• Agriculture and Agri-Food Canada, 2014: Impact of climate change on Canadian agriculture. Accessed 26 August 2015. [Available online at http://www.agr.gc.ca/eng/science-and-innovation/agricultural-practices/climate/future-outlook/impact-of-climate-change-on-canadian-agriculture/?id=1329321987305.]

  • An, H., and R. Carew, 2015: Effect of climate change and use of improved varieties on barley and canola yield in Manitoba. Can. J. Plant Sci., 95, 127139, doi:10.4141/cjps-2014-221.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Antle, J. M., 1983: Testing the stochastic structure of production: A flexible moment-based approach. J. Bus. Econ. Stat., 1, 192201, doi:10.1080/07350015.1983.10509339.

    • Search Google Scholar
    • Export Citation
  • Antle, J. M., 2010: Asymmetry, partial moments and production risk. Amer. J. Agric. Econ., 92, 12941309, doi:10.1093/ajae/aaq077.

  • Antle, J. M., J. E. Mu, and J. Abatzoglou, 2013: Climate and the spatial-temporal distribution of winter wheat yields: Evidence from U.S. Pacific Northwest. 2013 AAEA & CAES Joint Annual Meeting, Washington, DC, Agricultural and Applied Economics Association, 2617. [Available online at http://ageconsearch.umn.edu/bitstream/150393/2/AAEA_2013_JianhongMu_%232617.pdf.]

  • Ash, G. H. B., C. F. Shaykewich, and R. L. Raddatz, 1992: Moisture risk assessment for spring wheat on the eastern prairies: A water-use simulation model. Climatol. Bull., 26 (2), 6578.

    • Search Google Scholar
    • Export Citation
  • Ash, G. H. B., C. F. Shaykewich, and R. L. Raddatz, 1993: The biologically important thermal character of the eastern prairie climate. Climatol. Bull., 27 (1), 320.

    • Search Google Scholar
    • Export Citation
  • Asseng, S., I. Foster, and N. C. Turner, 2011: The impact of temperature variability on wheat yields. Global Change Biol., 17, 9971012, doi:10.1111/j.1365-2486.2010.02262.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bueckert, R. A., and J. M. Clarke, 2013: Annual crop adaptation to abiotic stress on the Canadian prairies: Six case studies. Can. J. Plant Sci., 93, 375385, doi:10.4141/cjps2012-184.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cabas, J., A. Weersink, and E. Olale, 2010: Crop yield response to economic, site and climatic variables. Climatic Change, 101, 599616, doi:10.1007/s10584-009-9754-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Campbell, C. A., F. Selles, R. P. Zentner, B. G. McConkey, S. A. Brandt, and R. C. McKenzie, 1997: Regression model for predicting yield of hard red spring wheat grown on stubble in the semiarid prairie. Can. J. Plant Sci., 77, 4352, doi:10.4141/P96-043.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Canola Council of Canada, 2015: Estimated acreage of HT and conventional canola, 1995–2010. Accessed 2 April 2015. [Available online at http://www.canolacouncil.org/markets-stats/statistics/estimated-acreage-and-percentage/.]

  • Chen, C.-C., B. A. McCarl, and D. E. Schimmelpfennig, 2004: Yield variability as influenced by climate: A statistical investigation. Climatic Change, 66, 239261, doi:10.1023/B:CLIM.0000043159.33816.e5.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, S., X. Chen, and J. Xu, 2013: Impacts of climate change on corn and soybean yields in China. 2013 AAEA & CAES Joint Annual Meeting, Washington, DC, Agricultural and Applied Economics Association, 149739. [Available online at http://ageconsearch.umn.edu/bitstream/149739/2/AAEA2013.pdf.]

  • Chipanshi, A., 2013: Tabulation weather data request. Agriculture and Agri-Food Canada, National Agro-Climate Information Service, Regina, SK, Canada, accessed 8 August 2013.

  • Cole, M. A., R. J. Elliott, T. Okubo, and Y. Zhou, 2013: The carbon dioxide emissions of firms: A spatial analysis. J. Environ. Econ. Manage., 65, 290309, doi:10.1016/j.jeem.2012.07.002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Deschênes, O., and M. Greenstone, 2007: The economic impacts of climate change: Evidence from agricultural output and random fluctuations in weather. Amer. Econ. Rev., 97, 354385, doi:10.1257/aer.97.1.354.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Di Falco, S., and J.-P. Chavas, 2009: On crop biodiversity, risk exposure, and food security in the highlands of Ethiopia. Amer. J. Agric. Econ., 91, 599611, doi:10.1111/j.1467-8276.2009.01265.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Government of Saskatchewan, 2015: Saskatchewan agricultural exports. Accessed 23 March 2015, 25 pp. [Available online at http://publications.gov.sk.ca/documents/20/83881-Saskatchewan%20Agriculture%20Exports%202015.pdf.]

  • Grisé, J., 2013: Effect of climate change on farmers’ choice of crops: An econometric analysis. M. S. thesis, Department of Bioresource Policy, Business and Economics, University of Saskatchewan, 143 pp.

  • He, Y., and Coauthors, 2013: Spring wheat yield in the semiarid Canadian prairies: Effects of precipitation timing and soil texture over recent 30 years. Field Crops Res., 149, 329337, doi:10.1016/j.fcr.2013.05.013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hoechle, D., 2007: Robust standard errors for panel regressions with cross-sectional dependence. Stata J., 7, 281312. [Available online at http://www.stata-journal.com/sjpdf.html?articlenum=st0128.]

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hopkins, J. W., 1935: Weather and wheat yields in western Canada. I. Influence of rainfall and temperature during the growing season on plot yields. Can. J. Res., 12, 306334, doi:10.1139/cjr35-027.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Im, K. S., M. H. Pesaran, and Y. Shin, 2003: Testing for unit roots in heterogenous panels. J. Econom., 115, 5374, doi:10.1016/S0304-4076(03)00092-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • IPCC, 2014: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. C. B. Field et al., Eds., Cambridge University Press, 1132 pp. [Available online at http://www.ipcc.ch/pdf/assessment-report/ar5/wg2/WGIIAR5-PartA_FINAL.pdf.]

  • Isik, M., and S. Devadoss, 2006: An analysis of the impact of climate change on crop yields and yield variability. Appl. Econ., 38, 835844, doi:10.1080/00036840500193682.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jones, J. W., and Coauthors, 2003: The DSSAT cropping system model. Eur. J. Agron., 18, 235265, doi:10.1016/S1161-0301(02)00107-7.

  • Just, R. E., and R. D. Pope, 1979: Production function estimation and related risk considerations. Amer. J. Agric. Econ., 61, 276284, doi:10.2307/1239732.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kutcher, H. R., J. S. Warland, and S. A. Brandt, 2010: Temperature and precipitation effects on canola yields in Saskatchewan, Canada. Agric. For. Meteor., 150, 161165, doi:10.1016/j.agrformet.2009.09.011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lemmen, D. S., and F. J. Warren, Eds., 2004: Climate change impacts and adaptation: A Canadian perspective. Natural Resources Canada, 201 pp. [Available online at http://www.environment.msu.edu/climatechange/canadaadaptation.pdf.]

    • Crossref
    • Export Citation
  • Lobell, D. B., and C. B. Field, 2007: Global scale climate-crop yield relationships and the impacts of recent warming. Environ. Res. Lett., 2, 014002, doi:10.1088/1748-9326/2/1/014002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., and J. I. Ortiz-Monasterio, 2007: Impacts of day versus night temperatures on spring wheat yields: A comparison of empirical and CERES model predictions in three locations. Agron. J., 99, 469477, doi:10.2134/agronj2006.0209.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., and M. B. Burke, 2008: Why are agricultural impacts of climate change so uncertain? The importance of temperature relative to precipitation. Environ. Res. Lett., 3, 034007, doi:10.1088/1748-9326/3/3/034007.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., and M. B. Burke, 2010: On the use of statistical models to predict crop yield responses to climate change. Agric. For. Meteor., 150, 14431452, doi:10.1016/j.agrformet.2010.07.008.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., W. Schlenker, and J. Costa-Roberts, 2011: Climate trends and global crop production since 1980. Science, 333, 616620, doi:10.1126/science.1204531.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCaig, T. N., 1997: Temperature and precipitation effects on durum wheat grown in southern Saskatchewan for fifty years. Can. J. Plant Sci., 77, 215223, doi:10.4141/P96-144.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCarl, B. A., X. Villavicencio, and X. Wu, 2008: Climate change and future analysis: Is stationarity dying? Amer. J. Agric. Econ., 90, 12411247, doi:10.1111/j.1467-8276.2008.01211.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mearns, L. O., 1988: Technological change, climate variability, and winter wheat yields in the Great Plains of the United States. Ph.D. dissertation, University of California, Los Angeles, 223 pp.

  • Mendelsohn, R., and M. Reinsborough, 2007: A Ricardian analysis of US and Canadian farmland. Climatic Change, 81, 917, doi:10.1007/s10584-006-9138-y.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meng, T., R. Carew, W. J. Florkowski, and A. M. Klepacka, 2016: Modeling temperature and precipitation influences on yield distributions of canola and spring wheat in Saskatchewan. Agricultural and Applied Economics Association (formerly the American Agricultural Economics Association) 2016 Annual Meeting, Boston, MA, Agricultural and Applied Economics Association, 38 pp. [Available online at http://purl.umn.edu/235251.]

  • Mkhabela, M. S., P. Bullock, S. Raj, S. Wang, and Y. Yang, 2011: Crop yield forecasting on the Canadian prairies using MODIS NDVI data. Agric. For. Meteor., 151, 385393, doi:10.1016/j.agrformet.2010.11.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nagy, C., and R. Gray, 2012: Rate of return to the research and development expenditure on zero tillage technology development in western Canada (1960-2010). Department of Bioresource Policy, Business and Economics, University of Saskatchewan, 35 pp.

  • Nalley, L. L., A. P. Barkley, and A. M. Featherstone, 2010: The genetic and economic impact of the CIMMYT wheat breeding program on local producers in the Yaqui Valley, Sonora Mexico. Agric. Econ., 41, 453462, doi:10.1111/j.1574-0862.2010.00457.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ortiz-Bobea, A., and R. E. Just, 2013: Modeling the structure of adaptation in climate change impact assessment. Amer. J. Agric. Econ., 95, 244251, doi:10.1093/ajae/aas035.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Özdoğan, M., 2011: Modeling the impacts of climate change on wheat yields in northwestern Turkey. Agric. Ecosyst. Environ., 141, 112, doi:10.1016/j.agee.2011.02.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Potgieter, A., H. Meinke, A. Doherty, V. O. Sadras, G. Hammer, S. Crimp, and D. Rodriguez, 2013: Spatial impact of projected changes in rainfall and temperature on wheat yields in Australia. Climatic Change, 117, 163179, doi:10.1007/s10584-012-0543-0.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Price, D. T., D. W. McKenney, L. A. Joyce, R. M. Siltanen, P. Papadopol, and K. Lawrence, 2011: High-resolution interpolation of climate scenarios for Canada derived from general circulation model simulations. National Resources Canada (Northern Forestry Centre) and U.S. Forestry Service Information Rep. NOR-X-421, 104 pp.

  • Qian, B., R. De Jong, R. Warren, A. Chipanshi, and H. Hill, 2009a: Statistical spring wheat yield forecasting for the Canadian prairie provinces. Agric. For. Meteor., 149, 10221031, doi:10.1016/j.agrformet.2008.12.006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Qian, B., R. De Jong, and S. Gameda, 2009b: Multivariate analysis of water-related agroclimatic factors limiting spring wheat yields on the Canadian prairies. Eur. J. Agron., 30, 140150, doi:10.1016/j.eja.2008.09.003.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Robertson, G. W., 1974: Wheat yields for 50 years at Swift Current, Saskatchewan in relation to weather. Can. J. Plant Sci., 54, 625650, doi:10.4141/cjps74-112.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saiyed, M., P. R. Bullock, H. D. Sapirstein, G. J. Finlay, and C. K. Jarvis, 2009: Thermal time models for estimating wheat phenological development and weather-based relationships to wheat quality. Can. J. Plant Sci., 89, 429439, doi:10.4141/CJPS07114.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sauchyn, D., and S. Kulshreshtha, 2008: Prairies. From Impacts to Adaptation: Canada in a Changing Climate 2007, D. S. Lemmen et al., Eds., Government of Canada, 275–328.

  • Sauchyn, D., and Coauthors, 2009: Saskatchewan’s natural capital in a changing climate: An assessment of impacts and adaptation. Saskatchewan Ministry of the Environment from the Prairie Adaptation Research Collaborative, 165 pp. [Available online at http://www.parc.ca/pdf/research_publications/pp2009-02_web.pdf.]

  • Schlenker, W., and M. J. Roberts, 2006: Nonlinear effects of weather on corn yields. Rev. Agric. Econ., 28, 391398, doi:10.1111/j.1467-9353.2006.00304.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schlenker, W., and M. J. Roberts, 2009: Nonlinear temperature effects indicate severe damages to U.S. crop yields under climate change. Proc. Natl. Acad. Sci. USA, 106, 15 59415 598, doi:10.1073/pnas.0906865106.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schlenker, W., W. M. Hanemann, and A. C. Fisher, 2006: The impact of global warming on U.S. agriculture: An econometric analysis of optimal growing conditions. Rev. Econ. Stat., 88, 113125.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, E. G., and D. L. Young, 2000: Requiem for summer fallow. Choices, 15, 2425.

  • Statistics Canada, 2013: Estimated areas, yield and production of principal field crops by small area data regions. Census of Agricultural Regions, CANSIM table 001-0071. Accessed 20 January 2015. [Available online at http://www5.statcan.gc.ca/cansim/a26?lang=eng&retrLang=eng&id=0010071&paSer=&pattern=&stByVal=1&p1=1&p2=-1&tabMode=dataTable&csid=.]

  • Statistics Canada, 2015: Estimated areas, yield, production and average farm price of principal field crops, in metric units, annual. CANSIM table 001-0010. Accessed 4 March 2015. [Available online at http://www5.statcan.gc.ca/cansim/a26?lang=eng&retrLang=eng&id=0010010&&pattern=&stByVal=1&p1=1&p2=-1&tabMode=dataTable&csid].

  • Tack, J., A. Harri, and K. Coble, 2012: More than mean effects: Modeling the effect of climate on the higher order moments of crop yields. Amer. J. Agric. Econ., 94, 10371054, doi:10.1093/ajae/aas071.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tack, J., A. Barkley, and L. L. Nalley, 2014: Heterogeneous effects of warming and drought on selected wheat variety yields. Climatic Change, 125, 489500, doi:10.1007/s10584-014-1185-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thornton, P. K., P. G. Jones, G. Alagarswamy, and J. Andresen, 2009: Spatial variation of crop yield response to climate change in East Africa. Global Environ. Change, 19, 5465, doi:10.1016/j.gloenvcha.2008.08.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Urban, D., M. J. Roberts, W. Schlenker, and D. B. Lobell, 2012: Projected temperature changes indicate significant increase in interannual variability of U. S. maize yields. Climatic Change, 112, 525533, doi:10.1007/s10584-012-0428-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Van Kooten, G. C., 1992: Economic effects of global warming on agriculture with some impact analyses for Southwestern Saskatchewan. Saskatchewan in a Warmer World: Preparing for the Future, E. E. Wheaton, V. Wittrock, and G. D. V. Williams, Eds., Saskatchewan Research Council, Publ. E-2900-17-E-92, 159–184.

  • Wang, J., R. Mendelsohn, A. Dinar, J. Huang, S. Rozelle, and L. Zhang, 2009: The impact of climate change on China’s agriculture. Agric. Econ., 40, 323337, doi:10.1111/j.1574-0862.2009.00379.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, J., E. Wang, and D. L. Liu, 2011: Modelling the impacts of climate change on wheat yield and field water balance over the Murray–Darling Basin in Australia. Theor. Appl. Climatol., 104, 285300, doi:10.1007/s00704-010-0343-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Weber, M., and G. Hauer, 2003: A regional analysis of climate change impacts on Canadian agriculture. Can. Public Policy, 29, 163179, doi:10.2307/3552453.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Williams, G. D. V., R. A. Fautley, K. H. Jones, R. B. Steward, and E. E. Wheaton, 1988: Estimating effects of climatic change on agriculture in Saskatchewan. Assessments in Cool Temperate and Cold Regions, M. L. Parry, T. R. Carter, and N. T. Konijin, Eds., Vol. 1, The Impact of Climatic Variations on Agriculture, Kluwer Academic Publishers, 219–379.

  • View in gallery

    Saskatchewan canola and spring wheat yield and seeded area, 1985–2014.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 919 833 80
PDF Downloads 794 725 62

Analyzing Temperature and Precipitation Influences on Yield Distributions of Canola and Spring Wheat in Saskatchewan

View More View Less
  • 1 Department of Urban Studies and Planning, Massachusetts Institute of Technology, Cambridge, Massachusetts
  • | 2 Agriculture and Agri-Food Canada, Pacific Agri-Food Research Centres, Summerland, British Columbia, Canada
  • | 3 Department of Agricultural and Applied Economics, University of Georgia, Griffin, Georgia
  • | 4 Department of Production Engineering, Warsaw University of Life Sciences, Warsaw, Poland
© Get Permissions
Full access

Abstract

The IPCC indicates that global mean temperature increases of 2°C or more above preindustrial levels negatively affect such crops as wheat. Canadian climate model projections show warmer temperatures and variable rainfall will likely affect Saskatchewan’s canola and spring wheat production. Drier weather will have the greatest impact. The major climate change challenges will be summer water availability, greater drought frequencies, and crop adaptation. This study investigates the impact of precipitation and temperature changes on canola and spring wheat yield distributions using Environment Canada weather data and Statistics Canada crop yield and planted area for 20 crop districts over the 1987–2010 period. The moment-based methods (full- and partial-moment-based approaches) are employed to characterize and estimate asymmetric relationships between climate variables and the higher-order moments of crop yields. A stochastic production function and the focus on crop yield’s elasticity imply choosing the natural logarithm function as the mean function transformation prior to higher-moment function estimation. Results show that average crop yields are positively associated with the growing season degree-days and pregrowing season precipitation, while they are negatively affected by extremely high temperatures in the growing season. The climate measures have asymmetric effects on the higher moments of crop yield distribution along with stronger effects of changing temperatures than precipitation on yield distribution. Higher temperatures tend to decrease wheat yields, confirming earlier Saskatchewan studies. This study finds pregrowing season precipitation and precipitation in the early plant growth stages particularly relevant in providing opportunities to develop new crop varieties and agronomic practices to mitigate climate changes.

Retired.

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author e-mail: Wojciech J. Florkowski, wojciech@uga.edu

Abstract

The IPCC indicates that global mean temperature increases of 2°C or more above preindustrial levels negatively affect such crops as wheat. Canadian climate model projections show warmer temperatures and variable rainfall will likely affect Saskatchewan’s canola and spring wheat production. Drier weather will have the greatest impact. The major climate change challenges will be summer water availability, greater drought frequencies, and crop adaptation. This study investigates the impact of precipitation and temperature changes on canola and spring wheat yield distributions using Environment Canada weather data and Statistics Canada crop yield and planted area for 20 crop districts over the 1987–2010 period. The moment-based methods (full- and partial-moment-based approaches) are employed to characterize and estimate asymmetric relationships between climate variables and the higher-order moments of crop yields. A stochastic production function and the focus on crop yield’s elasticity imply choosing the natural logarithm function as the mean function transformation prior to higher-moment function estimation. Results show that average crop yields are positively associated with the growing season degree-days and pregrowing season precipitation, while they are negatively affected by extremely high temperatures in the growing season. The climate measures have asymmetric effects on the higher moments of crop yield distribution along with stronger effects of changing temperatures than precipitation on yield distribution. Higher temperatures tend to decrease wheat yields, confirming earlier Saskatchewan studies. This study finds pregrowing season precipitation and precipitation in the early plant growth stages particularly relevant in providing opportunities to develop new crop varieties and agronomic practices to mitigate climate changes.

Retired.

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author e-mail: Wojciech J. Florkowski, wojciech@uga.edu

1. Introduction

The latest Intergovernmental Panel on Climate Change (IPCC 2014) report indicates that, without adaptation measures, global mean temperature increases of 2°C or more above preindustrial levels are expected to have negative effects on agricultural crops such as wheat, corn, and rice in both temperate and tropical regions of the world. Higher temperatures associated with decreasing relative humidity conditions can lead to severe drought and affect yield potential and impact crop production. Canadian climate model projection studies indicate a gradual decline in annual precipitation in the prairie province of Saskatchewan (Price et al. 2011) that is likely to affect agricultural crops (Lemmen and Warren 2004). Not only is unreliable precipitation likely to be a climate constraining factor, but the lengthening of the frost-free period is expected to influence crop choices and the timing of cultural practices in the crop-growing season of Saskatchewan (Grisé 2013).

Given the projected changes in annual precipitation and temperature in Saskatchewan, the aim of this paper is to estimate the effect of precipitation and temperature on crop yield distributions by a full- and partial-moment-based approach. Such moment-based methods are proposed as a flexible way to characterize and estimate the asymmetric relationships between climate variables and the higher-order moments of crop yields (Antle 2010). This study contributes to the earlier work by Antle (2010), Antle et al. (2013), and Schlenker and Roberts (2009) by treating precipitation and temperature as separable inputs consistent with the estimation approach by Ortiz-Bobea and Just (2013), who argue that monthly temperature and precipitation variables have different production effects on the output distribution. The full- and partial-moment-based approach is applied to yields of two major agricultural field crops (canola, spring wheat) in the province of Saskatchewan, to illustrate its flexibility to capture the variances and skewness effects of climate and nonclimate variables on the positive and negative yield distributions. Saskatchewan is considered Canada’s breadbasket, having some of the richest soils, and is a major producer of wheat and canola. In 2014, Saskatchewan’s production of spring wheat and canola totaled, respectively, 9.1 and 7.9 million tonnes or 43% and 48% of Canada’s spring wheat and canola production (Statistics Canada 2015). Saskatchewan is the leading Canadian field crop exporter with CAD $2.2 billion and $2.5 billion, respectively, of nondurum wheat and canola exports in 2014 (Government of Saskatchewan 2015). Climatic changes captured by the quantified effects of precipitation and temperature on crop yield distributions measure their economic significance for farmers, the province and international commodity markets.

In the next section we present a literature review, followed by the theoretical framework and data description in the third and fourth sections, respectively. In the fifth and sixth sections the estimation strategy and results are presented. The final section summarizes our conclusions and identifies areas for future research.

2. Literature review

Many econometric modeling studies have explored the effect of climate change on crop production. One of the earlier Canadian climate studies looked at the impact of weather conditions on wheat yields in western Canada (Hopkins 1935). Most of the earlier climate crop yield studies were by agronomists and meteorologists who analyzed how weather/crop yield effects varied over the crop growth life cycle. The two basic approaches adopted by agronomists included simulation–crop growth models (Jones et al. 2003; Lobell and Ortiz-Monasterio 2007; Qian et al. 2009a,b; Lobell and Burke 2010; Asseng et al. 2011; Wang et al. 2011; Özdoğan 2011; Urban et al. 2012; Potgieter et al. 2013) and regression/correlation analyses (Robertson 1974; McCaig 1997; Kutcher et al. 2010; He et al. 2013). Crop growth models required detailed plant physiological data, and in combination with simulated weather data from global circulation models, they were able to predict how crop yields such as wheat responded to climatic weather conditions.

The economic literature adopted the Just and Pope (1979) production function (Chen et al. 2004; Isik and Devadoss 2006; McCarl et al. 2008; An and Carew 2015) and hedonic models (Deschênes and Greenstone 2007; Mendelsohn and Reinsborough 2007; Wang et al. 2009) to analyze the impact of climate change on agricultural output and profit. Because of unreasonable restrictions in the Just–Pope modeling approach, recent empirical studies have adopted flexible moment-based approaches (Antle 1983, 2010) to analyze the effect of climate on the distribution of crop yields (Antle et al. 2013; Tack et al. 2014).

In general, the multiple empirical approaches adopted to study the effects of climate change on agricultural output have provided mixed results attributed partly to model specifications, weather data employed, and country location. In terms of temperature effects, global warming is expected to have a negative impact on global yields of wheat and maize (Lobell and Field 2007), with wheat and maize yields declining, respectively, by 5.5% and 3.8% (Lobell et al. 2011). These temperature yield reduction effects may offset some of the crop yield improvements achieved from technological progress. Employing a regression model that allows for spatial dependence in crop yields across counties, Chen et al. (2013) show that Chinese maize and soybean yields are expected to be adversely affected by higher temperatures by the end of the century with larger yield reductions for soybean than for maize. Flexible regression models employing finer-scale weather data reveal that temperature thresholds above 29°C (maize), 30°C (soybeans), and 32°C (cotton) can have harmful effects on crop yields (Schlenker and Roberts 2009). Apart from crop yield reductions, variability of crop yields is likely also to be impacted by warmer temperatures. Urban et al. (2012) find that, without adaptation, U.S. maize yield variability is expected to increase as a result of projected changes in temperature. Schlenker et al. (2006), employing a hedonic approach and a nonlinear transformation of the temperature variables, conclude that different warming temperature scenarios will result in a 10%–25% decrease in U.S. farmland values.

While quite a number of studies have concentrated on the adverse effects of global warming on crop yields or agricultural profits, a few studies employing cumulative growing season weather variables have examined the combined effects of temperature and precipitation on agricultural output. Climate model projections of future changes in temperature and precipitation suggest that uncertainties in growing season temperatures will have a greater impact on crop production relative to the changes in precipitation (Lobell and Burke 2008). Chen et al. (2004) found increases in precipitation decreased yield variability for U.S. wheat, maize, and cotton, while increasing sorghum production risk. Conversely, higher temperatures decreased cotton and sorghum yield variability but showed mixed results for wheat depending on the functional form employed.

High-latitude countries like Canada are likely to benefit from global warming, especially in the northern regions and the southern and central prairies (Agriculture and Agri-Food Canada 2014). While drier weather is projected to have the greatest impact on the Canadian prairies, in terms of expanding the growing season and the production of higher value crops such as soybeans (Weber and Hauer 2003), the major climate change challenges will be changes in water availability in the summer season, greater frequencies of droughts, and developing crop adaptation strategies (Sauchyn and Kulshreshtha 2008).

Our study employs Statistics Canada (2013) Saskatchewan crop yield and production data for 20 crop districts that differ in soil and climate characteristics. We employ partial-moment functions to test for asymmetric input effects on output and analyze how yield distributions for different crops (canola, spring wheat) respond to pregrowing season precipitation and monthly temperature and precipitation over the crop-growing season. When investigating the effect of climate variables on crop yields, important statistical features (autocorrelation and spatial correlation) were taken into account. However, some unobservable common factors (e.g., crop production practices, agricultural subsidies, access to resources, and ability to cope) cannot be quantitatively measured and thus are likely to result in complex patterns of spatial and temporal dependence of crop yield between the cross-sectional units (Thornton et al. 2009; Chen et al. 2013; Cole et al. 2013). Many empirical studies have ignored spatial correlation in climate variables, which may lead to severely biased standard error estimates. Therefore, to ensure that statistical inference is valid, appropriately adjusting for spatial and temporal autocorrelation of crop yields across crop districts is particularly important (Hoechle 2007). The benefits of this study will help policymakers and scientists develop improved adaptation strategies to lower downside risk vulnerability and mitigate yield variability from unpredictable weather events.

3. Theoretical framework

A stochastic production function is described as
e1
where yt represents crop yield, xt represents observed characteristics for climate and nonclimate variables, f(xi; β) = E[g(yi)|xi] is the mean function, and εt is a random disturbance term with the variance and skewness given as f2(xt; β2) > 0; and f3(xt; β2) (Di Falco and Chavas 2009).
Following Antle (2010) and Antle et al. (2013), crop district average crop yields follow a distribution g(yi,t | xi,t), where yi,t equals crop yield in crop district i and period t, and xi,t represents the corresponding observed and nonclimate variables. According to Tack et al. (2012), various functional forms of g(⋅) are related to different types of moment. Antle (1983) utilized the identity function g(yi,t) = yi,t and the model conditions the raw moment on explanatory variables. In contrast, Schlenker and Roberts (2006, 2009) utilized the natural logarithm function g(yi,t) = ln(yi,t), and this model conditions the natural logarithm moment on xi,t, while Tack et al. (2012) utilized the higher-power function g(yi,t) = jN and models the jth higher-order raw moments. In our study, the crop yield’s elasticity is our major interest, thus the natural logarithm function is adopted in this study. The mean function, which is transformed prior to estimating the higher-moment functions, is described as
e2
where the moments are assumed to be linear functions of the exogenous variables, while εi,t is a random error with mean zero. Equation (2) can be very flexible in terms of incorporating quadratic and interaction terms. Employing Eq. (2), the higher-moment function for crop yields is given as
e3
where the errors () are correlated across all equations and require correction for heteroskedasticity using weighted least squares or a heteroskedastic-consistent estimator (Antle 2010). One advantage of the moments-based model (2) is that it contains a different parameter vector βj, for each moment equation. According to Antle (2010), specification of the mean function is important to the properties of the higher-order-moment estimated residuals. However, one disadvantage of Eq. (3) is that it limits the effects that conditioning variables have on asymmetry related to the negative and positive deviations from the mean (Antle et al. 2013). To address the asymmetric limitations of the full-moment model, Antle (2010) and Antle et al. (2013) employed Eq. (3) to derive the partial-moment model, which is described as
e4
e5
where βj,n and βj,p are parameters that can characterize the effects of xi,t on the negative and positive sides of yield distribution in the jth higher-moment equation, respectively; and and are the corresponding error terms in the negative and positive sides of distribution equations.
The empirical model for the full-moment model is described as
e6
where is the jth moment function for average crop yield (canola, spring wheat) in crop district i and period t, SDAi,t equals the seeded area (canola, spring wheat), SFAi,t is the share of summer fallow area or management measure to conserve soil moisture for the following year’s crop, TEMi,t equals air temperature or growing degree-days during the growing season (May–September), PPREi,t equals precipitation during the pregrowing season (October–April) that captures moisture previously stored in the soil from snowfall, PREi,t equals precipitation during the growing season (May–September), XHTi,t equals the number of excessive heat days during the growing season with temperatures greater than 30°C, and TIMi,t is a time trend variable that captures technological (e.g., new varieties adopted) and agronomic management improvements.

4. Data description and sources

The agriculture sector in Saskatchewan is sensitive to the effects of climate change, with the southern region of the province being more susceptible to fluctuations in summer precipitation (Grisé 2013). Agriculture production in Saskatchewan has changed over the years with diversified cropping systems and larger planted areas devoted to pulses (peas, lentils) and canola. This has been facilitated in part by the adoption of zero- or minimum-tillage technological practices. By 2008, over 50% of the seeded area in Saskatchewan was devoted to zero tillage (Nagy and Gray 2012). While the adoption of zero-tillage practices offers several environmental and agronomic benefits (e.g., soil conservation), it has contributed partly to the reduction in soil moisture conserving practices, like summer fallow, which declined from 5.9 million ha in 1987 to 1.1 million ha in 2014 (Statistics Canada 2015). In the 1980s, summer fallow as a conservation practice typically occurred in the drier areas of the province to increase soil water reserves and was used primarily by wheat producers (Williams et al. 1988).

This study is based on a comprehensive field crop dataset collected by Statistics Canada (2013) on the total annual crop area seeded/harvested, summer fallow area, yield, and production of all the major crops grown in Canada by province at the crop district level. The time period coverage selected for this study was based on the availability of comparable spring wheat and canola yield, planted area, and summer fallow area data for the 20 crop districts over the 1987–2010 period. Saskatchewan’s crop districts are located in three distinct agroclimatic zones (subhumid, semiarid, and arid) that correspond to the black/dark-gray, dark-brown, and brown soils (Mkhabela et al. 2011). Because of lower annual precipitation in the brown soil zones located primarily in the southwestern part of the province, crop yields tend to be lower than yields in the black/dark-gray or dark-brown soil zones (Table 1). Generally, crop districts in the southern areas of the province experience the warmest winter and summer months, while the northern part of the province receives the highest annual precipitation (Grisé 2013). By 2100, increases in maximum temperature in the semiarid zone of the prairies are projected to increase from 2.5° to 4.5°C, coupled with increases in interannual variation in annual precipitation (Price et al. 2011). Saskatchewan’s agriculture production vulnerability is attributed to extreme environmental variations from such unpredictable climatic conditions.

Table 1.

Average canola and spring wheat yield (kg ha−1) by crop districts (CDs; see online at http://www.saskseed.ca/images/varieties2016.pdf) for 1987–2010. Brown’s test suggests that we cannot reject the equality of variances between areas. Source: Statistics Canada (2013).

Table 1.

Figure 1 shows how canola and spring wheat yield and seeded area have varied over the years 1985–2014. Crop yields have increased over time, which may be attributed to a combination of continuous cropping, improved genetics, better agronomic management practices, and favorable climatic conditions. Despite the significant positive spring wheat yield trends, seeded area decline has been much more precipitous than canola. The spring wheat seeded area decrease over the last three decades has been attributed, in part, to the introduction of new crops like pulses and canola into crop rotations coupled with higher relative commodity prices (Grisé 2013).

Fig. 1.
Fig. 1.

Saskatchewan canola and spring wheat yield and seeded area, 1985–2014.

Citation: Journal of Applied Meteorology and Climatology 56, 4; 10.1175/JAMC-D-16-0258.1

The climate of Saskatchewan can be characterized as consisting of long cold winters, warm summers, and insufficient precipitation during the growing season (Williams et al. 1988). However, recent trends suggest that Saskatchewan is getting less cold associated with a greater increase in daily minimum temperatures and warming temperatures in the winter and early spring (Sauchyn et al. 2009). Summary weather statistics for the 20 crop districts are shown in Table 2. Weather data from Environment Canada meteorological weather stations included actual daily observations and modeled data (gridded data). The nearest grid points from 10-km gridded data were used to fill any missing observations. The weather data for crop districts provided by Agriculture and Agri-Food Canada were for the following weather data categories: average daily temperatures, minimum temperature, maximum temperature, and total daily precipitation (Chipanshi 2013).

Table 2.

Descriptive statistics of climate and nonclimate variables for canola and spring wheat, 1987–2010.

Table 2.

Apart from pregrowing season precipitation, cumulative growing season precipitation, and cumulative growing season growing degree-days (GDD), intraseasonal weather variables for growing season precipitation and GDD were constructed since studies (e.g., Robertson 1974; Kutcher et al. 2010) have shown that canola and spring wheat phenological crop growth stages1 respond differently to seasonal weather conditions. Since the development and growth stages of spring wheat follow a monthly pattern, Robertson (1974) considered the monthly averages of weather in measuring the response of Saskatchewan spring wheat to seasonal weather climatic patterns using field-plot experimental conditions. Moisture from growing season precipitation and the amount of rainfall in the months preceding the growing season are the principal climatic factors influencing wheat production in the prairies (Ash et al. 1992; Van Kooten 1992). In our study, GDD is calculated as the sum of positive values of the average [(maximum + minimum)/2] daily air temperatures minus the minimum temperature (5°C) required for growth (Campbell et al. 1997). Crops like canola and wheat with lower threshold growth temperatures (5°C) tend to have lower spatial variability throughout the eastern prairies (Ash et al. 1993).

GDD measures the combined effects of temperature and growing season length and provides a useful approach for estimating wheat phenological development (Saiyed et al. 2009). Since GDD does not adequately account for the effects of extreme temperatures, we constructed another weather variable (number of days in the growing season with temperatures > 30°C) to capture extreme heat days. It is suggested that summer daytime temperatures exceeding 30°C can adversely affect flowering and reproductive growth of crops in the Canadian prairies (Bueckert and Clarke 2013).

5. Estimation strategy

As part of our estimation strategy, prior to estimating the full- and partial-moment model we undertook several diagnostic tests. First, we tested for stationarity of the variables in the mean equation employing the Im–Pesaran–Shin (IPS; Im et al. 2003) unit-root test. The IPS test allows for unbalanced panel data where the null hypothesis is that the panels contain a unit root. Table 3 shows the results of the IPS unit-root test for canola and spring wheat, which indicates that all the variables are stationary or integrated of order zero [I(0)]. This implies that the null hypothesis of a unit root is rejected. Second, we tested for autocorrelation by employing the Wooldridge test where the null hypothesis is that there is no first-order autocorrelation in the panel data. The F statistic (p value) in Table 4 indicates that the null hypothesis is rejected, and therefore the moments-based model is applied to the autocorrelation-transformed data.

Table 3.

IPS (Im et al. 2003) panel unit-root test results for canola and spring wheat yields, climate and nonclimate variables. IPS allows unbalanced panel data, and the statistic is employed because of fixed time period. Cross-sectional mean is removed. The null hypothesis is all panels contain unit roots.

Table 3.
Table 4.

Test for autocorrelation in the panel data model (the F statistic is shown, with the p value in parentheses). This is a Wooldridge test, with degrees of freedom 1 of 1 and degrees of freedom 2 of 19. The null hypothesis is that no first-order autocorrelation exists in the panel data. Models 1 and 2 are full-moment models using cumulative and intrasectional climate variables, respectively.

Table 4.

Another test undertaken was to determine if the panel data have fixed or random regional effects. The Sargan–Hansen statistics test (Table 5) rejects the null hypothesis that the coefficients from the random effects are consistent with the coefficients from the fixed effects model. The test result indicates the existence of fixed effects that were used in the estimation of the mean function. We tested for heteroscedasticity prior to estimating the mean function. The modified Wald test results (Table 6) rejected the null hypothesis of groupwise homoscedasticity in the fixed effects regression. Consequently, the heteroskedasticity-consistent standard errors are estimated in the mean equation.

Table 5.

Test of fixed effect vs random effect in the panel data model. The Sargan–Hansen statistic is reported. The null hypothesis is that coefficients from random effect are consistent to coefficients from the fixed effect. Models 1 and 2 are full-moment models using cumulative and intrasectional climate variables, respectively. Here, DF is degrees of freedom.

Table 5.
Table 6.

Modified Wald test of groupwise heteroskedasticity for crop yield. The null hypothesis is groupwise homoskedasticity in fixed effects regression model. Models 1 and 2 are full-moment models using cumulative and intrasectional climate variables, respectively.

Table 6.

The Pesaran’s test employed rejected the null hypothesis of cross-sectional independence (Table 7), which indicates the standard errors in the mean equation are adjusted in estimating for cross-sectional dependence. Furthermore, in the model specification of the mean function, nonlinear quadratic weather terms and interaction weather terms were tried but contributed to multicollinearity (Table 8), and were subsequently deleted. The estimates reported in Tables 912 are the parameter elasticities.

Table 7.

Pesaran’s test of cross-section correlation for crop yield. The null hypothesis is cross-sectional independence. Models 1 and 2 are full-moment models using cumulative and intrasectional climate variables, respectively.

Table 7.
Table 8.

Test for multicollinearity in the panel data model using mean variance inflation factor (VIF). Because all of the mean VIF are smaller than 10 there is no multicollinearity. Models 1 and 2 are full-moment models using cumulative and intrasectional climate variables, respectively.

Table 8.
Table 9.

Full-moment functions (first, second, third) results: canola and spring wheat yield (using cumulative climate variables). Prais–Winsten regression; heteroskedastic panels corrected standard errors (PCSEs) in parentheses for mean equation.

Table 9.
Table 10.

Full-moment functions (first, second, third) results: canola and spring wheat yield (using intraseasonal climate variables). Prais–Winsten regression; heteroskedastic PCSEs in parentheses for mean equation.

Table 10.
Table 11.

Partial-moment functions (second and third): canola and spring wheat (using cumulative climate variables). Clustered robust standard errors are in parentheses. The mean equations report the Driscoll–Kraay standard errors adjusted for spatial dependence.

Table 11.
Table 12.

Partial-moment functions (second and third): canola and spring wheat (use intraseasonal climate variables). Clustered robust standard errors are in parentheses.

Table 12.

6. Results and discussion

As discussed in the theoretical framework section, the full-moment model captures how factors have different effects on the major moments of crop yield distribution (i.e., mean, variance, and skewness), while the partial-moment model provides more flexibility and extends the full-moment approach by allowing the asymmetric effects of factors on the two tails of yield distribution (positive and negative deviations from the mean). It is notable that this study utilized the natural logarithm function (discussed in the theory section above), thus full and partial moments discussed in the following section are actually the natural logarithm moments of crop yield. Results of both the full-moment model and partial-moment model are displayed in Tables 912, with two model specifications (cumulative versus intraseasonal weather effects).

a. Full-moment model

Full-moment function (mean, variance, skewness) results for canola and spring wheat are shown in Table 9 (model 1) and Table 10 (model 2) for two alternative weather variable definitions. In model 1, weather pertains to the cumulative growing season, while weather in model 2 pertains to intraseasonal weather events. The full-moment results have a good fit and include significant climate variables, except the third-order full-moment functions for canola and spring wheat (Table 9). The odd-order full-moment functions often have a bad fit (Antle et al. 2013). As shown in Table 9, higher growing degree-days or heat units during the growing season increases the average yield of both canola and spring wheat, with a larger effect on canola yields. Specifically, a 10% increase in GDD enhances the mean yield of canola and spring wheat by 4.3% and 3.4%, respectively. Increasing GDD reduces canola yield variability but with little significant effect on the variance of spring wheat yield. Our results are in agreement with an earlier study, which reported that higher mean temperature increases winter wheat yield in the Pacific Northwest with an elasticity effect of 0.45 at the sample mean (Antle et al. 2013). Our results indicate the temperature stress variable (the number of days with growing season temperatures greater than 30°C) reduces both canola and spring wheat average yield and increases spring wheat yield variability. A 20% increase in the number of days with extremely high temperature (about three additional days) decreases average yields by 0.6% for canola (8 kg ha−1) and spring wheat (11.7 kg ha−1).

Pregrowing season precipitation increases both canola and spring wheat average yields (with elasticity effects of 0.10 and 0.12, respectively) and lowers their yield variance. These results are consistent with the findings of an earlier study (Williams et al. 1988) that found conserved soil moisture in the winter season was correlated with wheat yield in the Canadian prairies. Precipitation during the pregrowing season has a significant positive effect on the skewness of spring wheat yield, but this is not the case for canola yield. Therefore, increases in pregrowing season precipitation reduce downside risk vulnerability for spring wheat yield and help avoid crop failure.

Effects of the crop seeded or planted area differ for canola and spring wheat yield. An increase in canola seeded area lowers yield since more marginal land is cultivated as canola seeded area expanded over the years. The positive effect of spring wheat seeded area on average yield is consistent with results from previous Canadian studies for Ontario soybeans (Cabas et al. 2010). The share of summer fallow in total cropped area is significant and negatively affects the mean and variance of canola yield but positively influences yield skewness. The combination of adopting zero-tillage technologies, continuous cropping, diversified rotational cropping systems, and new cultivars appears to have displaced summer fallow as a moisture conserving measure since the mid-1960s (Smith and Young 2000). Mearns (1988), employing a similar technology variable (ratio of fallowed area to total sown area), found it impacts year-to-year variability of wheat yields in the U.S. Great Plains.

The time trend variable as a measure of technical improvements from improved crop genetics, fertilization, and management practices statistically increases the mean yields of both crops and lowers yield variance of canola with little significant effect on spring wheat yield variance. The increases in canola yield may be consistent with the rapid number of herbicide-tolerant/hybrid canola varieties adopted in the prairies since the mid-1990s (Canola Council of Canada 2015).

Table 10 shows the full-moment functions with a more detailed specification of weather variables to coincide with major stages of the crop growth cycle during the growing season. Overall, September GDD increases the average yield and reduces the yield variance for both canola and spring wheat; conversely, higher July GDD decreases the mean yield and increases the yield fluctuation for both crops; June GDD has a mixed effect on the mean of canola (positive) and spring wheat yield (negative); the positive effect of May and August GDD on the average yield is only significant for canola and spring wheat, respectively. In elasticity terms, a 10% increase of September GDD enhances the mean yield of canola and spring wheat by 2.8% and 3.3%, respectively. The negative effects of July GDD on crop yields are relatively large, with elasticity effects of 0.92 and 0.54 for canola and spring wheat, respectively. Our study results correspond to the observations from a previous study that showed that high temperatures in the month of July adversely affect the flowering period and consequently impact canola seed quality (Kutcher et al. 2010). For wheat, the month of July is associated with the flowering period stage of growth or kernel development (Robertson 1974). Increases in June GDD bolster the mean yield of canola by 2.1%, while decreasing the average yield of spring wheat by 3.0%. Meanwhile, GDD in June has a positive effect on the yield variability with a negative effect on the skewness of spring wheat output. May GDD significantly increases the average yield of canola (elasticity of 0.23) with a modest positive effect on the skewness (elasticity of 0.03) of wheat output. In contrast, the effects of August GDD are significant only on the mean yield (elasticity of 0.33) and variability of spring wheat but not for canola.

Extreme weather conditions, where the cumulative number of days in the growing season exceeded 30°C, negatively and significantly affect both canola and spring wheat yields. This result is consistent with a previous Mexican study in which a similarly defined heat stress variable was found to negatively affect mean wheat yields (Nalley et al. 2010). Higher temperatures affect the growth pattern of wheat during the growing season because, as temperature increases, there is a consequential reduction of soil moisture available for plant crop growth (Van Kooten 1992).

Precipitation in the early part of the growing season affects the mean, variance, and skewness differently for both canola and spring wheat. In general, positive effects of precipitation during pregrowing (October–April) and growing seasons (May–September) on the average crop yields have been confirmed. A 10% increase in pregrowing season precipitation increases the average yield of canola and spring wheat by 0.7% and 0.9%, respectively. May precipitation increases the average yield (elasticity of 0.09) and skewness for both crops while decreasing their yield variance. June precipitation increases the average yield of both canola and spring wheat but has little significant effect on the variance or skewness. Wheat experiences rapid development growth in June, which is the month with the highest precipitation, and is generally in the stem elongation and head emergence stage of growth by the end of June (Robertson 1974).

The positive effect of July precipitation on canola in mitigating yield loss is in agreement with the results of an earlier Saskatchewan study (Kutcher et al. 2010). Van Kooten (1992) found similar results to our study where the months of May and June precipitation positively affected spring wheat yield in southwestern Saskatchewan. September precipitation only has a significant positive effect on the variance and skewness of canola yield.

b. Partial-moment model

The second and third partial-moment functions for model 1 and model 2 are shown in Tables 11 and 12. The partial-moment functions are defined in absolute terms (parameter signs are opposite to full moments for the odd-order negative moments) and are based on deviations above (positive) and below (negative) the mean (Antle 2010). Likelihood ratio (LR) test statistics show that symmetry restrictions are rejected for both second and third partial-moment functions. This result shows that, unlike the full-moment functions, the partial-moment functions show different results for the second-order moments and provide a better specification for estimating higher-order moments.

The partial-moment function results (Table 11) differ with respect to climate and nonclimate effects on canola and spring wheat yield distributions. In general, the impacts of climate changes such as cumulative growing season GDD and extreme temperatures on the yield distribution of canola are only significant in the positive partial moments. The temperature stress (>30°C) variable increases the positive variability and skewness of canola and the negative variability of spring wheat yield, while cumulative GDD during the growing season reduces the positive variance and skewness of canola yield. For spring wheat yield, pregrowing season precipitation and temperature stress variable also have asymmetric effects on two tails of the yield distribution, but unlike canola, the impacts are only significant on the deviation below the mean yield (negative moments). Pregrowing season precipitation reduces the negative second- and third-order partial-moment functions for spring wheat. Increasing the number of extremely high-temperature days increases the yield variability of spring wheat, especially significantly at the deviation below the mean level.

Combining the results from both full- and partial-moment function estimation indicates that increased GDD during the crop-growing season reduces the overall fluctuation of canola yield (full moments), while such a significant impact originates from decreasing the variance of the deviation above the mean level of the yield distribution. In addition, the correlation between the extreme high temperature and the overall variance of canola yield is not significant, but such a relationship is confirmed on the upper tail of the distribution.

The partial-moment function results with intraseasonal weather variables to coincide with the crop growth stages are shown in Table 12. Overall, the partial-moment functions present similar results as the full-moment functions, but the effects of climate on the negative tail of both crop yield distributions are more significant and have a larger magnitude than on the positive tail. Temperature variance effects on yield distributions are more significant in the earlier part of the canola growing season when compared with spring wheat. For spring wheat the phenological growth stage of emergence occurs in mid-May, and by the end of the month it has reached the fifth-leaf or the beginning of internode elongation stage. By the end of June spring wheat is headed out, while in early July anthesis takes place followed by kernel development (Robertson 1974). GDD in May reduces the variability of both the positive and negative tail of yield distribution for canola, and the effect on the negative tail is slightly larger. For canola, it is usually seeded in the month of May with stem elongation and flowering beginning in late June to early July (Kutcher et al. 2010). Conversely, GDD in June and July increases the second positive and negative moments of spring wheat yield distributions. GDD in September reduces the variance of spring wheat yield distribution on both tails, and this is similarly the case for canola yield, but only significant on the negative tail. September is usually the ripening or harvest period for both spring wheat and canola. May precipitation reduces the negative variance for both crop (canola, spring wheat) yield distributions, and also reduces the positive variance of spring wheat yield. Precipitation in the pregrowing season decreases the variability of spring wheat yield on the negative tail.

7. Conclusions

Global warming is expected to affect the productivity of Saskatchewan agriculture and influence how agricultural producers adapt to the adverse effects. This study employed a flexible moments-based approach over the 1987–2010 period to analyze how canola and spring wheat yield distributions respond to changes in precipitation and temperature.

Results obtained from the full-moment functions with alternative specification of the weather variables show different effects of nonclimate and climate variables, with the latter dissimilarities particularly distinct. Full-moment functions show that the model where climate variables are disaggregated by months of the growing season provides detailed insights about GDD and precipitation effects on the yield distribution that are not captured in the model with weather variables cumulated over the growing season.

The incorporation of the monthly GDD and monthly precipitation during the growing season in the full-moment function discerns the effects throughout the crop growth cycle. For canola, GDD has both positive and negative effects, respectively, on the mean in May, June, and September and on the variance in May and September. However, the effect of July GDD lowers the mean and increases the variance, which offsets the positive outcomes in the other months of the growing season. Similarly, the effect on variance is positive and by far larger than the decreasing effect in May and September. Changing weather patterns and more frequent occurrences of hot weather in July will likely have a strong negative effect on average canola yields, while increasing its variance.

The disaggregation of GDD on average spring wheat yield and risk provided even more discerning weather insights than for canola output. The aggregated effect of GDD suggested an increase in the average wheat yield and a decrease in variance as the number of GDD increases during the growing season. By contrast, the model specification with five monthly GDD figures shows no discernable effect of GDD in May, but hot weather in June and July hurts the average spring wheat yield and increases its variance. Only after the initial plant growth has been completed, does warm weather contribute to an increase of average yield, while lowering its variance as indicated by the GDD effect in August and September.

The partial-moment functions provide further insights into the nonclimate and climate change variables. Specifically, the effects are made with respect to the average canola and spring wheat yields with the emphasis on the upside and downside influence. Overall, the model with aggregate climate variables explains canola yield better than spring wheat yield.

Among the climate change variables, the number of growing degree-days has the only sizable significant effect and reduces the positive variance and skewness of canola yield. The effect of hot weather has a similar directional effect, but of very small magnitude. For spring wheat, pregrowing season precipitation reduces the positive yield variance, while hot weather increases the positive yield variance by a small amount, which is twice the size of a similar effect on canola yields.

Clearly the partial-moment functions of the canola and wheat models that include disaggregated climate change variables provide far more insights than the version with aggregated climate change measures. GDD reduces the positive and negative variance of canola yield by similar amounts when such days occur in May. In subsequent months of the growing season, only the GDD in September reduces the negative variance and skewness of canola yield. Growing season precipitation reduces the negative variance of canola yield if it rains in May, but has the opposite (although minimal) effect in September. The number of exceedingly hot days and the time trend has similar effects on canola as in the model with aggregated climate variables.

The weather effects on spring wheat are far more acute. The monthly GDD in June and July increases both variances, but the negative variance is affected far more than the positive variance, especially in July, showing asymmetrical effects and consequences for wheat yields. However, GDD in September reduces both variances, especially the negative variance and skewness. In terms of precipitation, May precipitation decreases both variances, and the reduction of negative variance is particularly large given the critical role of moisture in this stage of crop growth. In addition, pregrowing season precipitation also reduces the negative variance.

In summary, results of the current study are focused on canola and spring wheat produced in Saskatchewan, Canada. Despite the regional nature of data used in the study, results contribute to the global literature on the effects of climate change. First, the study results support the use of highly disaggregated temperature and precipitation data. Second, the application of the full- and partial-moment functions to discern the effects on average yields and their variance suggests the use of the partial-moment function in future studies. The applied approach stresses the importance of accounting for spatial correlation, autocorrelation, and specifically asymmetrical response of yields of different crops to climate variables. Disregarding such effects can bias the estimates and increase the prediction error. The specific outcomes of this study also suggest considerably stronger effects of changing temperatures than precipitation, supporting findings of Lobell and Burke (2008). The effects of higher temperatures measured by the monthly GDD broaden insights of earlier studies of other world regions suggesting a decrease in wheat yields (Lobell and Field 2007; Lobell et al. 2011), while also confirming results of earlier studies focused on Saskatchewan, but using a different methodological approach (Kutcher et al. 2010). The current study finds that pregrowing season precipitation and precipitation in the early stages of plant growth are particularly relevant, supporting previous studies showing general effect of precipitation on wheat yield variability reduction in other regions (Chen et al. 2004) and specific field experimental studies on spring wheat in the Canadian prairies (He et al. 2013).

Acknowledgments

An earlier version of this study was presented at the Agricultural and Applied Economics Association meeting in 2016 (Meng et al. 2016), and much of the material published in this article first appeared under the label “Agricultural and Applied Economics Association (formerly the American Agricultural Economics Association) 2016 Annual Meeting, 31 July–2 August, Boston, Massachusetts” at the AgeconSearch website (ageconsearch.umn.edu).

REFERENCES

  • Agriculture and Agri-Food Canada, 2014: Impact of climate change on Canadian agriculture. Accessed 26 August 2015. [Available online at http://www.agr.gc.ca/eng/science-and-innovation/agricultural-practices/climate/future-outlook/impact-of-climate-change-on-canadian-agriculture/?id=1329321987305.]

  • An, H., and R. Carew, 2015: Effect of climate change and use of improved varieties on barley and canola yield in Manitoba. Can. J. Plant Sci., 95, 127139, doi:10.4141/cjps-2014-221.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Antle, J. M., 1983: Testing the stochastic structure of production: A flexible moment-based approach. J. Bus. Econ. Stat., 1, 192201, doi:10.1080/07350015.1983.10509339.

    • Search Google Scholar
    • Export Citation
  • Antle, J. M., 2010: Asymmetry, partial moments and production risk. Amer. J. Agric. Econ., 92, 12941309, doi:10.1093/ajae/aaq077.

  • Antle, J. M., J. E. Mu, and J. Abatzoglou, 2013: Climate and the spatial-temporal distribution of winter wheat yields: Evidence from U.S. Pacific Northwest. 2013 AAEA & CAES Joint Annual Meeting, Washington, DC, Agricultural and Applied Economics Association, 2617. [Available online at http://ageconsearch.umn.edu/bitstream/150393/2/AAEA_2013_JianhongMu_%232617.pdf.]

  • Ash, G. H. B., C. F. Shaykewich, and R. L. Raddatz, 1992: Moisture risk assessment for spring wheat on the eastern prairies: A water-use simulation model. Climatol. Bull., 26 (2), 6578.

    • Search Google Scholar
    • Export Citation
  • Ash, G. H. B., C. F. Shaykewich, and R. L. Raddatz, 1993: The biologically important thermal character of the eastern prairie climate. Climatol. Bull., 27 (1), 320.

    • Search Google Scholar
    • Export Citation
  • Asseng, S., I. Foster, and N. C. Turner, 2011: The impact of temperature variability on wheat yields. Global Change Biol., 17, 9971012, doi:10.1111/j.1365-2486.2010.02262.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bueckert, R. A., and J. M. Clarke, 2013: Annual crop adaptation to abiotic stress on the Canadian prairies: Six case studies. Can. J. Plant Sci., 93, 375385, doi:10.4141/cjps2012-184.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cabas, J., A. Weersink, and E. Olale, 2010: Crop yield response to economic, site and climatic variables. Climatic Change, 101, 599616, doi:10.1007/s10584-009-9754-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Campbell, C. A., F. Selles, R. P. Zentner, B. G. McConkey, S. A. Brandt, and R. C. McKenzie, 1997: Regression model for predicting yield of hard red spring wheat grown on stubble in the semiarid prairie. Can. J. Plant Sci., 77, 4352, doi:10.4141/P96-043.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Canola Council of Canada, 2015: Estimated acreage of HT and conventional canola, 1995–2010. Accessed 2 April 2015. [Available online at http://www.canolacouncil.org/markets-stats/statistics/estimated-acreage-and-percentage/.]

  • Chen, C.-C., B. A. McCarl, and D. E. Schimmelpfennig, 2004: Yield variability as influenced by climate: A statistical investigation. Climatic Change, 66, 239261, doi:10.1023/B:CLIM.0000043159.33816.e5.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, S., X. Chen, and J. Xu, 2013: Impacts of climate change on corn and soybean yields in China. 2013 AAEA & CAES Joint Annual Meeting, Washington, DC, Agricultural and Applied Economics Association, 149739. [Available online at http://ageconsearch.umn.edu/bitstream/149739/2/AAEA2013.pdf.]

  • Chipanshi, A., 2013: Tabulation weather data request. Agriculture and Agri-Food Canada, National Agro-Climate Information Service, Regina, SK, Canada, accessed 8 August 2013.

  • Cole, M. A., R. J. Elliott, T. Okubo, and Y. Zhou, 2013: The carbon dioxide emissions of firms: A spatial analysis. J. Environ. Econ. Manage., 65, 290309, doi:10.1016/j.jeem.2012.07.002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Deschênes, O., and M. Greenstone, 2007: The economic impacts of climate change: Evidence from agricultural output and random fluctuations in weather. Amer. Econ. Rev., 97, 354385, doi:10.1257/aer.97.1.354.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Di Falco, S., and J.-P. Chavas, 2009: On crop biodiversity, risk exposure, and food security in the highlands of Ethiopia. Amer. J. Agric. Econ., 91, 599611, doi:10.1111/j.1467-8276.2009.01265.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Government of Saskatchewan, 2015: Saskatchewan agricultural exports. Accessed 23 March 2015, 25 pp. [Available online at http://publications.gov.sk.ca/documents/20/83881-Saskatchewan%20Agriculture%20Exports%202015.pdf.]

  • Grisé, J., 2013: Effect of climate change on farmers’ choice of crops: An econometric analysis. M. S. thesis, Department of Bioresource Policy, Business and Economics, University of Saskatchewan, 143 pp.

  • He, Y., and Coauthors, 2013: Spring wheat yield in the semiarid Canadian prairies: Effects of precipitation timing and soil texture over recent 30 years. Field Crops Res., 149, 329337, doi:10.1016/j.fcr.2013.05.013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hoechle, D., 2007: Robust standard errors for panel regressions with cross-sectional dependence. Stata J., 7, 281312. [Available online at http://www.stata-journal.com/sjpdf.html?articlenum=st0128.]

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hopkins, J. W., 1935: Weather and wheat yields in western Canada. I. Influence of rainfall and temperature during the growing season on plot yields. Can. J. Res., 12, 306334, doi:10.1139/cjr35-027.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Im, K. S., M. H. Pesaran, and Y. Shin, 2003: Testing for unit roots in heterogenous panels. J. Econom., 115, 5374, doi:10.1016/S0304-4076(03)00092-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • IPCC, 2014: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. C. B. Field et al., Eds., Cambridge University Press, 1132 pp. [Available online at http://www.ipcc.ch/pdf/assessment-report/ar5/wg2/WGIIAR5-PartA_FINAL.pdf.]

  • Isik, M., and S. Devadoss, 2006: An analysis of the impact of climate change on crop yields and yield variability. Appl. Econ., 38, 835844, doi:10.1080/00036840500193682.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jones, J. W., and Coauthors, 2003: The DSSAT cropping system model. Eur. J. Agron., 18, 235265, doi:10.1016/S1161-0301(02)00107-7.

  • Just, R. E., and R. D. Pope, 1979: Production function estimation and related risk considerations. Amer. J. Agric. Econ., 61, 276284, doi:10.2307/1239732.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kutcher, H. R., J. S. Warland, and S. A. Brandt, 2010: Temperature and precipitation effects on canola yields in Saskatchewan, Canada. Agric. For. Meteor., 150, 161165, doi:10.1016/j.agrformet.2009.09.011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lemmen, D. S., and F. J. Warren, Eds., 2004: Climate change impacts and adaptation: A Canadian perspective. Natural Resources Canada, 201 pp. [Available online at http://www.environment.msu.edu/climatechange/canadaadaptation.pdf.]

    • Crossref
    • Export Citation
  • Lobell, D. B., and C. B. Field, 2007: Global scale climate-crop yield relationships and the impacts of recent warming. Environ. Res. Lett., 2, 014002, doi:10.1088/1748-9326/2/1/014002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., and J. I. Ortiz-Monasterio, 2007: Impacts of day versus night temperatures on spring wheat yields: A comparison of empirical and CERES model predictions in three locations. Agron. J., 99, 469477, doi:10.2134/agronj2006.0209.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., and M. B. Burke, 2008: Why are agricultural impacts of climate change so uncertain? The importance of temperature relative to precipitation. Environ. Res. Lett., 3, 034007, doi:10.1088/1748-9326/3/3/034007.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., and M. B. Burke, 2010: On the use of statistical models to predict crop yield responses to climate change. Agric. For. Meteor., 150, 14431452, doi:10.1016/j.agrformet.2010.07.008.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobell, D. B., W. Schlenker, and J. Costa-Roberts, 2011: Climate trends and global crop production since 1980. Science, 333, 616620, doi:10.1126/science.1204531.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCaig, T. N., 1997: Temperature and precipitation effects on durum wheat grown in southern Saskatchewan for fifty years. Can. J. Plant Sci., 77, 215223, doi:10.4141/P96-144.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCarl, B. A., X. Villavicencio, and X. Wu, 2008: Climate change and future analysis: Is stationarity dying? Amer. J. Agric. Econ., 90, 12411247, doi:10.1111/j.1467-8276.2008.01211.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mearns, L. O., 1988: Technological change, climate variability, and winter wheat yields in the Great Plains of the United States. Ph.D. dissertation, University of California, Los Angeles, 223 pp.

  • Mendelsohn, R., and M. Reinsborough, 2007: A Ricardian analysis of US and Canadian farmland. Climatic Change, 81, 917, doi:10.1007/s10584-006-9138-y.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meng, T., R. Carew, W. J. Florkowski, and A. M. Klepacka, 2016: Modeling temperature and precipitation influences on yield distributions of canola and spring wheat in Saskatchewan. Agricultural and Applied Economics Association (formerly the American Agricultural Economics Association) 2016 Annual Meeting, Boston, MA, Agricultural and Applied Economics Association, 38 pp. [Available online at http://purl.umn.edu/235251.]

  • Mkhabela, M. S., P. Bullock, S. Raj, S. Wang, and Y. Yang, 2011: Crop yield forecasting on the Canadian prairies using MODIS NDVI data. Agric. For. Meteor., 151, 385393, doi:10.1016/j.agrformet.2010.11.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nagy, C., and R. Gray, 2012: Rate of return to the research and development expenditure on zero tillage technology development in western Canada (1960-2010). Department of Bioresource Policy, Business and Economics, University of Saskatchewan, 35 pp.

  • Nalley, L. L., A. P. Barkley, and A. M. Featherstone, 2010: The genetic and economic impact of the CIMMYT wheat breeding program on local producers in the Yaqui Valley, Sonora Mexico. Agric. Econ., 41, 453462, doi:10.1111/j.1574-0862.2010.00457.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ortiz-Bobea, A., and R. E. Just, 2013: Modeling the structure of adaptation in climate change impact assessment. Amer. J. Agric. Econ., 95, 244251, doi:10.1093/ajae/aas035.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Özdoğan, M., 2011: Modeling the impacts of climate change on wheat yields in northwestern Turkey. Agric. Ecosyst. Environ., 141, 112, doi:10.1016/j.agee.2011.02.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Potgieter, A., H. Meinke, A. Doherty, V. O. Sadras, G. Hammer, S. Crimp, and D. Rodriguez, 2013: Spatial impact of projected changes in rainfall and temperature on wheat yields in Australia. Climatic Change, 117, 163179, doi:10.1007/s10584-012-0543-0.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Price, D. T., D. W. McKenney, L. A. Joyce, R. M. Siltanen, P. Papadopol, and K. Lawrence, 2011: High-resolution interpolation of climate scenarios for Canada derived from general circulation model simulations. National Resources Canada (Northern Forestry Centre) and U.S. Forestry Service Information Rep. NOR-X-421, 104 pp.

  • Qian, B., R. De Jong, R. Warren, A. Chipanshi, and H. Hill, 2009a: Statistical spring wheat yield forecasting for the Canadian prairie provinces. Agric. For. Meteor., 149, 10221031, doi:10.1016/j.agrformet.2008.12.006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Qian, B., R. De Jong, and S. Gameda, 2009b: Multivariate analysis of water-related agroclimatic factors limiting spring wheat yields on the Canadian prairies. Eur. J. Agron., 30, 140150, doi:10.1016/j.eja.2008.09.003.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Robertson, G. W., 1974: Wheat yields for 50 years at Swift Current, Saskatchewan in relation to weather. Can. J. Plant Sci., 54, 625650, doi:10.4141/cjps74-112.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saiyed, M., P. R. Bullock, H. D. Sapirstein, G. J. Finlay, and C. K. Jarvis, 2009: Thermal time models for estimating wheat phenological development and weather-based relationships to wheat quality. Can. J. Plant Sci., 89, 429439, doi:10.4141/CJPS07114.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sauchyn, D., and S. Kulshreshtha, 2008: Prairies. From Impacts to Adaptation: Canada in a Changing Climate 2007, D. S. Lemmen et al., Eds., Government of Canada, 275–328.

  • Sauchyn, D., and Coauthors, 2009: Saskatchewan’s natural capital in a changing climate: An assessment of impacts and adaptation. Saskatchewan Ministry of the Environment from the Prairie Adaptation Research Collaborative, 165 pp. [Available online at http://www.parc.ca/pdf/research_publications/pp2009-02_web.pdf.]

  • Schlenker, W., and M. J. Roberts, 2006: Nonlinear effects of weather on corn yields. Rev. Agric. Econ., 28, 391398, doi:10.1111/j.1467-9353.2006.00304.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schlenker, W., and M. J. Roberts, 2009: Nonlinear temperature effects indicate severe damages to U.S. crop yields under climate change. Proc. Natl. Acad. Sci. USA, 106, 15 59415 598, doi:10.1073/pnas.0906865106.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schlenker, W., W. M. Hanemann, and A. C. Fisher, 2006: The impact of global warming on U.S. agriculture: An econometric analysis of optimal growing conditions. Rev. Econ. Stat., 88, 113125.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, E. G., and D. L. Young, 2000: Requiem for summer fallow. Choices, 15, 2425.

  • Statistics Canada, 2013: Estimated areas, yield and production of principal field crops by small area data regions. Census of Agricultural Regions, CANSIM table 001-0071. Accessed 20 January 2015. [Available online at http://www5.statcan.gc.ca/cansim/a26?lang=eng&retrLang=eng&id=0010071&paSer=&pattern=&stByVal=1&p1=1&p2=-1&tabMode=dataTable&csid=.]

  • Statistics Canada, 2015: Estimated areas, yield, production and average farm price of principal field crops, in metric units, annual. CANSIM table 001-0010. Accessed 4 March 2015. [Available online at http://www5.statcan.gc.ca/cansim/a26?lang=eng&retrLang=eng&id=0010010&&pattern=&stByVal=1&p1=1&p2=-1&tabMode=dataTable&csid].

  • Tack, J., A. Harri, and K. Coble, 2012: More than mean effects: Modeling the effect of climate on the higher order moments of crop yields. Amer. J. Agric. Econ., 94, 10371054, doi:10.1093/ajae/aas071.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tack, J., A. Barkley, and L. L. Nalley, 2014: Heterogeneous effects of warming and drought on selected wheat variety yields. Climatic Change, 125, 489500, doi:10.1007/s10584-014-1185-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thornton, P. K., P. G. Jones, G. Alagarswamy, and J. Andresen, 2009: Spatial variation of crop yield response to climate change in East Africa. Global Environ. Change, 19, 5465, doi:10.1016/j.gloenvcha.2008.08.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Urban, D., M. J. Roberts, W. Schlenker, and D. B. Lobell, 2012: Projected temperature changes indicate significant increase in interannual variability of U. S. maize yields. Climatic Change, 112, 525533, doi:10.1007/s10584-012-0428-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Van Kooten, G. C., 1992: Economic effects of global warming on agriculture with some impact analyses for Southwestern Saskatchewan. Saskatchewan in a Warmer World: Preparing for the Future, E. E. Wheaton, V. Wittrock, and G. D. V. Williams, Eds., Saskatchewan Research Council, Publ. E-2900-17-E-92, 159–184.

  • Wang, J., R. Mendelsohn, A. Dinar, J. Huang, S. Rozelle, and L. Zhang, 2009: The impact of climate change on China’s agriculture. Agric. Econ., 40, 323337, doi:10.1111/j.1574-0862.2009.00379.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, J., E. Wang, and D. L. Liu, 2011: Modelling the impacts of climate change on wheat yield and field water balance over the Murray–Darling Basin in Australia. Theor. Appl. Climatol., 104, 285300, doi:10.1007/s00704-010-0343-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Weber, M., and G. Hauer, 2003: A regional analysis of climate change impacts on Canadian agriculture. Can. Public Policy, 29, 163179, doi:10.2307/3552453.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Williams, G. D. V., R. A. Fautley, K. H. Jones, R. B. Steward, and E. E. Wheaton, 1988: Estimating effects of climatic change on agriculture in Saskatchewan. Assessments in Cool Temperate and Cold Regions, M. L. Parry, T. R. Carter, and N. T. Konijin, Eds., Vol. 1, The Impact of Climatic Variations on Agriculture, Kluwer Academic Publishers, 219–379.

1

For canola, it is seeded in May with stem elongation and flowering occurring in late June to early July with swathing taking place in mid-August to early September followed by harvesting in end of September (Kutcher et al. 2010). In contrast, for spring wheat, seeding, jointing, and heading takes place in May, June, and July, respectively. The phenological growth stage of soft dough and harvest takes place in early August and late August/September, respectively (Qian et al. 2009a).

Save