Observing the Diurnal Cycle of Corn Canopy Water Storage with Destructive and Microwave Measurements

The diurnal cycle in crop water tells us about stress; a healthy crop will lose substantial water during midday hours with transpiration and recharge its water through the roots at night while a water stressed crop will shut down, closing its stomata to conserve water, and have a relatively small change in diurnal water mass storage. Observing crop health in this way can give us estimates of carbon assimilation, which corresponds to yield that travels worldwide from the US Corn Belt. If this method of radio link measurement of vegetation water storage proves to be effective, it could develop into a means of monitoring crops and other vegetation on smaller scales more relevant to land surface processes and management activities than can be observed by satellite. Vegetation moisture quantity is seldom directly measured due to the large time and effort requirement of the process. Microwave remote sensing can sense water in vegetation; microwave radiation rotates groups of liquid water molecules and transports salt ions, inducing current, drawing power from the wave. The more water in vegetation the more a wave is attenuated (weakened) as it passes through. Thus far validation of these measurements has been mostly confined to discrete times and locations, attempting to validate satellite measurements with resolutions of multiple kilometers with point observations. A few other studies have utilized remote sensing from ground-based towers, but the destructive sampling validation is lacking. Our study aims to validate the absolute quantity and diurnal change in corn tissue water observed with a microwave radio link with destructive vegetation measurements within one field.

On August 3, 2022, we sampled vegetation throughout the day while our radio link system was recording signal strength in a central Iowa corn field. At Eleven periods from 6 am to midnight crops were sampled. We sampled 60 plants each period, finding their whole fresh mass in the field immediately after cutting. Plants were later dried to measure their dry mass (subtracting dry mass from fresh mass reveals water mass). A microwave transmitter-receiver pair was installed within the field, continuously measuring the propagation attenuation at frequencies of 900 MHz to 5 GHz. Meteorological measurements were also recorded on site. Dielectric models of vegetation state that for a crop in the early reproductive stages, microwave attenuation will increase with increasing temperature at frequencies less than ~2.5 GHz. The opposite is true at frequencies above ~2.5 GHz: we expect attenuation to decrease as temperatures increase. To measure the diurnal cycle in crop water storage, this attenuation temperature dependence must be filtered out, hence the need to measure at frequencies above and below 2.5 GHz as days with no diurnal temperature cycle are extremely rare during the growing season and would likely be feature heavy cloud cover and reduced transpiration.
Vegetation temperature was computed from upwelling longwave radiation. Central Iowa was under abnormally dry conditions at the time of this sampling day, but there was still enough water available in the soil for vegetation to utilize without greatly hindering the diurnal cycle in crop water quantity, as the day featured many hours of mostly sunny conditions (after cloudy conditions in the morning) at a typical vapor pressure deficit and light winds. This day also was a rarity in that there was no dew, allowing for accurate measurement of vegetation fresh mass. We hypothesize that after filtering for temperature that this diurnal cycle will be apparent for all measured frequencies, and that water mass will reach a maximum in the early morning (around dawn) and minimum in the early afternoon.

Our preliminary results indicate that vegetation water fraction held steady in the morning around 0.76. Between 11 am and noon the water mass fraction dropped significantly to around 0.75. At 5 pm water mass fraction jumped back up to near 0.76 and remained near there for the rest of the samples. These direct measurements confirm our hypothesis that water mass reaches minimum in the early afternoon, but minimum water content occurred over a period of several hours in the morning, which was unexpected. Overall, increasing frequency reduced radio signal strength received, which is believed to be a result of increased scattering as smaller wavelengths near 5 GHz are closer in size to corn leaves and ears. Some frequencies show an increase in signal strength just before noon. We will present a full analysis of the data at the conference Our method of validation is very effort intensive. We hope our work will lead to a more feasible approach of using radio wave propagation.