--REACCH Website   




Decision Support Tools


Choose a County:

REACCH Region Layer

Connecting Weather and Winter Wheat Yield

Yield Data

Visualize year-to-year variability in winter wheat yields over the past three decades and its relationship to variability in climate.

Yield Variability

Increases in yield are apparent across the Columbia basin due to advances in technology (management, varietals). To more clearly see the effects climate variability on yield, we remove the long-term yield trend indicated by the red line in the upper graph. Differences between annual yield and the trend line are shown in the lower graph where green bars indicate above-normal and orange bars indicate below-normal yields.

Climate-Yield Relationships

The variability in wheat yields can be visualized as a function of late spring (Apr-Jun) precipitation and temperature. The figure below shows each year as a bubble according to the precipitation and temperature observed during that year, with the bubble colored according to if the yield was above-normal (top 33%), normal (middle 33%), or below-normal (bottom 33%). You can add/remove all the bubbles of each yield category by clicking on the name in the legend.

About Figure

Future Climate

Projections of future climate variability over this county can be visualized by looking at a scatterplot of projections for late spring (Apr-Jun) precipitation and temperature from 20 global climate models for all years within a 30-yr future time period. Climate projections were bias corrected to local scales to facilitate use. The different models simulate climate differently leading to a spread of projections for each time period.


In the above figure, you can see that average Spring temperatures are generally projected to increase over the 21st century, while Spring precipitation is generally projected to stay the same.

REACCH Research

Graduate student Wenlong Feng from the Dept. of Geography, University of Idaho.
Graduate student, Wenlong Feng, (supervised by Dr. John Abatzoglou, UI) is studying the relationship between winter wheat yield and weather over the REACCH counties. Specifically, he is exploring the sensitivity of wheat yield during different phenological stages to weather (temperature, precipitation, actual evapotranspiration).


Wenlong adopted Ritchie’s (1991) growing degree day based wheat growth model to define the date boundaries for phenology stages, and then, extracted the weather data (4-km gridMET, Abatzoglou, 2012) in each phenology stage with these boundaries. In order to derive the wheat actual evapotranspiration, he used a growing degree day based FAO-56 model (Saadi et al. 2015) and a modified Thornthwaite water balance model (Willmott et al. 1985).

Wenlong assessed the climate impact on annual wheat yield variability using stepwise regression on detrended wheat yield (for spatial variation of climate-yields relationships) and panel linear regression (which assumes homogeneity in the climate-yield relationship).


Wenlong found that the most sensitive growth period for mean temperature and cumulative actual evapotranspiration is the period from flowering to maturity and the most sensitive growth period for cumulative precipitation is the period from booting to maturity.

Specifically, the panel model suggests that wheat yield decrease 47.6 kg/ha with increasing 1 degree Celsius in mean temperature from flowering to maturity; yield increase 2.9 kg/ha with increasing 1 mm in precipitation from booting to maturity; 1 mm increase of water deficit leads to 3.0 kg/ha decrease in yield; 1 degree day increase in heat degree day from flowering to maturity causes 1.9 kg/ ha decrease in yield. The actual evapotranspiration is insignificant in the panel model because it describes similar information as the precipitation and water deficit does. The panel model can explain 29.3% of the variance of the inter-annual yield changes.

There are three counties cannot be explained by the stepwise model: Yakima, WA; Benton, WA; and Idaho, ID. The minimum R-squared of the rest counties is 0.155, the maximum R-squared is 0.656, and the average value is 0.334.


Abatzoglou, J. T. (2013). Development of gridded surface meteorological data for ecological applications and modelling. International Journal of Climatology, 33(1), 121–131.

Ritchie, J. T. (1991). Wheat phasic development. In Modelling Plant and Soil Systems (Eds J. Hanks & J. T. Ritchie), pp. 31–54. Madison, WI: American Society of Agronomy

Saadi, S., Todorovic, M., Tanasijevic, L., Pereira, L. S., Pizzigalli, C., & Lionello, P. (2015). Climate change and Mediterranean agriculture: Impacts on winter wheat and tomato crop evapotranspiration, irrigation requirements and yield. Agricultural Water Management, 147, 103–115.

Willmott, C. J., Rowe, C. M., & Mintz, Y. (1985). Climatology of the terrestrial seasonal water cycle. Journal of Climatology, 5(6), 589–606.
Figure: Current Weather