IITM), of the total cropped area of 162 million hectares. In India 65 per Pune, generally show the same trend for temperature, heat cent of agriculture is heavily dependent on natural factors such waves, glaciers, droughts and floods, and sea level rise as by as rainfall, temperature, weather condition etc. In crops, wheat the Intergovernmental Panel on Climate Change of United has been chosen purposively since wheat has played important Nations (Raghava Reddy, 2010). Increase in global temperature role in achieving food security of the country. Global warming will affect the agriculture production in India. This paper becomes an alarming issue of concern in the developing world. analyzes statistically the atmospheric temperature, rainfall and Global warming is the increase in the average temperature of agriculture production data of wheat for Amravati district during the earth’s near surface air and oceans since the mid-twentieth the study period 1988 to 2005. According to IPCC reports, the century and its projected continuation. Global temperature will surface temperature of the earth has risen by 0.6 ± 0.20C over increase by 1.80C to 40C with an overall average increase of the 20th century. The increased temperature resulting from 2.80C in temperature (IPCC, 2007). The average global global warming is likely to reduce the profit from wheat temperature has risen by about 0.80C from pre-industrial level. cultivation. More recent studies done at the Indian Agricultural

HIND AGRICULTURAL RESEARCH AND TRAINING INSTITUTE D.T. DESHMUKH AND H.S. LUNGE

Research Institute indicate the possibility of loss of 4-5 million MATERIALS AND METHODS tons in wheat production with every rise of 10C temperature throughout the growing period even after considering benefits The data used in this paper are the yearly averages of of carbon fertilization (Agarwal, 2007). Previous studies total mean rainfall, minimum and maximum atmospheric concluded that the dry lands are greatly affected due to climate temperatures. The yearly averages were calculated from the change (Eid et al., 2007, Kurukulasuriya and Mendelsohn, monthly readings which are provided by the India 2008). The problem of effect of global warming on rainfall and Meteorological Department, Pune. Also year wise secondary agriculture production in PNG is studied and observed that data for area, production, and yield of wheat for Amravati district reduced rainfall is affecting the agriculture production of PNG were obtained from ‘Epitome of Agriculture- Part I and Part II’ (Rehman, 2008). It may be noted that in the last 50 years, the published in 2004-05 by State Agriculture Department, Mumbai. rise in temperature has been 0.13± 0.070C per decade. Very Correlation analysis and regression analysis are applied to recently, National Oceanic and Atmospheric Administration production data as well as to temperature and rainfall data. The (NOAA), climate agency of America, reported that the average p-values are obtained and tested at 5 per cent level of temperature of earth for the first four months of the year 2010, significance. Temperature, rainfall and agriculture production of 13.30 C is 0.690C above the 20th century average. While the global ocean surface temperature was 0.570C above the 20th century average of 160C (Fulekar and Kale, 2010). It was found that temperature increase has significant negative impact on agriculture production. Moreover, an increase in revenue was visualized with the increase in rainfall. The overall extent of negative impact of temperature is greater than the positive effect of rainfall in the region (Shakoor et al., 2011). Rainfed agriculture supports 40 per cent of the India’s population and contributes 44 per cent to the national food basket (Angles et al., 2011). These studies motivate us to study statistically the changes in temperature, rainfall, and agriculture production of wheat in Amravati district. Fig. A : Amravati district map

Table A : Year wise data for the eighteen years of Amravati district Year MMIN MMAX TMRF TMRF(J-S) mm Production Yield 0C 0C mm ‘00’ton. (kg/hectare) 1988 22.075 34.4166667 100.64167 247.875 463 1302 1989 21.74545 33.6545455 68.063636 171.175 342 1090 1990 19.33333 33.0416667 95.733333 222.4 365 1185 1991 19.7 34.7416667 43.483333 123.35 128 1153 1992 15.33333 35.0363636 74.745455 190.15 215 1304 1993 14.6 34.25 61.333333 134.55 191 1446 1994 0 31.9 33.52 167.6 229 1452 1995 0 33.55 39.16 - 214 1390 1996 - - - - 244 1574 1997 21.53333 32 122.6 116.4 130 783 1998 20.9 30.925 80.088889 199.46667 249 1283 1999 18.5 33.5583333 83.391667 223.675 325 1609 2000 17.95833 34.3083333 65.583333 155.15 170 1269 2001 17.30833 34.15 58.291667 132.525 150 1325 2002 21.80833 32.3083333 58.183333 166.25 115 1034 2003 21.77273 32.69 47.15 141.45 144 1246 2004 19.675 30.5 41.083333 94.2 120 1164 2005 19.73333 30.775 75.116667 200.15 233 1368 Source: IMD, Pune and district wise Agricultural Statistical Information, Commissionrate of Agriculture, Pune

Internat. Res. J. agric. Eco. & Stat. 4(1) March, 2012 : 18-24 HIND AGRICULTURAL RESEARCH AND TRAINING INSTITUTE 19 IMPACT OF GLOBAL WARMING ON RAINFALL & WHEAT PRODUCTION

data of wheat are made over the years i.e. time and therefore as against the district average of 924.56 mm. (36.4 are referred to as time series data, which is defined as a series inches). The soils of the district are derived from the Deccan of observations that varies over time. The time series is made trap (a basaltic rock). up of four components known as seasonal, trend, cyclical and Table Ashows the eighteen years data for mean of irregular (Patterson, 1987). Trend is defined as the general maximum and mean of minimum temperatures, total mean movement of a series over an extended period of time or it is rainfall, and total mean rainfall from the month of June to the long term change in the dependent variable over a long September. The some data for 1996 are not available. period of time (Webber and Hawkins, 1980). Trend is determined by the relationship between the two variables as temperature RESULTS AND DATA ANALYSIS and time, rainfall and time, and agriculture production and time. The statistical methods such as regression analysis, The results obtained from the present investigation as correlation, and coefficient of determination (Murray et al., well as relevant discussions have been summarized under 2000), t-test and p-values are used. following heads:

Study area: Linear regression: Recently, Vidarbha region has become infamous for a The linear regression line was fitted using the most large number of farmer suicides occurring. Amravati district is common method of principle of least squares. This method one of the six distressed districts of Vidarbha for which the calculates the best fitting line for the observed data by Government of India and Government of Maharashtra State minimizing the sum of the squares of the vertical deviations have announced the package of relief for the farmers. Vidarbha’s from each data point to the line. If a point lies exactly on the economy is primarily agricultural and it is less economically straight line then the algebraic sum of the residuals is zero. prosperous as compared to the rest of Maharashtra. Vidarbha Residuals are defined as the difference between an observation is the eastern region of Maharashtra State made up of Nagpur at a point in time and the value read from the trend line at that division and Amravati division. Nearly 89 per cent of cultivated point in time. A point that lies far from the line has a large area of Vidarbha is under rain fed farming. Amravati district is residual value and is known as an outlier or, an extreme value. situated between 20032’ and 21046’N latitudes and 76037’ and The equation of a linear regression line is given as: 78027’E longitudes which are located in Deccan plateau. The y = a + b x, district occupies an area of 12,212 sq. km. which is 3.96 per cent where, y is the observation on the dependent variable, x of Maharashtra State. It comprises 14 talukas namely, Amravati, is the observation on the independent variable, a is the intercept , Warud, Tiosa, Dhamangao Railway, Chandur Railway, of the line on the vertical axis, and b is the slope of the line. Nandgaon Khandeshwar, Bhatkuli, Chandur Bazar, Daryapur, In order to fit regression lines scatter diagrams of the Anjangaon Surji, Achalpur, Chikhaldara and Dharni. annual average temperature, rainfall and agriculture production The climate of the district is tropical and summer (dependent variables) against time (independent variable) in temperature can go up to higher than 440C (1110F). Amravati years were plotted (Fig. 1a and b). Linear regression lines were district faces extreme variations in temperature with very hot then fitted to determine the trends of temperature, rainfall and summer and very cold winter. The district receives rainfall from agriculture production. The drawing of the scattered diagrams south west monsoons mainly in the months of June, July, and the fitting of the regression lines were done in Microsoft August and September. Maximum rainfall occurs during July Excel. and August. Amravati has a tropical wet and dry climate with Fig. 1a and b indicate the trend line for rainfall in the hot, dry summers and mild to cool winters. Summer lasts from month of June to September. The production of wheat against March to June, monsoon season from July to October and time is decreasing, which implies that there was a negative winter from November to February. The highest and lowest linear relationship between rainfall in the month of June to temperatures ever recorded were 46.7 °C on 25th May, 1954 September and time, and production of wheat and time. and 5.0° C on 9th February, 1887, respectively. Fig. 2 a and b indicate the trend line for mean of minimum The soils of Amravati district have been developed on temperature and yield of wheat against year is increasing which the hilly and undulating topography. The soils, therefore, show implies the positive correlation between them. The trend line a wide variation in their depth. The soils on the hills and slopes for yield wheat against time nearly remained constant over the have shallow to medium depth while in the low-lying areas and years. river valleys, deep soils are formed due to the accumulation Fig. 3 a and b indicate the trend line for total mean rainfall and deposition of the soils from uplands. Thus, three main soil against mean of maximum temperature which is slightly types are obtained, viz., (1) Shallow soils, (2) Medium deep increasing and for rainfall in the month of June to September soils and (3) Deep soils. The annual rainfall varies from 721.36 against mean of maximum temperature is also increasing, which mm. (28.4 inches) at Anjangaon to 1,701.8 mm. (67.0 inches) at implies there was a very low positive linear relationship between

Internat. Res. J. agric. Eco.& Stat. 4 (1) March, 2013 : 18-24 20 HIND AGRICULTURAL RESEARCH AND TRAINING INSTITUTE D.T. DESHMUKH AND H.S. LUNGE

300 500 250 400 200 300 (J-S) 150 200 Production TMRF 100 100 50 0 0 1985 1990 1995 2000 2005 2010 1985 1990 1995 2000 2005 2010 Year Year

Fig. 1 (a &b) : Scatter diagram and the trend line for various variables of Amravati district

25 2000 20 1500 15 1000

10 Yield TMRF 5 500

0 0 1985 1990 1995 2000 2005 2010 1985 1990 1995 2000 2005 2010 Year Year

Fig. 2 (a &b) : The trend line for mean temperature and wheat yield total mean rainfall and mean of maximum temperature,and rainfall relationship between total mean rainfall and yield of wheat. in the month of June to September against mean of maximum Fig. 5 a and b indicate the trend line for production of temperature. wheat and yield of wheat against total mean rainfall in the Fig. 4 a and b indicate the trend line for production of month of June to September which is increasing and implies wheat against total mean rainfall which is increasing rapidly there was a high positive linear relationship between and implies that was a high positive linear relationship between production of wheat and total mean rainfall in the month of total mean rainfall and production of wheat, whereas yield of June to September but yield of wheat showed moderate positive wheat shows decreasing trend, which implies negative relationship.

150 300

100 200 TMRF 50 TMRF-JS 100

0 0 30 32 34 36 30 32 34 36 Temp.MMAX Temp.MMAX

Fig. 3 (a &b) : Trend line for total mean rainfall and mean of max. temperature

Internat. Res. J. agric. Eco. & Stat. 4(1) March, 2012 : 18-24 HIND AGRICULTURAL RESEARCH AND TRAINING INSTITUTE 21 IMPACT OF GLOBAL WARMING ON RAINFALL & WHEAT PRODUCTION

500 2000

400 1500 300 1000

200 Yield

Production 500 100

0 0 0 50 100 150 0 50 100 150 Rainfall Rainfall

Fig. 4 (a &b) : Trend line for production of wheat against total mean rainfall

Fig. 6a and b indicate the trend line for total mean rainfall the Karl Pearson’s formula for calculating the correlation against mean of minimum temperature which is increasing and coefficient ‘r’ is given by: implies there existed a positive relationship between them. 2 2 2 r   (xi – x) (yi – y) / Σ (x1 – x) (y1 – y) Similarly the trend line for yield of wheat against mean of minimum temperature is decreasing, which implies that there i=1,2,………..n. existed a negative relationship between yield of wheat and Testing the significance of the correlation co-efficient: mean of minimum temperature. In testing the significance of the correlation coefficient, the following null (H ) and alternative (H ) hypothesis were Correlation co-efficient: 0 1 considered. The correlation co-efficient determines the magnitude and Hypothesis: H : =0 against H : 0 strength of linear relationship between the two variables under 0 1 where,  is the population correlation coefficient. study. It always lies between –1 to +1. The value +1 indicating The appropriate test statistics for testing the above a perfect positive correlation and the value -1 indicating a perfect hypothesis is negative correlation (means all points would lie along a straight line and having a residual of zero). A correlation co-efficient r (n  2) closed to or equal to zero indicates no or very poor relationship t  d. f. = n-2 = 16 (1 - r2 ) between the variables. A positive correlation co-efficient indicates a positive (upward) relationship and a negative Significant value for t at 5 per cent level = 2.12 correlation co-efficient indicates a negative (downward) Table 1 represents the values of the correlation co- relationship between the variables. The correlation co-efficients efficients and the test statistics represented within bracket. between temperature, rainfall, agriculture production and time The correlation coefficients determine the strength of the were calculated as follows : linear relationship between the two variables. Table 1 shows the correlations between the variables as temperature, rainfall, Given the pairs of values (x1,y1), (x2,y2),…………….(xn,yn),

500 2000

400 1500 300 1000 200 Yield

Production 500 100 0 0 0 100 200 300 0 100 200 300 Rainfall Rainfall

Fig. 5 (a &b) : Trend line production of wheat and yield of wheat against total rainfall

Internat. Res. J. agric. Eco.& Stat. 4 (1) March, 2013 : 18-24 22 HIND AGRICULTURAL RESEARCH AND TRAINING INSTITUTE D.T. DESHMUKH AND H.S. LUNGE

2000 150 1500 100 1000 Yield

TMRF 50 500

0 0 0 10 20 30 0 10 20 30 Temp-MMIN Temp-MMIN

Fig. 6 (a &b) : Trend line of total mean reainfall against mean of min. temperature

Table 1 : Correlation co-efficients and t values Temp. MMAX Temp. MMIN PROD. YIELD TMRF 0.021391 0.51585 0.45559 -0.33934 (0.085584) (2.408604) (2.047159) (-1.44292) Insignificant Significant Insignificant Insignificant TMRF J-S 0.13039 0.052368 0.853379 0.416908 (0.526051) (0.20976) (6.548197) (1.834682) Insignificant Insignificant Significant Insignificant Temp. 0.204313 0.188155 MMAX (0.834863) (0.766307) Insignificant Insignificant Temp. MMIN 0.056516 -0.47178 (0.226426) (-2.14028) Insignificant Insignificant production and the yield of wheat. The highest positive = 1- 0.98554= 0.01446 correlation is seen between the rainfall in the month of June to As p-value is less than 0.05, the Null hypothesis is September and production of wheat, whereas there is a negative rejected and concluded that the correlation co-efficient between relationship between the total mean rainfall and the yield of total mean rainfall and mean of minimum temperature is highly wheat, mean of minimum temperature and yield of wheat. significant. The correlation co-efficient between total mean Similarly there is 0.51585 degree of relationship between the rainfall in the month of June to September and production of mean of minimum temperature and the total mean rainfall, wheat is also highly significant.The rest of the correlation co- whereas very low relationship was observed between the mean efficients are insignificant. of minimum temperature and the total mean rainfall in the month of June to September. Conclusion: The test statistic value for the total mean rainfal in the On the basis of present investigation, the trend analysis month of June to September and production of wheat is highly indicates that the mean of minimum temperature increased significant. Similarly the test statistic value for the total mean significantly over the years and production of wheat was rainfall and mean of minimum temperature is highly significant. decreasing sharply over the years for Amravati district, indicating the impact of global warming.While production of Calculating p-value: wheat and yield of wheat are very much dependent on total The p-value for a test of hypothesis is defined to be the mean rainfall in the month of June to September but total mean smallest level of significance at which the Null hypothesis is rainfall in the month of June to September showed decreasing rejected. trend over the years indicating the global warming impact. It is The p-values were calculated in the following manner. seen that the coefficient of determination R2 between total mean p-value= P(t> Observed value of the test statistic) rainfall in the month of June to September and production of = 1- P(t< Observed value of the test statistic) wheat was 0.728, means out of the total variations, 72.8 per For t= 2.4086, p= 1- P( t< 2.4086) cent variations can be explained and remaining 27.2 per cent

Internat. Res. J. agric. Eco. & Stat. 4(1) March, 2012 : 18-24 HIND AGRICULTURAL RESEARCH AND TRAINING INSTITUTE 23 IMPACT OF GLOBAL WARMING ON RAINFALL & WHEAT PRODUCTION variations are unexplained. This could be due to random Fulekar, M.H. and Kale, R.K. (2010). Impact of Climate Change: fluctuations or an additional variable that has not been Indian Scenario, University News, 48 (24) : 15-23. considered. Similarly, 26.6 per cent explained variations were Kurukulasuriya, P. and Mendelsohn, R. (2008). A Ricardian analysis observed between mean of minimum temperature and total mean of the impact of climate change on African cropland. AFJARE, rainfall and remaining 70.4 per cent variations are unexplained. 2: 1-23. This could be due to random fluctuations or an additional IPCC (2007). Climate change-A synthesis Report of the IPCC, variable that has not been considered. Technical Report, Inter-governmental Panel on Climate Change. The others like Usman Shakoor et al. (2011) has studied the problem of impact of climate change on agriculture and he Murray, R. Spigel and Stephens, Larry J. (2000). Schaum’s outlines Statistics, 3rd Ed., Tata McGraw. Hill, NEW DELHI (INDIA). found that there is a negative impact of temprature is greater than the positive effect of rainfall in Rawalpindi NATCOM (2004). India’s Initial National Communication to the region,Pakistan.The Shafiqur Rahman (2008) has observed that UNFCCC. National Communication Project, MoEF, Govt. of the temparatutre is incresing significantly over the thity six India. years time span, in the region PNG, while the rainfall shows Patterson, P.E. (1987). Statistical methods, Richard D. Irwin INC, decresing trend which affects the agriculture poduction of PNG. Homewood, IL. We, in our study, found that the agriculture production of wheat Raghava Reddy,P.(2010). Impact of weather extremes on Indian is decresing sharply over the eighteen years of span in Amravati agriculture in the context of climate change, University News, A district of Vidharbha,India. Weekly Journal of Higher Education, Association of Indian Universities, 48 (24), June 14-20. Authors’ affiliations: Shafiqur Rahman (2008). Effect of global warming on rainfall and H.S. LUNGE, Department of Statistics, Shri Shivaji Science College, AMRAVATI (M.S.) INDIA agriculture production, Internat. Rev. Business Res. Papers, 4 Email : [email protected] (4) : 319-329. Usman Shakoor, Abdul Saboor, Ali, Ikram and Mohsin, A.Q. (2011). LITERATURE CITED Impact of climate change on agriculture: Empirical evidence from arid region, Pak. J. Agri. Sci., 48(4) : 327-333. Agrawal, P.K.(2007). Climate change: Implications for Indian Agriculture. Jalvigyan Sameeksha, 22 : 37-46. Webber, J. and Hawkins, C. (1980). Statistical analysis applications to business and economics, Harper and Row, New York. Angles, S., Chinnadurai, M. and Sundar, A. (2011). Awareness on impact of climate change on dryland agriculture and coping mechanisms of dryland farmers, Indian J. Agric. Econ., 66 (3) : 365-372. Eid, H.M., El-Marsafawy, S.M. and Ouda, S.A. (2007). Assessing the economic impacts of climate change on agriculture in Egypt: A ricardian approach, Development Research Group, Sustainable Rural and Urban Development Team, The World Bank, Policy Research Working Paper 4293. *—*—*—*—*—*—*—*

Internat. Res. J. agric. Eco.& Stat. 4 (1) March, 2013 : 18-24 24 HIND AGRICULTURAL RESEARCH AND TRAINING INSTITUTE