About 20 years ago when I was a young undergraduate student, I started working on a project with the goal to develop the most accurate soil moisture sensor. Over the years, however, I have realized the complexity and difficulty of the task and that project is still ongoing. During my academic career, I also learned about other approaches to determine soil moisture and had the chance to investigate them.
It was in 2012 that I first became familiar with the work of Jackson et al. (1981) and USDA-ARS scientists at Bushland, TX on plant-based (thermal sensing) irrigation scheduling. Their papers were inspiring and changed my research path. Following in their footsteps, my team and I (at Washington State University w/ funding from METER Group) focused on plant-based water stress detection and irrigation scheduling. Below is a short summary of our efforts and what I see as the future of this research.
Objectives
In different studies since 2012, we have developed and evaluated various techniques for precision canopy and water management of crops through sensor-based decision making. We started by defining the following objectives:
create a site-specific irrigation control and monitoring system (we needed this system to continuously monitor water status of plants, determine their water requirements, and automatically schedule irrigation),
design and deploy a wireless network of soil, plant and microclimate sensors,
determine water requirements of the plants in real-time,
develop and assess irrigation scheduling algorithms based on plant-, soil-, and weather-based approaches, and
develop a sensor-based setup with plant-based models incorporated into its software (after calibration, this setup will be used for precise non-contact sensing of soil water content, soil water potential, and stem water potential in real-time by measuring canopy temperature and micrometeorological parameters)
Challenges
We wanted to use plant sensors instead of soil sensors to detect plant water stress and make irrigation decisions in apple tree orchards. But, the practice of using the crop water stress index (CWSI) for detecting water stress presented some challenges:
Available ET models like the Penman-Monteith did not explain stomatal regulations in most crops, and therefore the estimations were not reliable. Currently, scientists use either an empirical CWSI or a theoretical one developed using equations from FAO-56, but the basis for FAO-56 equations is alfalfa or grass, which isn’t similar to trees like apples. One of the main differences between grass and apple trees is that apple tree leaves are highly linked to atmospheric conditions. They control their stomata to avoid water loss.
automatic plant-based irrigation scheduling methods had never been used in apple trees and rarely in other tree crops.
One of the main differences between grass and apple trees is that apple tree leaves are highly linked to atmospheric conditions. They control their stomata to avoid water loss.
Crop-Specific Modeling Efforts
Therefore, we decided to develop our own crop water stress index, based on what we learned from the literature, physiology of apple trees and our research findings. We developed new models, water stress indices and algorithms for interpreting plant canopy signals to indirectly yet accurately determine soil water content. We developed two theoretical crop-specific models for estimating potential and actual transpiration of apple trees. We also managed to take our research further into developing several novel CWSI models, and an innovative computer-based irrigation-scheduling algorithm.
We developed a new theoretical crop water stress index specifically for apple trees. It accounts for stomatal regulations in apple trees using a canopy conductance sub-model.
We developed a new theoretical crop water stress index specifically for apple trees. It accounts for stomatal regulations in apple trees using a canopy conductance sub-model. It also estimates average actual and potential transpiration rates for the canopy area which is viewed by an infrared temperature (IRT) sensor or a thermal camera.
We established our new “Apple Tree” CWSI based on the energy budget of a single apple leaf, so “soil heat flux” was not a component in the modeling. The accuracy of this approach, however, greatly depends on the accuracy of the reference soil moisture measurement method. Because, we need to rely on soil data to fine-tune our plant-based models and algorithms. To establish a relationship between CWSI and soil water, we needed to measure soil water content in the root zone precisely. We used neutron probe (NP), which provides enough precision and volume of influence to meet our requirements. However, NP is a labor and time intensive method that does not allow for real-time measurements, posing a serious limitation. That is why we also used soil sensors to be able to monitor the soil more continuously.
To establish a relationship between CWSI and soil water, we measured soil water content in the root zone using a neutron probe.
Data Collection in the Field
To validate the models, we carried out measurements in several experimental and commercial orchards across Washington State in four different years. In one of the orchards, we measured soil water deficit using NP in the top 60 cm of the soil profile (root zone), and collected canopy surface temperature data using IRT sensors. We also carried out continuous measurements of soil water content and soil water potential using soil sensors in four additional orchards. Our team also measured microclimate data.
We carried out measurements in several experimental and commercial orchards across Washington State in four different years.
Results
The results were interesting. We found a strong correlation between the new CWSI model and soil water deficit in the root zone in apple trees in all of the orchards. The new model exhibited high sensitivity to mild variations in the soil water content (between Field Capacity and Management Allowable Deficit), suggesting it as a powerful indicator of water availability in the root zone. We expected to see correlations between plant and soil measurements, but such strong relationships were unexpected.
We found a strong correlation between the new CWSI model and soil water deficit in the root zone in apple trees in all of the orchards.
We found that both soil water deficit and soil water potential were highly correlated with thermal-based water stress index models in apple trees in the mildly-stressed range. But our most important finding was that the relationships were time-sensitive, meaning that they were valid only at a specific time of day. As expected, NP soil moisture data resulted in the best correlations and soil data obtained from commercial soil sensors were not as good, but still acceptable.
Our most important finding was that the relationships were time-sensitive, meaning that they were valid only at a specific time of day.
The measurements taken between 10:00 am and 11:00 am (late morning, time of maximum transpiration) were highly correlated with soil water deficit and soil water potential, but the “coefficient of determination” decreased quickly and significantly beyond this time window (about half in just one hour, and reached zero in the afternoon hours). This is a very important finding because researchers still think midday is the best time to measure CWSI. The apple trees showed an interesting behavior which was nothing like what we are used to seeing in row crops. They regulate their stomata in a way that transpiration rate is intense late in the morning (maximum) and late in the afternoon. During the hot hours of afternoon, they close their stomata to minimize water loss.
Similar Research
In the literature you will find other efforts, where researchers have tried to correlate remotely-sensed satellite-based thermal or NIR measurements to soil water content. However, these relationships are unable to predict the water status of plant root zone and can't be used for irrigation scheduling. The closest studies to ours have been able to find good relationships between CWSI and soil water content in the root zone near the end of the season at high soil water deficits in row crops. Paul Colaizzi, a research agricultural engineer with USDA-ARS at Bushland, TX, did his PhD research in part on the relationship between canopy temperature, CWSI, and soil water status in Maricopa, Arizona; also motivated by Jackson et al. (1981). Steve Evett and his team at Bushland, TX are continuing that research as they try to develop a relationship between CWSI and soil water status that will hold up. They are using a CWSI that is integrated over the daylight hours and have found good relationships between CWSI and soil water content in the root zone near the end of the season when plots irrigated at deficits begin to develop big deficits.
Future of Research
In the future, we hope to study different aspects of this approach, and to find a better way to monitor a large volume of soil in the root zone in real-time (as reference). Currently, neutron probe (NP) is the most precise method for measuring soil water deficit. However, NP is very labor-intensive and does not allow for real-time soil moisture measurements.
We would like to see how universal these equations can be. Right now, we suspect they are crop- and soil-specific, but by how much we do not know. We want to study other fruit trees, and perhaps even row crops, under other irrigation systems and climates. We need to monitor crops for health, as well, to make sure what we are measuring is purely a water stress signal.
We need to monitor crops for health, as well, to make sure what we are measuring is purely a water stress signal.
References
Jackson, R. D., Idso, S. B., Reginato, R. J., Pinter, P. J. Jr., 1981. Canopy temperature as a crop water stress indicator. Water Resour. Res.17, 1133–1138.
Osroosh, Y., 2020. “Internet of Plants” and Plant-based Irrigation Scheduling. https://www.duruntashlab.com/blog
Osroosh, Y., 2020. Do plant-based irrigation scheduling methods work? https://www.duruntashlab.com/blog
Osroosh, Y., Peters, R.T., Campbell, C., Zhang, Q., 2016. Comparison of irrigation automation algorithms for drip-irrigated apple trees. Computers and Electronics in Agriculture, 128: 87–99.
Osroosh, Y., Peters, R.T., Campbell, C., 2016. Daylight crop water stress index for continuous monitoring of water status in apple trees. Irrigation Science, 34(3): 209–219.
Osroosh, Y., Peters, R.T., Campbell, C., Zhang, Q., 2015. Automatic irrigation scheduling of apple trees using theoretical crop water stress index with an innovative dynamic threshold. Computers and Electronics in Agriculture, 118: 193–203.
Osroosh, Y., Peters, R.T., Campbell, C., 2015. Estimating potential transpiration of apple trees using theoretical non-water-stressed baselines. Journal of Irrigation and Drainage Engineering, 141(9): 04015009.
Osroosh, Y., Peters, R.T., Campbell, C., 2015. Estimating actual transpiration of apple trees based on infrared thermometry. Journal of Irrigation and Drainage Engineering, 141(8): 04014084.
Ferrer-Alegre, F., Mohamed, A.Z., Osroosh, Y., Bates, T., Campbell, C., Peters, R.T., 2019. A comparative study of irrigation scheduling based on morning, daylight and daily crop water stress index dynamic threshold (CWSI-DT) in apple trees. IX International Symposium on Irrigation of Horticultural Crops. June 17-20. Matera, Italy.
Mohamed, A.Z., Osroosh, Y., Peters, R.T., Bates, T., Campbell, C., Ferrer-Alegre, F., 2019. Morning crop water stress index as a sensitive indicator of water status in apple trees. ASABE Annual International Meeting. July 7-10. Boston, MA.
Can Canopy Measurements Determine Soil Moisture? (Part 1-2), 2016. environmentalbiophysics.org
Comments