top of page

Automating Crop Water Use Estimation: Combining Imaging and Microclimate Data

Updated: Apr 9

Over the last decade or so, I have encountered numerous publications exploring the use of RGB and multispectral imaging to determine canopy cover percentage and its relationship to crop coefficients. A recurring theme in almost all of these papers was the desire for an automated system to truly leverage this concept!


Fueled by my passion for applied research, coupled with my interest in sensors and automation, I embarked on a journey to automate this very process. The result is the development of a low-cost, all-in-one multimodal imaging system (Model: BINA Pro, DurUntash Lab, San Diego, CA) integrating both thermal and multispectral capabilities, with essential crop models embedded directly within. This system is designed to be versatile, suitable for both indoor and outdoor growing environments.


While the intricacies of the system itself are not the focus of this short article, I want to highlight how one can effectively combine imaging data with microclimate information (radiation, relative humidity, air temperature, wind speed) to accurately estimate the crop water use of plants, whether they are in pots, vegetables, or row crops.


This same concept of automation can be readily adapted to other imaging platforms, such as unmanned aerial vehicles (UAVs), through the integration of embedded systems and custom coding, or even implemented on a smartphone. However, for research purposes and small-scale farming operations, a low-cost handheld multimodal automatic imager offers distinct advantages. It eliminates data latency and empowers users with more immediate insights for improved decision-making.


Evapotranspiration-based Irrigation Scheduling

Currently, the predominant method for assessing plant and soil water status and scheduling irrigation relies on the use of soil moisture sensors. This raises the question: is there a viable alternative to directly measuring substrate moisture with sensors to determine the water requirements of potted plants?


Indeed, a compelling alternative lies in evapotranspiration (ET)-based irrigation scheduling, or basing irrigation decisions on the potential crop water use (ETc). This is typically estimated daily using an evapotranspiration model in conjunction with a crop coefficient (Kc).


Unfortunately, established crop coefficient values (or curves) are lacking for many crops, particularly those commonly grown in pots. The solution is to develop your own unique “crop coefficient curve” during a single growing cycle, which can then be applied for future measurements. Here’s a step-by-step guide on how to determine Kc for your specific crop:


  1. Select a representative number of plant pots.


  2. To determine the actual water consumption (ETa, which represents actual ET) of your plants, weigh the pots daily and calculate the change in water content, then divide this by the surface area of the top of the container. It is crucial that your plants are consistently well-watered throughout this experiment to ensure that ETa and the estimated potential crop ET are equivalent.


  3. Utilize microclimate data (radiation, relative humidity, air temperature, wind speed) and the Penman-Monteith evapotranspiration model (Allen et al., 1998) to calculate the daily potential water use (ETr or ETo, representing potential ET for a reference crop like alfalfa or grass).


  4. Divide the actual evapotranspiration (ETa) by the reference evapotranspiration (ETr) to obtain your daily crop coefficient (Kc). Plot these daily Kc values, and you will generate a crop coefficient curve as a function of the days since planting.


To achieve more accurate crop ET estimations that are responsive to actual crop growth rather than simply the number of days post-planting, consider incorporating the following additional steps:


  1. Capture a top-view image of the plant pots every day and determine the ground cover percentage (C, %). This step requires an RGB or, preferably, a multispectral camera and some basic image processing skills (Figure 1).


  2. Plot the ground cover percentage (C) values against the corresponding crop coefficient (Kc) values. From this data, develop a ground cover function for your specific plants, which will typically take the form of a polynomial equation (e.g., Kc = a x C² + b x C + m).


Moving forward, to determine the daily water needs of your plants, you will only require a top-view image of your plants and the corresponding microclimate data. It's important to note that this same methodology can be effectively applied to estimate the water use of vegetables and row crops as well.


Figure 1. Canopy coverage (%) is converted to crop coefficient (Kc) using a polynomial equation.
Figure 1. Canopy coverage (%) is converted to crop coefficient (Kc) using a polynomial equation.

Automated Estimation of Crop ET

To streamline this entire process, I have integrated crop coefficient functions (Bryla et al., 2010) for several crops, including lettuce, garlic, and tomatoes, into my handheld all-in-one multimodal imaging system (Model: BINA Pro, DurUntash Lab, San Diego, CA).


When I need to measure crop ET, I simply connect a commercially available microclimate unit to the system's SDI-12 port (Figure 2) and capture a top-down image of my plants. The BINA Pro then automatically calculates the crop ET and other relevant parameters for me, geotags and timestamps the results, and saves them in a '.csv' file.


Figure 2. The all-in-one imaging system (BINA Pro, DurUntash Lab, San Diego, CA) automatically estimates daily crop ET (ETc) using a top-view image of the crop to estimate crop cover (C, %), a polynomial equation that converts C to crop coefficient, and the Penman-Monteith ET model. Microclimate data is obtained from an external microclimate unit that is plugged into the BINA Pro SDI-12 port.
Figure 2. The all-in-one imaging system (BINA Pro, DurUntash Lab, San Diego, CA) automatically estimates daily crop ET (ETc) using a top-view image of the crop to estimate crop cover (C, %), a polynomial equation that converts C to crop coefficient, and the Penman-Monteith ET model. Microclimate data is obtained from an external microclimate unit that is plugged into the BINA Pro SDI-12 port.

In Conclusion:

The integration of imaging technology with microclimate data offers a powerful and increasingly automated pathway to refine irrigation scheduling through accurate estimation of crop evapotranspiration. By moving beyond traditional reliance on soil moisture sensors, this approach, particularly when implemented in low-cost, handheld multimodal systems, empowers researchers and small-scale farmers with timely and spatially relevant insights into plant water needs.


The ability to develop crop-specific coefficient curves based on canopy cover provides a more dynamic and growth-responsive method for determining water use compared to solely relying on the number of days after planting. The automation of this process, as demonstrated by the BINA Pro system, significantly reduces the complexities and time investment associated with these measurements, paving the way for more efficient water management practices and ultimately contributing to more sustainable and productive agriculture. This synergy between imaging, environmental sensing, and embedded crop models represents a significant step forward in precision agriculture.


References

Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration: guidelines for computing crop water requirements. Irrigation and Drainage Paper No. 56. FAO, Rome, Italy, 300 pp.


Bryla, D.R., Trout, T.J. and Ayers, J.E., 2010. Weighing Lysimeters for Developing Crop Coefficients and Efficient Irrigation Practices for Vegetable Crops. HortScience, 45: 1597-1604.

留言


bottom of page