Written by Basavaraj Amogi, Gwen Hoheisel, and Lav Khot, Washington State University and AgWeatherNet
What is an inversion
Eastern Washington fruit production occurs in a unique landscape between two mountain regions: the Cascades to the West and the Rockies to the East. In this low-lying region, inversions occur very naturally during the winter months.
Air is typically warmer near the surface and gradually cooler at higher altitudes. An inversion occurs when a layer of warm air settles above a cooler air mass near the ground (Fig. 1). Fruit farms on hillsides can experience distinct temperature gradients due to variations in both elevation along the slope of hill and atmospheric layering. During these inversions, wind machines can be very effective at mixing the air and reducing the risk of frost damage to fruit crops.
There are several indicators of an inversion. Clear skies, calm wind near sunrise or sunset, and the appearance of dew or frost on foliage (which occurs if canopy cools to its dewpoint) all suggest inversion conditions. Fog or smoke that remains close to the ground rather than dispersing also indicates a lack of air mixing. Inversions usually forms in the evening hours as air temperature drops and then dissipate the following morning once the sun warms the surface air, provided wind speeds remain low — generally below 3 – 5 mph.
Figure 1: Pictorial explanation of an inversion in a fruit farm with warm air trapping cool air near the ground.
Understanding inversion strength to optimize wind machine operations
Inversion strength refers to the temperature difference between a warmer air layer above and the cooler air at, or just above the canopy height. For example, the difference in air temperature at 30 feet above canopy and 6 feet above ground. A larger temperature difference indicates a stronger inversion.
Stronger inversions offer greater potential for warming by mixing the air. Wind machines are typically most effective when inversion strength is at least 2oF, as they typically raise the surface air temperatures by about 50% of the inversion strength. For example, if the air at 30 feet is 3oF warmer than at 6 feet, a wind machine can raise the surface temperature by about 1.5oF. Wind machines also perform best when the warm air is located between 15 and 75 feet above the canopy. However, if surface air is already well mixed by wind or insulated by thick clouds, the benefit of operating wind machines becomes very minimal.
Because inversion strength plays a crucial role in determining the effectiveness of wind machine operation during frost events, AgWeatherNet (AWN) has developed new tool to predict them using a tower station data and indicate when wind machine use may be most beneficial.
Predicting an inversion strength
Although some numerical weather prediction models from National Weather Service (NWS) often predict regional occurrences of inversion conditions, localized station-specific forecasts are often more helpful for growers. Hence, AWN researchers have developed machine learning models trained on NWS forecasts and historical tower station specific weather data, effectively reducing the forecast errors from NWS.
Comparisons between AWN’s station-specific forecasts and regional forecasts show that 86% of the predictions made by AWN fall within 1-2 °F of actual observations (Fig. 2). As with weather forecast models, prediction accuracies degrade over longer periods, meaning the prediction error can be as high as 4-5 oF with a 10-day forecast. Nonetheless, these station-specific forecasts remain valuable for estimating strength of an inversion for next 24 hours and determining optimal time for operating wind machines.
Figure 2. Distribution of AWN inversion forecast errors compared to actual observations across multiple stations. The black line represents the mean error, while the dashed lines indicate 1σ (68%), 2σ (95%), and 3σ (99.7%) confidence intervals.
Accessing inversion prediction tools at the AWN portal
AgWeatherNet web portal offers two primary ways to access inversion data in Washington State: (1) Map-Based Quick Access for observed inversion strength at AWN tower stations across the state, and (2) Forecasts for tower stations that have at least one year of historical weather data. To view observed inversion strength, go to the AWN homepage, scroll through the map interface, select “Inversion” at the top, and hover over a station icon to see current readings (Fig. 3). For forecast data, users can log in to AWN web portal, navigate to Weather → Weather Dashboard, select a relevant weather station, and open the “Air Temperature (30 FT)” tile for a detailed inversion forecast (Fig. 4).
Figure 3. AWN home screen with arrows on how to change the map view and select the inversion model. Hover on a number in the map to show the station and the inversion strength.
By clicking on the “Forecast” link (Fig. 4 insert), a more detailed view will appear (Fig. 5). This graph will show temperature at high elevation (30 feet) and near surface (6 feet). Warmer air must be present at higher elevation to mix the air with wind machines. If air is warmer at the surface, use of wind machines is not advised. Many fruit farms do use mid-row heaters to warm the surface air. The forecast does not account for that additional heat and the benefit it may provide in frost mitigation.
Figure 5: A more detailed view of the temperature change over time at high elevation and near surface to estimate inversion strength. If the temperature at 30 feet (orange line) is warmer than temperature at 6 feet, then ‘mixing could help’. But wind machine use is not advised if the opposite is true, cold air higher elevation and warm air near the surface.
Acknowledgments. This work has been supported in parts by the Washington State Department of Agriculture (WSDA) Specialty Crop Block Grant, NSF/USDA NIFA Cyber-Physical Systems (Award No: 2021-67021-34336), and USDA NIFA 0745 projects. We acknowledge prior modeling efforts by Dr. Matt Cann and model integration into the AWN web portal by Mr. Sean Hill and Mr. Aakash Shrestha.
Fruit Matters articles may only be republished with prior author permission © Washington State University. Reprint articles with permission must include: Originally published by Washington State Tree Fruit Extension Fruit Matters at treefruit.wsu.edu and a link to the original article.
