Modeling the Impact of Climate and Non-Climatic Factors on Cereal 1 Production: Evidence from Indian Agriculture 2

: The underpinned study examines the effects of climatic and non-climatic 22 factors on Indian agriculture, cereal production, and yield using the country-level time 23 series data of 1965–2015. With the autoregressive distributed lag (ARDL) bounds 24 testing approach, the long-term equilibrium association among the variables has been 25 explored. The results reveal that climatic factors like CO 2 emissions and temperature 26 adversely affect agricultural output, while rainfall positively affects it. Likewise, non- 27 climatic factors, including energy used, financial development, and labor force, affect 28 agricultural production positively in the long run. The estimated long-run results further 29 demonstrate that CO 2 emissions and rainfall positively affect both cereal production 30 and yield, while temperature adversely affects. The results exhibit that the cereal 31 cropped area, energy used, financial development, and labor force significantly and 32 positively impact the long-run cereal production and yield. Finally, pairwise granger 33 causality test confirmed that both climatic and non-climatc factors are significantly influencing agriculture and cereal production in India. Based on these results, 35 policymakers and governmental institutions should formulate coherent adaptation 36 measures and mitigation policies to tackle the adverse climate change effects on 37 agriculture and its production of cereals. 38

Agricultural production appears vulnerable to climatic changes and negatively impacts 69 human health, dairy and milk production, agricultural trade, and the price of food-grain  (Solomon et al. 2007). 89 In particular to the Asian emerging economy, India, the agriculture sector is still vital 90 in economic development, despite the recent decrease in gross domestic products. This 91 sector is continuously playing a pivotal role in food safety, poverty reduction, and job 92 creation, employing 52 percent of the labor force (Guntukula 2019). The diversity in 93 the agricultural sector is also high, i.e., a massive geographical area like natural 94 resources, crop production management, weather conditions. However, it has become 95 a more fragile and exposed area due to the low level of development and poor 96 adaptation policy (Birthal et al. 2014). Its 30 percent population is poor, and 50 percent 97 of farmers are still at a subsistence level of farming (Kumar et al. 2015), whereas more 98 than 60 percent population relay on agricultural activities (Pattanayak &Kumar 2014). temperature by 2 -4ºC, a surge in rainfall during the rainy season, and a 15 -20 percent 106 rise in precipitation. It will also impact agricultural productivity physically (Gupta et 107 al., 2014); evidence shows that cereal, rice, cotton, sugarcane, sunflower, and wheat 108 production significantly decreased (Gupta et al. 2014, Mall et al. 2006). The surge in 109 temperature by 1 to 2ºC will affect rice production by 3 to 17 percent in India (Aggarwal 110 &Mall 2002). In contrast, the influence of carbon fertilization on agriculture production 111 has predicted a loss for the country by 0-40 percent (Aggarwal (2008). Thus, as farmers lack proper financial resources to mitigate the effects of the 117 environment on agriculture, climate change is becoming a severe challenge for 118 economists, agriculturists, and policymakers to develop an advanced technique to 119 alleviate the effects of climate on agriculture activities (Singh et al. 2017). Besides, 120 most literature is found in developed countries, raising the concern for the country's 121 food security (Adger et al. 2003 1961  1965  1969  1973  1977  1981  1985  1989  1993  1997  2001  2005  2009  2013  2017 Cereal yield (kg per hectare) Cereal production (metric tons) to identify climatic and non-climatic factors that affects cereal production, and (iii) also 127 evaluating the combined effects of climatic and non-climatic factors on cereal yield.

128
The present comprehensive study significantly contributes to the existing literature as 129 we are the pioneer in exploring the short-and long-term impacts of climate change and 130 other important input factors on agricultural value added, cereal production and cereal 131 yield in the case of India using the ARDL framework and Granger causality tests.
132 Figure 2 demonstrates the conceptual framework presenting climatic and non-climatic 133 factors that may affect Indian agriculture, cereal production and cereal yield.   sugarcane, also predicted that change in climate adversely affects the sugarcane, 149 understandably, impacting the 40% of worldwide land used for agriculture production.    Table 1. Whereas, the 251 trend of logarithmically transformed all the variables is shown in Figure 3.  (1) can be 300 expressed as follows: The conditional ARDL model for Eq.
(2) expressed as follows: The conditional ARDL model for Eq. (3) expressed as follows: Following the cointegration tests based on Equations (4), (5), and (6), the error 335 correction models (ECM) for the agricultural value-added, cereal production, and cereal 336 yield specifications, for the present study, are specified as follows: Before applying the ARDL approach, we checked the orders of integration of the series.

369
The examined series is mixed orders of integration, as observed in the estimated 370 outcomes of both unit root tests include ADF and PP (see Table 2). The estimated 371 outcomes of both unit root tests suggested that the ARDL approach can be used for 372 examining the long-run and short-run interrelationships among variables.    Table 3). The 382 findings suggesting that the ARDL model can be applied for further estimation.    0.310 (0.577) ** and *** indicate the rejection of no cointegration at the 5 and 1% significance level, respectively. Table 6 reports the estimated long-and-short-run outcomes of the model (I), and Figure   415 4 shows the summarized long-run nexus among the variables. The predicted long-and-short-run coefficients for a climate like carbon dioxide and 421 mean temperate are significantly and negatively affecting agricultural value-added.     Table 7 reports the estimated long-and-short-run outcomes of Model (II) and Figure 7 469 shows the summarized long-run association among variables. fertilizers usage significantly increased cereal production. In this study, we applied 505 various diagnostic and stability tests to verify the estimated ARDL model. Table 6 506 reports the outcomes of various diagnostic tests. As shown in Table 7, all diagnostic 507 tests confirm that the ARDL is free from diagnostic problems. The CUSUM and 508 CUSUM square both stability tests show that the ARDL model is stable over the 509 sampled period (see Figures 8 and 9). 510 Table 7. ARDL model II: The impact of non-climatic and climatic factors on cereal 511 production 512

383
Model selection method: Akaike information criteria (AIC) Selected model: ARDL(1, 1, 1, 0, 0, 0, 2, 1, 2  We undertook the ARDL approach for identifying the non-climatic and climatic factors 527 impacting the yield of cereals. Table 8 presents the empirical long-and-short-run of the 528 ARDL model, and Figure 10 displays the summary of the long-run.  Table 8 shows that the coefficient of CO2 emission is positive in the long-run; however, 536 the coefficient of CO2 emission is negative in the short run. The coefficients of average 537 temperature in both the long-run and short-run have a significant negative effect on 538 cereal yield; therefore, a 1ºC increase in temperature will decrease the cereal yield by 539 1.844% and 2.252%, respectively. In coming decades, the crop productivity is more tests confirmed the constancy of the model (see Figure 11 and 12).  The pairwise Granger causality test is applied to explore the causal associations 568 between the study variables. The estimated results are summarized in Table 9, 569 indicating the existence of one-way causality between CO2 and agricultural value-  In order to verify the existence of causal links between variables, the results obtained 580 in the estimation of model II are reported in Table 10, showing the unidirectional 581 causality from CO2 and rainfall to cereal production while two-way causality explored 582 from temperature to cereal production. This means climatic factors significantly 583 influencing cereal production. Besides, the unidirectional causality from cereal cropped 584 area, energy consumption, and financial development to cereal production whereas 585 two-way causality discovered from gross capital formation and rural labour to cereal 586 production is varified. These results imply that non-climatc factors play an important 587 role to enhance cereal production and ensure food security in India.         The data will be available on request. 654 The authors declare that they have no conflict of interest.