Weather and Management Influence Cropland Evapotranspiration in Manitoba

After precipitation input, the water vapour that returns to the atmosphere as a result of surface evaporation and crop transpiration, or evapotranspiration (ET), is the most significant component governing the water balance of cropland and an important ecosystem service. ET often exceeds growing season precipitation on the Canadian Prairies and soil water availability, besides nitrogen, remains one of the most limiting factors for crop production in the region. On the other hand, moisture conditions can be highly variable on the Prairies and excess moisture following extreme precipitation events (particularly on the eastern Prairies) has also been a problem in recent years leading to waterlogging, late or no seeding, poor emergence, and total crop losses in the late spring and early summer. For example, over the past decade in the province of Manitoba payment on post-harvest crop insurance claims for yield loss by excess moisture (drowning) has been almost double that for drought. Recent trends and projected climate scenarios suggest that precipitation increases can be expected during the early growing season in the region. For this reason it is particularly important to further elucidate the role of ET in regulating the water balance of cropland using contemporary micrometeorological measurements and estimates. These will help assess the role that ET plays within the hydrological cycle presently and under anticipated climate change scenarios. As well, it is important to gain understanding of the impact that changes in land management could have on ET and the cropland soil water balance from both drought and flood mitigation perspectives.

The present study reports on the results of four years (2012, 2013, 2014 and 2015) of growing season (1 June to 30 September) ET data from two annual crop fields located on the same organic farm in the Assiniboine River Basin of southwestern Manitoba, Canada. Using the energy balance residual technique, actual ET (ET_a) was estimated as the latent energy flux calculated from the difference between net radiation (R_n) and the measured sensible and ground heat flux densities (H and G, respectively). A sonic anemometer-thermometer was used to directly measure H in each field using eddy covariance. Duplicated measurements of R_n were made in separate locations in each of the field's flux footprints, and G was estimated with duplicated arrays of heat flux plates and soil temperature/moisture probes. Weather data was collected on site and used to calculate reference ET (ET₀) based on the FAO56 Penman-Monteith equation. Actual crop coefficient (K_a) values were estimated as ET_a / ET₀.

The four growing seasons had contrasting climatic conditions, particularly those experienced in 2012 and 2014. In 2012, the site had approximately 50% of typical growing season precipitation and an average air temperature that was 1.6°C higher than normal (1981-2010). In 2014, there was over 140% of normal precipitation and an average air temperature close to the 30-year mean. The growing season conditions in 2013 and 2015 were much closer to normal. As a result, the median daily ET₀ for the 2012 growing season (4.6 mm/day) was significantly higher (p < 0.05) compared to the other three years (3.6 to 3.9 mm/day) of the study which were statistically similar. However, differences in median daily ET_a for the four growing seasons were only significantly different for one of the fields at the site between the 2015 (2.9 mm/day) or 2013 (3 mm/day) and the 2012 (1.9 mm/day) growing seasons. Contrasting seasonal patterns of daily R_n and ET_a were evident within growing seasons between the two fields at the farm and were significantly different according to paired statistical tests for two and three of the four years, respectively. These differences in ET_a between the two fields on the farm represent the impact of different crops (short season vs. long season, winter vs. summer annuals) and the timing of seeding and harvest operations.   

Average cumulative ET_a for the two fields at the site was 306 mm, 329 mm, 321 mm, and 360 mm for the 2012, 2013, 2014, and 2015 growing seasons which received 143 mm, 284 mm, 370 mm and 221 mm of rain, respectively. The median daily K_a (ET_a / ET₀) values were significantly different between 2012 (0.61) and 2014 (0.73) or 2015 (0.72) for one field at the site, and between 2012 (0.46) and all three (0.72 – 0.78) of the later growing seasons for the second field. The consistent reduction in ET_a relative to ET₀ in 2012 highlights the dominant role that lower than normal precipitation and reduced soil moisture played in limiting crop water use relative to the other years of the study. The excess moisture at the site in 2014 seems to have resulted in daily and cumulative seasonal ET_a rates similar to more normal years. However, it was likely the result of higher evaporation during waterlogged periods relative to transpiration rates as crop performance and biomass production were negatively impacted and seasonal differences in ET_a between the two fields present during the other years were muted in 2014. Implications of the research for the soil water balance at the site and possible impacts of future droughts and extreme precipitation events will be presented and discussed.