Methods for improving plant treatment

Abstract

The present disclosure relates to systems and method for using genetically engineered sensor plants to improve crop treatment efficiency and efficacy and to test crop treatments.

Description

METHODS FOR IMPROVING PLANT TREATMENTCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Application No. 63 / 530,186, filed August 1 , 2023. The content of this application is incorporated herein by reference in its entirety.INTRODUCTION

[0002] Plant stressors include both biotic and abiotic stressors. Biotic stressors may include, but are not limited to, insects, bacteria, viruses, fungi, and other plants, whereas abiotic stressors include, but are not limited to, environmental factors, such as water, temperature, wind, nutrient deficiency, and salinity. Mosa K, et al., Plant Stress Tolerance, (2017) doi: 10.1007 / 978-3-319-59379-1_1 . These stressors may further include chemical stressors that come from the application of various treatments associated with a stressor. While plants may be able to adapt to certain stressors, stressors can limit plant growth and also result in plant death. Appropriately addressing a stressor in a plant is therefore key for ensuring optimum plant health and yield if a plant is being grown as a crop.

[0003] Genetically engineered plants (i.e., plants having been gene edited) have improved plant resistance to pathogens, stressors, and disease, while allow for an increase in overall crop yield and efficiency for farmers. Genetically engineered plants, however, still require application of various treatments (e.g., herbicides, insecticides, fungicides, etc.) in order to adequately protect the plants from stressors. The widespread, indiscriminate application of treatments for various plant diseases and pathogens, and to promote plant health, has adverse environmental impacts and has led to a rise in resistance and / or tolerance to these treatments. Therefore, thereexists a need for more selective screening of plants needing treatment and / or the more selective application of treatments to plants within an agricultural environment. Further, there is a need for testing and analyzing the efficacy of treatments, new and old, for application to treat a specific plant disease, pathogen, or stressor.SUMMARY

[0004] In some embodiments, the present disclosure provides a method for curtailing excessive exposure of an agricultural environment to a chemical plant treatment, comprising: (a) capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; (b) detecting a signal in the spectral images of the sensor plants in response to a stressor; and (c) selectively applying a chemical treatment targeting the stressor to a subgroup of plants in the agricultural environment based on the detected signal.

[0005] In some embodiments, the present disclosure provides a method for improving plant treatment efficiency, comprising: (a) capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; (b) detecting a signal in the spectral images of the sensor plants in response to a stressor; and (c) selectively applying a treatment targeting the stressor to a subgroup of plants in the agricultural environment based on the detected signal.

[0006] In some embodiments, the present disclosure provides a method of screening plants for treatment, comprising: (a) capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor;(b) detecting a signal in the spectral images of the sensor plants in response to astressor; and (c) selecting a subgroup of plants in the agricultural environment for treatment based on the detected signal, wherein the treatment is achieved by a chemical targeting the stressor.

[0007] In some embodiments, the present disclosure provides a method of testing the efficacy of a treatment for plants, comprising: (a) applying a stressor to one or more sensor plants in an agricultural environment, wherein the sensor plants have been genetically engineered to produce a signal in response to a stressor; (b) capturing spectral images of the sensor plants in the agricultural environment; (c) detecting a signal in the spectral images of the sensor plants in response to the stressor; (d) selectively applying a treatment to the plants in the agricultural environment based on the detected signal; and (e) recording the change in signal after the treatment has been applied.

[0008] In some embodiments, the present disclosure provides a method of testing the efficacy of a treatment for plants, comprising: (a) capturing spectral images of sensor plants in the agricultural environment, wherein the sensor plants have been genetically engineered to produce a signal in response to a stressor; (b) detecting a signal in the spectral images of the sensor plants in response to the stressor; (c) selectively applying a treatment to the plants in the agricultural environment based on the detected signal; and (d) recording the change in signal after the treatment has been applied.

[0009] In some embodiments, the sensor plants described herein are dicots. In some embodiments, the sensor plants described herein are monocots.BRIEF DESCRIPTION OF THE FIGURES

[0010] FIG. 1 illustrates a pictorial representation of a sensor plant and a system method of detecting the sensor plant.

[0011] FIG. 2 illustrates a pictorial representation of a sensor plant and a system method of detecting the sensor plant.

[0012] FIGS. 3A-3C depict the use of a drought stressor sensor in tomato plants. FIG. 3A depicts fluorescent signals from well-watered drought sensor plants (left) and from well-watered control plants that always constitutively express a red fluorescent protein (right). FIG. 3B depicts the fluorescent signals from drought sensor plants (left) and from drought control plants that that always express a red fluorescent protein (right) during drought. FIG. 3C shows the change in fluorescent signal from exemplary unmodified tomato plants (M82), drought sensor plants (V298-27-8), and plants always constitutively expressing a fluorescent protein (DsRed) over the course of the drought experiment.

[0013] FIGS. 4A and 4B show the use of a fungal stressor sensor in soybean plants. FIG. 4A shows the onset of visual disease symptoms in soybean plants exposed to Cercospora spores, and FIG. 4B shows the induction of a fluorescent signal in the fungal sensor plants following exposure to Cercospora spores.DETAILED DESCRIPTION

[0014] As used herein, a “sensor plant’’ is referred to herein as a plant configured to signal presence of a particular stressor or set of stressors within and / or at the plant. A sensor plant can be genetically engineered to include a set of promoterreporter pairs (e.g., one promoter-reporter pair, three promoter- re porter pairs) configured to trigger generation of a detectable signal or signals by the sensor plant in the presence of a particular stressor or set of stressors. For example, a sensor plant can be genetically engineered to include a first promoter-reporter pair configured to trigger generation of a red fluorescence signal by the sensor plant in the presence of fungi. Thus, the sensor plant can generate a detectable signal that, when detected,may alert a user (e.g., a farmer, an agronomist, a botanist) associated with the sensor plant of a stressor or stressors present. Further, a sensor plant of a first plant type can be configured to signal presence of stressors in plants of the first type and / or of a different type. For example, a sensor corn plant can be configured to signal presence of stressors in corn plants. In another example, a sensor tomato plant can be configured to signal presence of stressors in potato plants.

[0015] As used herein, a “stressor” is referred to herein as a type of abiotic and / or biotic stress that may negatively affect plant health, such as pest, disease, chemicals, water, heat, and / or nutrient stresses or deficiencies. For example, a plant may experience an insect stressor corresponding to presence of an insect or insect population at the plant that may hinder plant growth and / or health.

[0016] As used herein, a “pressure” is referred to herein as a measurable and / or detectable presence of a particular stressor and / or set of stressors in plants (e.g., a stressor present in a sensor plant, in a cluster of sensor plants, in a crop of sensor plants (e.g., a field planted with sensor plants), in an agricultural environment comprising sensor plants, etc.). For example, the computer system can detect an insect stressor at a cluster of sensor plants and — based on features extracted from images of the cluster of sensor plants — estimate an insect pressure (e.g., measurable presence, distribution, magnitude) at this cluster. Thus, a pressure represents a measurable presence of a particular stressor.

[0017] As used herein, a “pressure gradient” is referred to herein as a distribution of pressures of a stressor (or stressors) across multiple sensor plants and / or sets (or clusters) of sensor plants in an agricultural environment. For example, a user may initially distribute three sets of sensor plants within an agricultural environment. Later, the computer system can access images of the agriculturalenvironment, recorded by an aerial sensor (e.g., a satellite), depicting the three sets of sensor plants. Based on features extracted from regions of the image depicting each set of sensor plants, the computer system can interpret a pressure gradient of a stressor in the agricultural environment. More specifically, the computer system can: interpret a first pressure of the stressor in a first set of sensor plants based on features extracted from a first region of the image depicting the first set of sensor plants; interpret a second pressure of the stressor in a second set of sensor plants based on features extracted from a second region of the image depicting the second set of sensor plants; interpret a third pressure of the stressor in a third set of sensor plants based on features extracted from a third region of the image depicting the third set of sensor plants; and interpret the pressure gradient of the stressor in the agricultural environment based on the first pressure, the second pressure, and the third pressure. Based on this pressure gradient, the computer system can interpret pressures of the stressor at various locations within the agricultural environment (e.g., via interpolation).

[0018] As used herein, a “user” is referred to herein as a person associated with an agricultural environment including sensor plants. The term “agricultural environment” includes, but is not limited to, an agricultural field, a crop of plants, a greenhouse, an arboretum, a laboratory, and / or other areas where a plant may be grown. For example, a user may refer to a farmer associated with a particular agricultural environment. In another example, a user may refer to an agronomist associated with a particular crop of plants. In another example, a user may refer to a scientist studying or developing sensor plants and / or treatments of stressors in sensor plants and non-sensor plants.

[0019] The following description of embodiments of the invention is not intended to limit the invention to these embodiments but rather to enable a person skilled in theart to make and use this invention. Variations, configurations, implementations, example implementations, and examples described herein are optional and are not exclusive to the variations, configurations, implementations, example implementations, and examples they describe. The invention described herein can include any and all permutations of these variations, configurations, implementations, example implementations, and examples.Sensor Plants

[0020] The present disclosure includes genetically engineered sensor plants that produce a detectable signal in response to a stressor. In some embodiments, to generate a sensor plant, plant cells can be genetically engineered to couple a known reporter gene with a certain biological process. Molecular genetic techniques (e.g., gene or genomic editing) can be implemented to associate an expression of the reporter gene with certain abiotic and / or biotic stresses and traits. Therefore, the reporter gene can act as a signal of an abiotic and / or biotic stress or trait in the plant cells. For example, the sensor plant can be modified to fluoresce (i.e., absorb photons at one frequency and emit photons at a different frequency), exhibit a change in pigmentation (i.e., produce pigment proteins that result in a phenotypic change in the plant cell), or produce another detectable signal (e.g., heat) in the presence of (and proportional to) a disease or stressor. In this example, the sensor plant can be modified to fluoresce in the presence of one or more disease or stressors, such as: fungi, bacteria, nematode, parasites, viruses, insects, chemicals, heat, competing plant species, water stress, nutrient stress, phytoplasmal disease, etc. In another example, the sensor plant can be modified to signal presence of a stressor via bioluminescence of the sensor plant. In yet another example, the sensor plant can be modified to signal presence of a stressor via a pigmentation change of the sensor plant. In anotherexample, the plant can be modified to signal presence of a stressor via the release of heat energy.

[0021] Sensor plants, as described herein, may be any plant that is capable of being genetically engineered, including, but not limited to, soybeans (Glycine max), cotton (Gossypium species), corn (Zea mays), canola (Brassica napus). In some embodiments, the sensor plants are dicots. In some embodiments, the sensor plants are monocots. Methods of genetically engineering plant cells are known to those skilled in the art, and any acceptable method of genetic engineering (e.g., gene or genomic editing) that is compatible with a given plant species may be employed to produce sensor plants.

[0022] Plant cells can be genetically engineered (e.g., gene or genomic editing) to include promoter and reporter pairs that indicate presence of certain stressors in a plant or crop of plants. A promoter includes genetic regulatory elements that drive expression of mRNA at a specific time and place that is subsequently translated into a functional protein. Promoter activity is representative of native biological processes that occur when a particular stress is present in the plant. To detect presence of these stressors, a known reporter gene that expresses a certain signal can be coupled to the promoter of choice. Therefore, when the plant's cells express the promoter associated with a certain stressor, the reporter tagged to the promoter is also expressed and thus detectable. Some reporter signals (e.g., fluorescent signals, pigments, etc.) exist naturally in plants without genetic modification. These signals can be enhanced by selective breeding, gene or genomic editing, new breeding techniques, and / or other plant selection techniques. Each of these reporter genes can produce an optical signal that is distinguishable from the plant itself. A combination of reporter genes can be used as well, to indicate various plant stressors present in theplant or crop. In some embodiments, the promoter-reporter pairs and production of the signals in response to the stressor do not interfere with the functioning of other genes in the plant.

[0023] In some embodiments, the sensor plants described herein are plants that are already growing and can be transplanted to an agricultural environment for continued growth and monitoring.

[0024] As described herein, a sensor plant includes a first promoter-reporter pair including: a first promoter that activates in the presence of a first stressor at the sensor plant; and a first reporter coupled to the first promoter and configured to exhibit a first signal that is detectable (e.g., a signal in the electromagnetic spectrum) in response to activation of the first promoter by the first stressor.

[0025] In an embodiment, the sensor plants described herein further include a second promoter-reporter pair including: a second promoter that activates in the presence of a second stressor at the sensor plant; a second reporter coupled to the second promoter and configured to exhibit a second signal that is detectable (e g., a signal in the electromagnetic spectrum) in response to activation of the first promoter by the second stressor, the second signal different from the first signal.

[0026] In some embodiments, the sensor plants described herein further include a third promoter that activates in the presence of a third stressor at the sensor plant, and a third reporter being coupled the third promoter and configured to exhibit a third signal that is detectable (e.g., a signal in the electromagnetic spectrum) in response to activation of the third promoter by the third stressor, the third signal different from the first signal and the second signal.

[0027] In some embodiments, the sensor plants described herein further include a fourth promoter that activates in the presence of a fourth stressor at thesensor plant, and a fourth reporter being coupled the fourth promoter and configured to exhibit a fourth signal that is detectable (e.g., a signal in the electromagnetic spectrum) in response to activation of the fourth promoter by the fourth stressor, the fourth signal different from the first signal, the second signal, and the third signal.

[0028] In some embodiments, the sensor plants described herein further include a fifth promoter that activates in the presence of a fifth stressor at the sensor plant, and a fifth reporter being coupled the fifth promoter and configured to exhibit a fifth signal that is detectable (e.g., a signal in the electromagnetic spectrum) in response to activation of the fifth promoter by the fifth stressor, the fifth signal different from the first signal, the second signal, the third signal, and the fourth signal.

[0029] In some embodiments, the sensor plants described herein further include a sixth promoter that activates in the presence of a sixth stressor at the sensor plant, and a sixth reporter being coupled the sixth promoter and configured to exhibit a sixth signal that is detectable (e.g., a signal in the electromagnetic spectrum) in response to activation of the sixth promoter by the sixth stressor, the sixth signal different from the first signal, the second signal, the third signal, the fourth signal, the fifth signal, and the sixth signal.

[0030] FIG. 1 provides a pictorial representation of a sensor plant as described herein, wherein the sensor plant comprises one or more promoter-reporter pairs. An optical sensor monitors the sensor plant and captures images of the plant and any signals produced by the sensor plant via the promoter-reporter pair. The collected images can be compared to a stored model in the computer system to identify the presence of a particular stressor, and the identification of a stressor can automatically trigger the generation of specific prompts for a user (e.g., a prompt to apply an insecticide in response to the detection of a signal responsive to an insect stressor).

[0031] In some embodiments, the sensor plants described herein include a first promoter-reporter pair including: a first promoter configured to activate in the presence of a first stressor within a first magnitude range at the sensor plant; and a first reporter coupled to the first promoter and configured to exhibit a first signal in the electromagnetic spectrum in response to activation of the first promoter by the first stressor. In this variation, the sensor plant also includes a second promoter-reporter pair including: a second promoter configured to activate in the presence of the first stressor within a second magnitude greater than the first magnitude range at the sensor plant; and a second reporter coupled to the second promoter and configured to exhibit a second signal in the electromagnetic spectrum in response to activation of the second promoter by the second stressor.

[0032] In some embodiments, the sensor plants described herein include: a first promoter that activates at a first time over a first duration in response to a first stressor presence in the sensor plant; a second promoter that activates at a second time for a second duration in response to the first stressor presence in the sensor plant, the second time succeeding the first and preceding the termination of the first duration; and a reporter coupled to the first and second promoter that, in response to activation of the first promoter, exhibits a first signal over the first duration for detection of the first stressor; and, in response to activation of the second promoter, exhibits a second signal over the second duration for detection of the first stressor.

[0033] In an embodiment, multiple promoters can be tagged to one reporter such that the sensor plant outputs a signal for a particular stressor over an extended duration of time. For example, a set of three promoters linked to fungal stress can be tagged with the red fluorescence protein reporter. At an initial time, presence of the first promoter can trigger the expression of the red fluorescence protein in responseto a certain fungal pressure. At a second time, as the signal produced by the first promoter decreases, presence of the second promoter can trigger the continued expression of the red fluorescence. And again, at a third time, a third promoter can trigger the expression of the red fluorescence in the plant. Therefore, genetic engineering techniques can be implemented to string together multiple promoter sequences and tag this string of promoters with a reporter gene for identifying which promoter sequences are expressed in the plant, thus extending the detection window.

[0034] In some embodiments, plant cells can be genetically engineered to include combinatorial reporter-promoter pairs that present different signals responsive to different stressors and / or pressures (e.g., promoter-reporter pairs that signal presence of more than one type of stressor). In some embodiments, the combinatorial reporter-promoter pairs include, but are not limited to, promoter-reporter pairs that detect fungi, bacteria, nematode, parasites, viruses, insects, chemicals, heat, water stress, nutrient stress, phytoplasmal disease, and combinations thereof. The computer system can distinguish between the different signals produced by the combinatorial promoter-reporter pairs and detect when a signal is being produced by the sensor plant. The computer system can additionally leverage a model to interpret these signals, including deriving more information than the sum of the set of reporters, such as: a type of fungus in addition to presence of a fungal pressure; or proportion of water stress to heat stress.

[0035] In some embodiments, the sensor plant can be configured to include a first quantity of promoters and a second quantity of reporters less than the first quantity of promoters. For example, expression of the red fluorescent protein can signal presence of a certain water pressure, and expression of the yellow fluorescent protein can signal presence of a certain heat pressure. However, the expression of both thered fluorescent protein and the yellow fluorescent protein can signal either presence of both a certain water pressure and heat pressure, or presence of a third pressure, such as a certain insect pressure. Therefore, fluorescence of the sensor plant can be combined with knowledge of disease frequency, common disease locations, and common disease times to isolate a particular plant stressor present in the agricultural environment. In some embodiments, a first, second, and third fluorescing compound are each coupled to a first, second, and third biological process, respectively. In some embodiments, a fourth biological process is coupled to the first and second fluorescing compound; a fifth biological process is coupled to the second and third fluorescing compound, a sixth biological process is coupled to the first and third fluorescing compound; and a seventh biological process is coupled to the first, second, and third fluorescing compound. In this embodiment, the detection of all three fluorescing compounds in a plant can signal each of the following: activation of the sixth biological process; activation of the first, second, and third biological process; activation of the first and fifth biological process; activation of the fourth and third biological process; activation of the sixth and second biological process. These biological processes can be distinguished to enable detection of different processes occurring in these plant cells — and therefore different stressors present at the plant. For example, the computer system can prompt the crop manager to treat all possible diseases or a specific disease that can be catastrophic if not treated quickly. In another example, a farmer or agronomist may retrieve a sample from the plant and test for each possible disease to initiate an appropriate course of action. Further, by selectively identify the presence of one or more particular stressors, application of treatments, including chemical treatments, can be selectively applied to limit overexposure to treatments that can have a negative environmental impact (e.g., chemical runoff, release ofunnecessary greenhouse gases, tolerance to treatments, etc.) that can ultimately impact plant health, including crop yield.

[0036] In some embodiments, the promoter and reporter pairs can be implemented by tagging one reporter to one promoter. For example, if a red fluorescent protein is tagged to a promoter sequence indicative of fungal stress in a sensor plant, the promoter sequence, and therefore the red fluorescent protein, can express in the plant cells when the fungal infection in the plant cells rise above a minimum fungal stressor threshold. Similarly, if an anthocyanin pigment protein is tagged to a promoter sequence indicative of an insect stress in a sensor plant, the promoter sequence, and therefore the anthocyanin pigment, can express in the plant cells when the stress from an insect rises above a minimum insect stressor threshold.

[0037] In some embodiments, each sensor plant type for a particular crop is configured to produce a signal responsive to one plant stressor — that is, one sensor plant type includes one promoter-reporter pair configured to produce a signal for one type of stressor. For example, a first sensor plant type for a particular crop (e.g., corn) includes a promoter- re porter pair configured to output a signal responsive to a fungi pressure; and a second sensor plant type for this particular crop includes a different promoter-reporter pair configured to output a signal responsive to an insect pressure.

[0038] In some embodiments, promoter-reporter pairs configured to output signals for multiple distinct stressors are integrated into one sensor plant type for a particular crop. For example, one sensor plant type for a particular crop contains promoter-reporter pairs configured to produce: a luminescent signal responsive to fungi pressure; a pigmentation change responsive to insect pressure; and a red fluorescence signal responsive to phosphorus deficiency. Thus, one plant or cluster of plants of this sensor plant type can be sensed to detect multiple discrete pressures.Detection System

[0039] In the detection system described herein, a computer system (e.g., a local computing device, a remote server, a computer network) identifies a stressor present at a sensor plant based on signals (e.g., fluorescent signal emitted in the electromagnetic spectrum) generated by the sensor plant and captured in spectral images taken by an optical sensor, wherein the sensor plant has been genetically engineered to signal environmental conditions adverse to plant health or growth (i.e., abiotic and / or biotic stressors). In some embodiments, the computer system extracts features (e.g., intensities at particular wavelengths corresponding to specific components, such as proteins) from the spectral images and interpret presence and / or magnitude of a particular stressor exposed to the sensor plant based on these features. In some embodiments the computer system interprets the signals detected based on a stored model linking plant stressors to wavelengths of interest based on known characteristics of promoter and reporter genes in the sensor plant — and before such stressors are visually discernible in the visible spectrum (i.e., with an unaided human eye). In some embodiments, the computer system interprets the presence and / or magnitude of the stressor at the sensor plant and / or other plants nearby based on signals generated by the sensor plant. The computer system then selectively generates and distributes prompts for a user for mitigating the stressor at the sensor plant and / or nearby plants based on the signals that are detected in the spectral images captured by the optical sensor.

[0040] In the detection system described herein, an optical sensor (e.g., a multi- spectral or hyperspectral camera, spectrometer, etc.) captures spectral images of an agricultural environment. In some embodiments, the optical sensor is fixed. In some embodiments, one or more fixed optical sensors are mounted on a pole within anagricultural environment such that the optical sensor can collect images of sensor plants within an agricultural environment, such as hourly or daily, and upload these images (e.g., via a computer network) to the computer system. In some embodiments, the optical sensor is mobile. In some embodiments, one or more mobile optical sensors are mounted on a truck, tractor, or other farm implement that can intermittently capture images of these clusters of sensor plants when driven on an access road along this agricultural environment, such as multiple times in one day per week or while spraying or performing other value add actions. In some embodiments, mobile optical sensors are mounted on an airplane, a helicopter, a drone, a satellite, or other aerial device. In some embodiments, optical sensors used in the detection system described herein comprise fixed optical sensors, mobile optical sensors, and combinations thereof, such that more than one optical sensor may be used to capture spectral images of sensor plants within the agricultural environment.

[0041] In some embodiments, the one or more optical sensors capture spectral images of sensor plants at different frequencies and at different locations within an agricultural environment to achieve greater spatial resolution. In some embodiments, as an optical sensor captures spectral images of sensor plants within an agricultural environment, additional data is recorded and associated with the image. In some embodiments, GPS data associated with an optical sensor and / or imaged sensor plant, date and time, weather, and other conditions is recorded and associated with the corresponding image. Thus, if a specific sensor plant is producing a signal in response to a stressor, the computer system can provide environmental conditions and the exact location (e.g., coordinates) of a particular sensor plant.

[0042] In some embodiments, the computer system extracts magnitudes (e.g., intensities) of wavelengths of the sensor plant reporter signal from the spectral imagescollected by one or more optical sensors. In some embodiments, the computer system implements a stored model to interpret pressure (e.g., presence and / or magnitude) of a particular stressor in an agricultural environment over time based on magnitudes of the extracted wavelengths associated with a particular stressor.

[0043] In some embodiments, spectral images are collected by one or more optical sensors intermittently and inconsistently (e.g., less temporal resolution). In some embodiments, spectral images are collected by one or more optical sensors consistently (e.g., greater temporal resolution). In some embodiments, the images are collected hourly, daily, every other day, semi-weekly, weekly, bi-weekly, and / or monthly. In some embodiments, the computer system leverages data extracted from intermittent and inconsistent images recorded by an optical sensor in combination with consistent data extracted from images recorded by an optical sensor over an agricultural environment, to expand stressor predictions across the crop. Further, the computer system can converge on a more precise model for predicting pressures across the crop over time based on data extracted from these images, such as via incorporations of machine learning algorithms.Detection Method

[0044] In the methods of using the detection system described herein, the computer system can detect and interpret signals generated by sensor plants by extracting features from images of sensor plants in an agricultural environment that correlate to presence of particular stressors at the sensor plants.

[0045] In some embodiments, the computer system accesses digital images (e.g., spectral images) of sensor plant(s), agricultural environment, and / or plant canopy (e.g., sensor plants and surrounding plants) captured by an optical sensor deployed at the sensor plant(s) and / or plant canopy. In some embodiments, the opticalsensor can include: an optomechanical fore optic that enables measurement of fluorescent and non-fluorescent targets; and a digital spectrometer or digital camera that records images through the optomechanical fore optic. The computer system can thus access images recorded by the optical sensor and process these images according to the method described herein to detect reporter signals and interpret stressors present in these plants. More specifically, the computer system can: access images (e.g., spectral) of sensor plants recorded by the optical sensor (e.g., digital spectrometer); extract wavelengths of the compounds of interest from these images; and identify stressors present at the sensor plants based on these wavelengths.

[0046] In some embodiments, the computer system accesses images of sensor plants captured by an optical sensor, such as from a handheld camera, a handheld spectrometer, a mobile phone, a satellite, or from any other device that includes a high-resolution spectrometer, includes band-specific filters, or is otherwise configured to detect wavelengths of electromagnetic radiation fluorescence, luminescence, or changes in wavelengths of visible light produced by the sensor plant in the presence of a particular stressor.

[0047] In some embodiments, the computer system implements different instrumentation depending on the compound of interest, as the wavelengths of different compounds are each best observed under different conditions and may require distinct modes of detection. In some embodiments, the computer system accesses images, captured by a handheld spectrometer, of sensor plants configured to emit a fluorescent signal (e.g., red fluorescence) in the presence of a stressor; and access images, captured by a handheld camera, of sensor plants configured to exhibit a change in pigmentation (production of an anthocyanin pigment protein) in the presence of a stressor.

[0048] In some embodiments, the computer system accesses images of sensor plants collected at particular times of day and / or time intervals so as to maximize detectability of signals generated by sensor plants. For example, for a sensor plant configured to produce a bioluminescent signal in the presence of a particular stressor or stressors, the computer system can access images of the sensor plant collected at night when other signals generated by the sensor plant and its surroundings are minimized.

[0049] The computer system described herein can detect and interpret pressures of stressors in sensor plants via active and / or passive modes of detection. In some embodiments, the computer system implements passive detection to detect a signal generated by sensor plants — without excitation of the sensor plants — in the presence of a stressor or stressors. In some embodiments, the computer system implements active detection to detect a signal generated by sensor plants — in response to excitation of the sensor plant (e.g., via external illumination) — in the presence of a stressor or stressors. In some embodiments, the computer system implements a detection method in which sensor plants are illuminated in an oscillating light for excitation such that the response to that illumination can be isolated.

[0050] In some embodiments, the computer system detects solar-induced fluorescent signals generated by sensor plants via narrow-wavelength measurements near dark spectral features in incident solar radiation. Narrow band techniques associated with Fraunhofer lines (from absorption in the solar atmosphere) and Telluric lines (which originate from absorption of molecules in Earth's atmosphere) enable measurement of the optical signals in daylight, without implementing external illumination. Implementing this measurement technique allows for both specificity and accuracy of measuring small, obscure signals, as well as the ability to collectmeasurements both on the ground and airborne. Therefore, it is possible to collect images of the sensor plants from a large range of distances. The computer system detects these solar-induced fluorescent signals and extracts insights into pressures of stressors at sensor plants generating these signals. In some embodiments, the computer system can: access a first feed of spectral images captured by a first optical spectrometer; interpret a first pressure of a stressor in the first set of sensor plants based on solar-induced fluorescence measurements extracted from a first image in the first feed of images; access a reporter model linking solar-induced fluorescence measurements extracted from spectral images to pressures of stressors for sensor plants; and interpret a first pressure in the first set of sensor plants based on a first solar-induced fluorescence measurement extracted from the first image.

[0051] In some embodiments, the computer system accesses data from a single sensor plant in an agricultural environment. In some embodiments, the computer system accesses images collected by an optical sensor configured to install (e.g., clamp) onto a leaf or stalk of the sensor plant and to capture close-range images of fluorescing surfaces on the sensor plant at a high frequency (e.g., once per minute, once per hour). In some embodiments, the computer system accesses images collected by an optical sensor that is not installed onto the sensor plant. In some embodiments, one or more optical sensors are used to capture images of the sensor plant. In these examples, the computer system can upload images captured by the one or more optical sensors to a remote database via a cellular network, or images can be downloaded to a mobile device or vehicle via a local ad hoc wireless network when a mobile device or vehicle is nearby, and then uploaded from the mobile device or vehicle to the remote database for further analysis as described herein.

[0052] In some embodiments, the computer system accesses images of a set (e.g., cluster) of sensor plants collected by one or more optical sensors. For example, the computer system can access images of sensor plants in an agricultural environment recorded by an optical sensor mounted to a boom or column located in a center of the sensor plants to capture close-range images of fluorescing surfaces on sensor plants in the sensor plants at a high frequency (e.g., once per hour, once per day). The computer system can extract insights from these close-range images of the sensor plants to interpret pressures of a particular stressor(s) the sensor plants. Further, by interpreting pressures in the sensor plants from images recorded by an optical sensor located at the sensor plants, the computer system can extract insights into pressures in a subregion of the agricultural environment including particular sensor plants as well as adjacent subregions.

[0053] In some embodiments, the computer system accesses images of an agricultural environment comprising sensor plants (e.g., a field comprising entirely sensor plants). For example, the computer system can access images of sensor plants in the agricultural environment recorded by one or more optical sensors (e.g., fixed optical sensors, mobile optical sensors, aerial optical sensors, etc.) captured at a high frequency (e.g., once per hour, once per day). In some embodiments, the images are captured at a lower frequency (e.g., once per week, twice per month, once per month). The computer system can extract insights from these images of the sensor plants to interpret pressures of a particular stressor(s) in the sensor plants. Further, by interpreting pressures in the individual sensor plants from images recorded by an optical sensor, the computer system can extract insights into pressures in the agricultural environment as a whole as well.

[0054] In some embodiments, a user manually collects data for sensor plants on a handheld device. For example, the computer system can access images (e.g., spectral images) of sensor plants in an agricultural environment collected by a mobile device (e.g., a smartphone) operated by a user to capture close-range images of the sensor plants at a lower frequency (e.g., once per week, biweekly). In some embodiments, the computer system additionally accesses images of the sensor plants captured by other optical sensors in an agricultural environment. In this implementation, the computer system can upload images to a remote database via a cellular network or automatically upload images via a native or web-based agricultural application executing on the handheld device. The computer system can interpret pressures in the sensor plants directly from features extracted from spectral images to generate a high-resolution, short-interval time series representation of the health of the sensor plants and / or agricultural environment.

[0055] In some embodiments, the computer system implements ground-based mobile imaging to extract insights into the health of sensor plants and agricultural environment by collecting images from optical sensors installed in manned or unmanned vehicles. For example, the computer system can access images (e.g., spectral images) of sensor plants collected by an optical sensor configured to install (e.g., mount) into a bed of a truck operated by a user. In the example, the user may drive the truck along an edge of or through the agricultural environment in order to capture images of the sensor plants as the truck moves through the agricultural environment. The computer system can then upload these images to the remote database, timestamped and georeferenced, and access these images upon upload or at a later time.

[0056] In some embodiments, the computer system accesses images of sensor plants and / or an agricultural environment recorded by an aerial sensor configured to capture images (e.g., spectral images) of the agricultural environment and sensor plants. For example, the computer system can access images of sensor plants and / or an agricultural environment collected by an optical sensor configured to install (e.g., mount) onto a drone or other aerial device. Alternatively, in an agricultural environment comprising non-sensor plants with clusters of sensor plants, a drone or other aerial device may be used to scan regions of the agricultural environment where sensor plant clusters are located to collect images of these sensor plants.

[0057] In some embodiments, the computer system accesses images of a cluster of sensor plants, multiple clusters of sensor plants, and / or a crop of sensor plants in an agricultural environment recorded by an aerial sensor (e.g., long-duration, high-altitude UAVs or a satellites such as OCO-2 or GOSAT) configured to capture long-range images of sensor plants. For example, the computer system can access images collected by a satellite sensor configured to collect long-range images of sensor plants at a low frequency (e.g., once per week, biweekly, once per month). In some embodiments, the computer system can access images collected by a commercial satellite sensor configured to collect long-range images of sensor plants at relatively higher frequencies (e.g. , once per day, multiple times per week).

[0058] The computer system can access images of sensor plants captured at set intervals or particular times of day in order to increase likelihood of detection of signals, to allow for the rapid testing of treatments and / or screening of plants for treatment, and to detect pressures of stressors in sensor plants and crops including sensor plants at early stages before these pressures expand in magnitude or negatively affect crop yield. For example, the computer system can access images ofsensor plants in an agricultural environment recorded by one or more optical sensors to monitor pressures of stressors indicative of plant health and to prompt users (e.g., a farmer) associated with the agricultural environment to selectively mitigate these pressures once detected (e.g., above a threshold pressure). Alternatively, a user manually monitoring an agricultural environment may not visibly see or detect pressures of stressors in the crop until after a pressure has significantly damaged plants in the crop. Thus, the computer system can lower risk or probability of pressures spreading throughout an agricultural environment and across crops into other agricultural environments, and increase overall crop yield. The computer system additionally enables users to selectively and specifically apply a given treatment that is particular for a stressor that a sensor plant is experiencing. The selective application of treatments reduces overexposure or unnecessary exposure to a given treatment, which can slow the development of tolerance and / or the environmental impact that can arise from exposure to the treatment. Further, sensor plants can be configured to output signals of relatively large magnitudes (e.g., greater intensity) responsive to pressures of stressors at relatively low magnitudes. Sensor plants can include promoters configured to activate and / or inactive within a short period of time (e.g., minutes or hours) of an initial infection, deficiency at the sensor plant, or based on the current level of a stressor at the sensor plant. The computer system can then detect a signal generated from activation or inactivation of the promoter in the sensor plant. Based on detection of the signal, the computer system can recommend a minimal treatment to mitigate a pressure in the sensor plant.

[0059] The computer system can regularly monitor sensor plants at set frequencies such that pressures of stressors in sensor plants are detected early while limiting cost and effort by users (e.g., farmers, agronomists) associated with anagricultural environment. For example, the computer system can: access a feed of images of sensor plants in an agricultural environment recorded at a set frequency (e.g., twice per day, daily, weekly); interpret a pressure of a stressor in the sensor plants based on features extracted from a first image, in the first feed of images; and, in response to the pressure exceeding a threshold pressure, generate a prompt to a user associated with the agricultural environment to address the stressor in the sensor plants in the agricultural environment, and / or in neighboring plants. In this example, if the pressure falls below the threshold pressure, the computer system can continue accessing images, in the first feed of images, at the set frequency, to continue monitoring the pressure of the stressor in the sensor plants. In some embodiments, the computer system can generate a prompt alerting the user of the pressure of the stressor. Thus, the computer system enables the user to regularly monitor the health of sensor plants and / or plants in the agricultural environment associated with the user while minimizing physical travel to the agricultural environment, treatment of sensor plants, and / or testing of sensor plant health by the user.

[0060] In some embodiments, the computer system implements both high frequency and lower frequency measurements to more precisely interpret and predict stressors in sensor plants and the agricultural environment. In this implementation, the computer system can combine high-resolution, short-interval time series representation of the health of sensor plants with features extracted from low- frequency, wider field-of-view images sensor plants to predict the health of sensor plants and the agricultural environment. For example, the computer system can access a first feed of images recorded at a first frequency (e.g., twice per day, once per day, biweekly) by a fixed sensor facing \sensor plants in an agricultural environment. Additionally, the computer system can access a second feed of images,of the agricultural environment including the sensor plants, recorded by a mobile sensor (e.g., deployed by a user associated with the agricultural environment) at a second frequency less than the first frequency (e.g., weekly, every two weeks). From images in these feeds, the computer system can derive a model linking features extracted from images in the first feed of images to pressures of stressors at both the sensor plants and the agricultural environment. Thus, the computer system can predict pressures across the region of the agricultural environment at the first frequency based on features extracted from images in the first feed. The computer system can regularly confirm and / or rectify the model based on features extracted from images in the second feed at the second frequency.

[0061] In some embodiments, the computer system can extract features from these images of sensor plants to interpret pressures in sensor plants. For example, the computer system can: access a first feed of images of sensor plants in an agricultural environment; and interpret a first pressure of a stressor in the sensor plants based on a first set of features extracted from a first image, in the first feed of images. More specifically, the computer system can: extract a first feature, in the first set of features, from the first image, the first feature corresponding to a first pixel of the first image; extract a second feature, in the set of features, from the first image, the second feature corresponding to a second pixel of the first image; and estimate a representative feature based on a combination of the first feature and the second feature; access a reporter model linking features extracted from images in the first feed to pressures of the first stressor at the sensor plants; and interpret the first pressure of the first stressor in the sensor plants based on the representative feature and the reporter model. Thus, based on features extracted from images collected by the one or more optical sensors, the computer system can interpret a pressure of astressor at a sensor plant or sensor plants based on a reporter model linking characteristics (e.g., intensity of wavelength) to a particular stressor (e.g., insects, heat, fungi) and / or pressure of the particular stressor.

[0062] The computer system can extract features (e.g., intensities at particular wavelengths) from images of sensor plant(s), a cluster of sensor plant(s), and / or an agricultural environment including sensor plants to interpret pressures of stressors in these sensor plants. In order to extract these features, the computer system can distinguish sensor plants from non-sensor plants, if any are in the agricultural environment, in these images.

[0063] In some embodiments, the computer system identifies locations in an agricultural environment that include sensor plants and extracts features from images or regions of images corresponding to these locations. For example, the computer system can access georeferenced images of sensor plants in an agricultural environment recorded by a ground-based mobile sensor. The computer system can: access a position and orientation of the ground-based mobile sensor when the images were captured; access a set of GPS coordinates corresponding to locations of sensor plants in the agricultural environment; and identify sensor plants in the images based on the position and orientation of the ground-based mobile sensor and the GPS coordinates of sensor plants.

[0064] In some embodiments, the computer system can identify sensor plants in images of sensor plants and non-sensor plants based on a baseline signal generated only by sensor plants. For example, sensor plants can be configured to generate a baseline signal within a first wavelength band at which non-sensor plants do not generate any signal. Further, these sensor plants can be configured to generate a signal within a second wavelength band responsive to pressures of a stressor at thesensor plant, the second wavelength band distinct from the first wavelength band. Thus, the computer system can check subregions of images of clusters of sensor plants or crops including sensor plants for this baseline signal within the first wavelength band, to identify regions of the images including sensor plants and / or clusters of sensor plants.

[0065] In some embodiments, the computer system can identify sensor plants in aerial images of crops (e.g., sensor plants and non-sensor plants) by overlaying images with a mask configured to hide non-sensor plants and highlight sensor plants. For example, the computer system can generate a mask for a particular agricultural environment including five clusters of sensor plants distributed throughout the agricultural environment, the mask defining an opaque layer including five transparent regions corresponding to the five clusters. The computer system can then: overlay the mask over an image of the crop captured by an aerial sensor; apply null pixel values to regions of the crop covered by the opaque layer; and extract features (e.g., intensity measurements) from the five transparent regions corresponding to the five clusters of sensor plants in the crop.

[0066] It is appreciated that the computer system described herein can implement any combination of these methods of data collection (e.g., instrumentation, frequency, range) to collect high-quality data that enable rapid, targeted responses to certain plant stressors and therefore increase yield of sensor plants and / or non-sensor plants nearby in the same agricultural environment. In some embodiments, the computer system accesses high-resolution images recorded by a high-resolution optical sensor (e.g., a RGB camera, a multispectral camera or spectrometer, a thermal or IR camera) mounted to a pole located in an agricultural environment and configured to capture high-resolution images of the sensor plants at a high frequency (e.g., threetimes per day) each day and upload these images to a remote database. The computer system can extract features (e.g., intensity at particular wavelengths) from these high- resolution images to interpret pressures of a stressor at the first cluster of sensor plants. Additionally, the computer system can access low-resolution images recorded by a satellite sensor configured to capture low-resolution images of the entire agricultural environment, at a low frequency (e.g., once per two-week interval). The computer system can extract features (e.g., intensity at particular wavelengths) from these low-resolution images to interpret pressures of the stressor in sensor plants in the agricultural environment. The computer system can derive a model linking pressures of the stressor one or more sensor plants to the pressures of the stressor at other sensor plants and / or non-sensor plants in the agricultural environment based on the daily behavior of the sensor plants and the biweekly behavior of the sensor plants in the agricultural environment; and interpolate behavior of the agricultural environment as a whole.Sensor Plant Distribution

[0067] In some embodiments, sensor plant traits are incorporated into a genetically modified organism (GMO) plant genome as part of a GMO stack already present in GMO seeds, which can then be planted to produce an entire crop of sensor plants. The sensor plants can be configured to generate several distinct signals that represent an array of stresses and can be planted in clusters within the field — as described above — wherein all plants in one cluster comprise the same promoterreporter pair(s) configured to produce a signal for a particular biotic or abiotic stressor (or a particular set of biotic and / or abiotic stressors). In some embodiments, the sensor plants are planted exclusively in an agricultural environment. In some embodiments, sensor plants containing the same promoter- re porter pairs are planted along the fulllength of one crop row in the field with sensor plants in the two adjacent crop rows comprising different promoter-reporter pairs configured to produce signals for different biotic or abiotic stressors; in this example, this pattern of rows comprising sensor plants with different promoter-reporter pairs is repeated along the full length of the field. In another example, sensor plants comprising the same promoter-reporter pairs are planted in rectilinear clusters, such as in adjacent five-meter-long segments of five consecutive crop rows with sensor plants in the adjacent clusters comprising different promoter-reporter pairs configured to produce signals for different biotic or abiotic stressors; in this example, this grid around of clusters of sensor plants comprising the same promoter-reporter pairs is repeated along the full length and width of the field.

[0068] In an embodiment, sensor plants can be planted in exclusively in an agricultural environment. In some embodiments, the sensor plants are planted in clusters — alongside non-sensor plants bearing the same fruit or of similar crop type, and / or alongside sensor plants of a different type (e.g., sensor plants detecting a different stressor) in order to maintain high signal-to-noise ratios and sensing capabilities for a crop.

[0069] By thus clustering sensor plants in one-dimensional or two-dimensional groups of plants configured to produce signals for the same stressors, the crop as a whole can produce high-amplitude signals — characterized by high signal-to-noise ratios — for multiple different biotic and / or abiotic stressors in discrete rows or regions of the field. As described above, stressors indicated by these rows or clusters of plants configured to produce signals for the same stressors can then be interpolated or extrapolated across the entire field to predict pressures across the entire crop.

[0070] Therefore, in the aforementioned embodiments, because each plant in the field exhibits sensing capabilities, the entire crop can be monitored directly, thecomputer system can generate a pressure map of biotic and / or abiotic stressors for the crop as a whole based on signals produced by these plants during one period of time (e.g., on one day) and detected by a fixed or mobile local or remote sensors. By repeating this process to develop new pressure maps for the field over time, the computer system can monitor stressors across the field over time and serve data and / or recommendations for proactive mitigation of these stressors. The computer system can also implement this process to update the pressure map for the field following a stressor treatment at the field, thereby enabling a field operator to directly assess efficacy of this stressor treatment and to make more informed treatment decisions for the field in the future. The user can screen for plants needing treatment so that treatments can be selectively applied as needed, rather than over an entire agricultural environment. Further, once applying a particular treatment to the field based on these interpreted pressures, the computer system can continue to measure and detect signals generated by the sensor plants and therefore assess efficacy of the particular treatment based on new pressures interpreted from these signals.

[0071] In some embodiments, rather than planting the sensor plants as seeds (such as in row crops), seedlings, and / or saplings, the sensor plants can be grafted onto existing plants. Grafting may be useful for perennial crops and other high value crops, such as almond trees or grape vines. A scion or leafy portion of the sensor plant may be grafted into a portion of the desired plant, for example on the middle portion of a tree trunk. For example, a scion of a sensor grape vine can be grafted into the trunk of a mature grape, such that the scion portion of the mature grape vine can implement the sensing technology, providing a representation of the health of the mature grape vine. As grafting sensor plants into existing plants is, initially, a more time-consuming process, the grafting method may be useful for perennial crops, whichdo not require replanting each year. These plants are trimmed at the end of each season but, when the leaves bloom the following season, the sensing capabilities will still be present. Therefore, the grafts only need one application to last the lifetime of the plant.

[0072] In some embodiments, the sensor plants can be planted in clusters or exclusively — rather than mixed with the non-sensor plants — when a field is planted. Thus, in embodiments where an agricultural environment entirely comprises sensor plants, individual plants can be monitored and can also inform a user of the overall health of an agricultural environment. In some embodiments, where sensor plants are planted alongside non-sensor plants or sensor plants of a different type (e.g., detecting a different stressor), rather than mixing a sensor plant for a particular stressor with the non-sensor plants of the same or similar plant type, these sensor plants can be planted in clusters or exclusively in designated sensor plant regions in the field, such as in specific crop rows (e.g., every 50thcrop row) or in target segments of crop rows (e.g., three-row-wide, three-meter-long clusters with a minimum of 20 crop rows or 20 meters between adjacent clusters of sensor plants). Thus, by clustering these sensor plants, adjacent to or surrounded by non-sensor plants or sensor plants of a different type (e.g., detecting a different stressor) in the same field, stress-related signals produced by these sensor plants may exhibit high contrast with adjacent non-sensor plants or sensor plants of a different type and thus yield a high signal-to-noise ratio for presence of the particular stressor in the field. For example, by planting multiple instances of the sensor plant in a small region of the field, a red fluorescing reporter output by these sensor plants may be more easily distinguished against a non- fluorescent background of adjacent non-sensor plants or against a different reporter output produced by a sensor plant of a different type. Similarly, if multiple sensor plantsare planted in one row in the field, this cluster of sensor plants can produce a cumulative signal — indicating presence of an insect pressure as the insect pressure migrates across a crop — characterized by a greater signal-to-noise ratio than a lone sensor plant in this row, and this cluster of sensor plants may also yield greater spatial information regarding direction and scope of the insect pressure moving across the field than a lone sensor plant in this row.

[0073] Clusters of sensor plants can be planted with non-sensor plant crops in the field, wherein clusters of sensor plants contain at least one sensor plant for each stressor or in which each sensor plant includes a promoter for each plant stressor. For example, batches of sensor plants — including at least one promoter for at least one stressor — can be planted in clusters in a field with other non-sensor plants. In another implementation, clusters of sensor plants are grouped by promoter. In this implementation, a first cluster of water pressure sensing plants, a second cluster of fungi pressure sensing plants, and a third cluster of insect pressure sensing plants are planted in discrete groups in the field. In this implementation in which the sensor plants containing the same reporter are planted together in clusters, these clusters may output stronger, higher-amplitude, lower noise signals that are more easily identifiable by a fixed, local-mobile, or remote sensor when a corresponding pressure is present in the field.

[0074] The location of sensor plant clusters can also be selected to enable detection of certain plant stressors with greater accuracy and / or reduced noise. In one example in which a user is physically present to collect stressor data from an agricultural environment — such as via a sensor mounted on a vehicle or via a handheld device — the clusters of sensor plants can be planted near the edges of the crop to enable quick access for the farmer. In this example, because sensor plantclusters are located near the edge of a crop, a user may collect samples from these sensor plants and test these samples directly for plant stressors in order to verify pressures indicated by reporters in these sensor plant clusters. In another example, sensor plants are planted in the center of the crop to increase proximity to each plant in the crop, and therefore potentially increase sensing capabilities or the likelihood of detecting a disease migrating across the crop.

[0075] In yet another example, if a user’s crop shares an edge with another user’s crop, it might be desirable to plant a row of insect pressure sensor plants along the shared edge in order to quickly detect a migrating insect population immediately as they enter the crop. In another example, if there is a lower elevation portion of a crop, a cluster of water pressure sensor plants may be planted in this area, in order to detect when this area is collecting an excess amount of water. A cluster can also be planted at the highest elevation portion of the crop, where plant dehydration might be prevalent. In some embodiments, sensor plants are planted throughout an entire agricultural environment.

[0076] In the implementation described above in which sensor plants are distributed in clusters throughout a field, the sensor plants can be identified and distinguishable from the non-sensor plants in order to improve efficiency of data collection. For example, if a user is using a handheld device to collect images of the clusters on a weekly basis, a marker can be placed in the field such that the cluster is easily located. In another example, where satellite images are used to collect images of crops, the coordinate location of clusters can be obtained in order to collect wavelength measurements of the sensor plants.

[0077] In another implementation, the sensor plants are mixed with the nonsensor plants and also planted together in clusters. The clusters of solely sensor plantscan be evenly distributed throughout a crop or in optimized locations. The clusters of sensor plants can be analyzed more frequently, such as by a drone that scans the clusters of sensor plants each day to collect aerial images. A satellite can collect images of the crop as a whole less frequently, collecting data for both the clusters of sensor plants and the individual sensor plants mixed in with the rest of the crop. The health of the entire crop or agricultural environment can be predicted by the computer system based on the timestamped and georeferenced images of the sensor plants.

[0078] In one implementation, sensor plants can be transplanted as seedlings into a crop. For example, a sensor strawberry plant may be initially transplanted as a seedling to a field of strawberry plants. In another implementation, sensor plants can be sown as seeds into a crop. For example, a sensor soybean plant may be initially sown as a seed into a crop of soybean plants. In yet another implementation, sensor plants can be grafted onto existing perennial crops. For example, a sensor grape scion sensor can be grafted to a grape producing vine.

[0079] In some embodiments, the agricultural environment is a controlled environment, such as a greenhouse (e.g., glass roof or factory farm), a growth chamber, or another enclosed growing structure. Sensor plants grown in controlled environments can be regularly monitored for detection of pressures of stressors at the sensor plants. In one implementation, sensor plants can be grown in an enclosed growing structure via vertical farming.

[0080] Sensor plants grown in these controlled environments can be transplanted to other locations (e.g., commercial agricultural environments) to serve as sensor plants. Alternatively, sensor plants grown in controlled environments can be monitored for detection of pressures of a stressor or stressors under particular controlled environmental conditions (e.g., climate, region, presence of other plants) inthe controlled environment. The computer system can interpret pressures in these sensor plants in the greenhouse environment and extract insights into plants (e.g., in an agricultural environment) under similar environmental conditions based on pressures in the sensor plants.

[0081] The computer system can more frequently monitor sensor plants in a controlled agricultural environment (e.g., an indoor growing facility, a greenhouse, etc.) than sensor plants located in an agricultural environment due to the smaller area of the controlled agricultural environment. Therefore, the computer system can extract further insights into these sensor plants grown in the controlled agricultural environment. For example, by interpreting daily pressures of a particular stressor in sensor plants in a greenhouse, the computer system can more precisely converge on a model linking features extracted from images collected of the sensor plants to pressures of the particular stressor. The computer system can then better model pressures of the particular stressor in a controlled agricultural environment including sensor plants of a same type and / or including these sensor plants once transplanted by a user associated with an agricultural environment.

[0082] Sensor plants, as described herein, grown in a controlled agricultural environment can also function as a screening tool for the evaluation of treatments for plant stressors. In some embodiments, sensor plants are grown in a controlled agricultural environment and are exposed to a stressor, wherein the sensor plants comprise a promoter-reporter pair that is specific for the stressor the sensor plants are exposed to. Once the sensor plants exhibit a detectable signal that is captured in images captured by one or more optical sensors, various treatments may be applied to the sensor plants. Changes in the signal produced by the sensor plants can then be captured in subsequent images collected by the one or more optical sensors. Analysisof the change in signal can provide insights into the efficacy of different treatments for a stressor, and can enhance and accelerate treatment development and testing. Multiple treatments can be tested across a controlled environment comprising many sensor plants. The treatments being applied to the sensor plants can be newly developed, or treatments that have already been created. Further, the treatments may be applied in different forms. In some embodiments, the treatment is applied as a liquid (e.g., applied through a sprayer). In some embodiments, the treatment is applied as a solid. In some embodiments, the sensor plants and detection system as herein described analyze not only the treatment itself, but the efficacy of different application methods and treatment schedules in order to determine the optimal formulation and use of a treatment.Analysis Performed by the Detection System

[0083] The computer system can: access images (e.g., spectral) of the sensor plants; extract features indicative of stressors and pressures corresponding to these stressors in these sensor plants; interpolate or extrapolate pressures of particular stressors in these sensor plants to other plants (e.g., sensor and non-sensor plants) in the same agricultural environment (and in nearby fields); and then generate realtime prompts or treatment decisions for these crops in order to increase efficiency of crop treatments and maintenance over time and maintain or increase yield from the agricultural environment.

[0084] In one implementation, the computer system: extracts wavelength measurements for specific compounds in a region of an image depicting one or more sensor plants; and transforms these wavelength measurements into a pressure map (e.g., presence, magnitude) of a particular stressor or stressors in the one or more sensor plants. For example, if the computer system detects — in this region of theimage — a specific wavelength for a compound associated with a fungal disease, the computer system can access a model linking wavelength of the compound of interest to the fungal stressor and then pass the intensity of this wavelength in this region of the image into the model to estimate the fungal pressure (e.g., in the form of “percent” pressure) in the one or more sensor plants. Based on the fungal pressure for the specific sensor plant, the computer system can generate a prediction of the fungal pressure for other plants (e.g., other sensor plants and / or non-sensor plants) surrounding or nearby the one or more sensor plants.

[0085] FIG. 2 provides a pictorial representation of the analysis performed by the computer system. An optical sensor captures images of a sensor plant and extracts different wavelengths that can be associated with a specific promoter-reporter pair. The computer system can compare the extracted wavelengths to previously established signals and / or standards (e.g., a threshold level of a stressor) that indicates the presence of a particular stressor. The computer system can then determine which stressor(s) are present in a sensor plant, and subsequently produce guidance or instructions for a user to address the identified plant stressor.

[0086] In the foregoing example, to generate the model linking intensity of wavelengths to pressures of stressors, a user may collect samples from a leaf or the soil sensor plant to detect plant stressors. The samples can be tested to identify the specific type and pressure of a stressor present at the leaf, while the wavelength of the compound in the plants associated with the disease can be measured from the images collected. A model depicting the relationship between the detected wavelength of the compounds of interest and the pressure magnitude can then be generated (e.g., by the computer system) based on these empirical data. Subsequently, the computer system can automatically (and autonomously) predict pressures throughout the cropbased on features extracted from images of sensor plants rather than based on physical samples collected by a user. Alternatively, this model can be generated based on lab data prior to deployment of the sensor plants to the agricultural environment and can be linked to sensor plants deployed during the subsequent growing season.

[0087] In a crop with multiple clusters of sensor plants or with sensor plants distributed throughout the crop, the images collected both on the ground and aerially can be accessed by the computer system to output a pressure map for the crop. The pressure map can display the locations of specific disease and stressors, and can be updated or combined to display the spread or elimination of certain pressures over time. The map can display interpolated pressure data for regions of the crop, including where no sensor plants are located. In one implementation, images can be collected multiple times per day from a camera located on a pole in the center of a cluster of sensor plants. Additionally, satellite images of the entire crop, including other sensor plant clusters, can be collected biweekly. The data collected daily from the single cluster can be used to model the behavior of the other clusters, based on the biweekly wavelength measurements of disease compounds in the rest of the clusters. The regions of the crop between clusters, or the “non-sensor” regions, can also be modeled by interpolation (e.g., via machine learning algorithms). To confirm presence of a stressor and to interpret a pressure of this stressor, a user may collect samples of the sensor plant itself or of the surrounding soil.

[0088] For example, the computer system can access a feed of images from a remote database, the first feed of images timestamped and georeferenced, and uploaded to the remote database via a wireless network from a device located on a post in the center of a first cluster of sensor plants in an agricultural environment at a frequency of one image every hour; access satellite images of the agriculturalenvironment, including a set of clusters of sensor plants, the satellite images collected biweekly; interpret a pressure of a stressor in the first cluster based on the model linking features extracted from the feed of images to stressor and pressures of stressors; interpolate the pressure of the set of clusters and of all plants in the agricultural environment, based on the model and the feed of images from the remote database and the satellite images; generate a pressure map including locations of a pressure in an agricultural environment; magnitude of the pressure; locations of sensor plant clusters; a first timestamp indicating the time the map is generated and a second timestamp indicating a time for which the map is representative; generate prompts or treatment recommendations for this agricultural environment based on the pressure map; and, deliver the pressure map and corresponding prompts or treatment recommendations to a user associated with the agricultural environment.

[0089] After generating a pressure map based on the measured wavelengths of specific compounds in the plants, the computer system can prompt a user associated with the agricultural environment to take certain actions in order to combat plant stressors. In one implementation, a farmer may plant a row of insect sensor plant seeds on an edge of a soybean field, for monitoring the border between the farmer's crop and a neighboring crop. Each day, an optical device mounted to a pole in the row of sensor plants can capture images of the sensor plants. From these images, the computer system can measure the wavelengths of compounds associated with the insect related disease, and display a certain insect pressure magnitude on the edge of the map where the row of sensor plants is located. Based on the insect pressure magnitude and the times at which images were collected, the computer system can display a predicted current insect pressure magnitude for the surrounding area in the crop and prompt the farmer to make certain decisions such as: whether to treat thecrop with insecticide for the insects dependent on the pressure magnitude reading; which areas of the crop to treat for insect disease; and an extent of treatment in different regions of the crop. After initial treatment, as more images are collected and more data becomes available, the computer system can update the pressure map and prompt the farmer to implement an updated treatment plan with this new information, and make improved treatment decisions for future insect related diseases. The output pressure map provides a means for the farmer to be alerted to a disease or stress in the crop at the onset, as well as access predictions for what may happen in response to certain treatments or to applying no treatment. Over time, as more data is collected and various treatments are applied to the crop based on stressors indicated by signals output by sensor plants in the field, the computer system can develop models to predict responses of plants and plant stressors to certain treatments, such as a magnitude change in signal output by a sensor plant for a known stressor responsive to a particular magnitude of treatment applied to the field.

[0090] The computer system can generate real-time prompts or treatment decisions for these crops in order to increase efficiency of crop treatments and maintenance over time and maintain or increase yield from the agricultural environment. For example, in response to interpreting a pressure of a particular stressor, in sensor plants, above a threshold pressure, the computer system can generate a prompt to address the particular stressor in plants proximal the sensor plants. More specifically, the computer system can: isolate a first action, in a set of actions defined for sensor plants, linked to the particular stressor; and transmit a notification to perform the first action in the agricultural environment to mitigate the particular stressor to a computing device of a user associated with the agricultural environment. Thus, the computer system can update users (e.g., agronomists,farmers, field owners) regarding plant health and / or suggest treatments for mitigating pressures of stressors in plants.

[0091] In some embodiments, the computer system can derive a pressure model linking pressures of a particular stressor at a first set of sensor plants (e.g., one sensor plant, a cluster of sensor plants, an agricultural environment comprising entirely sensor plants) to pressures of the particular stressor at the second set of sensor plants. By developing this pressure model, the computer system can minimize data collection of all sensor plants in a particular region (e.g., agricultural environment) by relating pressures in sensor plants in a single set of sensor plants to other sets of sensor plants in the agricultural environment.

[0092] In some embodiments, the computer system can: access a first feed of images recorded at a first frequency by a fixed sensor (e.g., a camera mounted to a beam in a center of an agricultural environment) facing a first set of sensor plants in an agricultural environment; access a second image of a second set of sensor plants in the agricultural environment, the second image recorded by a mobile sensor (e.g., camera of a mobile device of a user associated with the agricultural environment) during a first time period; interpret a first pressure of a stressor in the first set of sensor plants during the first time period based on a first set of features extracted from a first image, in the first feed of images, captured during the first time period; and interpret a second pressure of the stressor in the second set of sensor plants during the first time period based on a second set of features extracted from the second image. Based on the first pressure interpreted at the first set of sensor plants and the second pressure interpreted at the second set of sensor plants, the computer system can derive a pressure model associating pressure of the stressor at the first set of sensor plants with pressure of the stressor at the second set of sensor plants.

[0093] Once the computer system derives the pressure model, the computer system can continue accessing images from the first feed to interpret pressures at the first set of sensor plants and at the second set of sensor plants based on the model. For example, during a second time period, the computer system can: interpret a third pressure of the stressor in the first set of sensor plants based on a third set of features extracted from a third image, in the first feed of images, captured during the second time period; and predict a fourth pressure of the stressor in the second set of sensor plants during the second time period based on the third pressure and the model. Therefore, the computer system can predict pressure at the second set of sensor plants based on images of the first set of sensor plants from the first feed, without accessing additional images of the second set of sensor plants. Alternatively, the computer system can continue collecting images of the second set of sensor plants at a second frequency less than the first frequency to ensure precision of the pressure model and to update the pressure model over time. Further, the computer system can collect images of other sets of sensor plants and develop additional pressure models linking pressures in sensor plants of these other sets of sensor plants across the particular region to the first set of sensor plants in the agricultural environment, thus enabling predictions of pressures of the particular stressor in the set of sensor plants across the agricultural environment based on information extracted from images of the first set of sensor plants.

[0094] Based on this predicted fourth pressure at the second set of sensor plants, the computer system can generate a prompt or transmit a notification to a user associated with the agricultural environment. For example, in response to the fourth pressure in the second set of sensor plants exceeding a threshold pressure, thecomputer system can generate a prompt to address the stressor in plants proximal the second set of sensor plants in the agricultural environment.

[0095] In some embodiments, the computer system can derive a gradient model associating pressures of a particular stressor at a first set of sensor plants (e.g., one sensor plant, a cluster of sensor plants) to pressures at subregions of an agricultural environment including the first set of sensor plants (e.g., a pressure gradient in the agricultural environment). By developing this gradient model, the computer system can minimize data collection of all sensor plants in a particular region (e.g. , agricultural environment) by relating pressure gradients in the particular region (e.g., pressures in sensor plants across the particular region) to a single set of sensor plants in the agricultural environment. Further, the computer system can correct for deviations in pressures interpreted at the first set of sensor plants based on the gradient model.

[0096] In some embodiments, the computer system can: access a first feed of images recorded at a first frequency by a fixed sensor (e.g., a camera mounted to a pole in an agricultural environment) facing a first set of sensor plants in an agricultural environment; access a second image of a region of the agricultural environment comprising the first set of sensor plants, the second image recorded by a mobile sensor (e.g., an aerial sensor, a drone, a satellite) during a first time period; interpret a first pressure of a stressor in the first set of sensor plants during the first time period based on a first set of features extracted from a first image, in the first feed of images, captured during the first time period; interpret a first pressure gradient of the stressor in sensor plants in the region of the agricultural environment during the first time period based on a second set of features extracted from the second image; and derive a gradient model associating pressure of the stressor at the first set of sensor plants andpressure gradient of the stressor in the region of the agricultural environment based on the first pressure of the stressor and the first pressure gradient.

[0097] Upon deriving the gradient model, the computer system can rectify the first pressure gradient based on the first pressure of the stressor at the first set of sensor plants and the gradient model. Further, the computer system can predict pressure gradients of the particular stressor based on features extracted from images in the first feed. For example, the computer system can: interpret a second pressure of the stressor in the first set of sensor plants during a second time period based on a third set of features extracted from a third image, in the first feed of images, captured during the second time period; and predict a second pressure gradient of the stressor in the region of the agricultural environment during the second time period based on the second pressure and the model.

[0098] From this pressure gradient, the computer system can monitor pressures at various subregions of the agricultural environment. If the computer system predicts a high pressure of the particular stressor at a particular subregion of the agricultural environment, the computer system can flag this subregion and generate a prompt to a user associated with the agricultural environment to address the particular stressor in this subregion. For example, the computer system can, in response to the second pressure gradient predicting a third pressure in a subregion of the agricultural environment and exceeding a threshold pressure, generate a prompt to address the stressor in plants occupying the agricultural environment proximal the subregion of the agricultural environment. Further, based on the pressure gradient, the computer system can generate a pressure map. The computer system can include this pressure map in the prompt for the user.

[0099] Further, the computer system can refine the gradient model by interpreting pressures from additional sets of sensor plants in the agricultural environment. In one implementation, the entire agricultural environment is sensor plants (e.g., having no non-sensor plants). In this implementation, the computer system interprets the first pressure gradient based on features extracted from the second image recorded by a mobile sensor. The computer system can combine this low-resolution pressure gradient data for the entire agricultural environment of sensor plants with the high-resolution pressure data for the first set of sensor plants to develop a more precise gradient model for predicting pressure gradients of the entire agricultural environment.

[0100] In another implementation, in which clusters of sensor plants are planted within an agricultural environment of non-sensor plants, the computer system can interpret the first pressure gradient based on features extracted from regions of the second image, recorded by the mobile sensor, regions including the first set of sensor plants and (at minimum) a second set of sensor plants. In this implementation, the computer system can interpret a pressure of the particular stressor at the first set of sensor plants based on the first image and interpret a second pressure of the particular stressor at the first set of sensor plants based on the second image. The computer system can then: derive a gradient model associating pressure of the particular stressor at the first set of sensor plants with pressure gradient of the first stressor in the agricultural environment based on the second pressure and the first pressure gradient, both extracted from the second image; and rectify the first pressure gradient of the particular stressor in the agricultural environment based on the first pressure and the model.

[0101] The computer system can leverage data corresponding to a particular agricultural environment or crop to develop an annual model for modeling pressures of stressors in the particular agricultural environment. For example, during a first season and for a particular crop, the computer system can extract insights into: water movement across the particular crop; sun exposure across the crop (e.g., daily, weekly, monthly, seasonally); and timing of pressures of other stressors such as insects, fungi, and nutrient deficiencies. The computer system can input each of these insights into an annual model for predicting conditions of the crop at the beginning of next season and throughout the next season. Then, at the start of the next season, the computer system can predict initial conditions of the crop based on the model. Further, the computer system can suggest farming practices to a user associated with the crop based on these predicted initial conditions, such as types of seed hybrid to plant and / or different blends of soil to lay. As the season continues, the system can update the annual model accordingly.

[0102] Further, based on the annual model, the computer system can predict and / or suggest agricultural products and / or treatments best suited for this agricultural environment. For example, the computer system can predict a first pressure of a stressor in plants in the agricultural environment at a particular time based on the annual model. Based on the predicted first pressure, the user may apply a new treatment to these plants at the beginning of a season in order to mitigate the predicted first pressure. Later, the computer system can interpret a second pressure in plants in the agricultural environment at the particular time based on data recorded by a sensor in the agricultural environment. If the second pressure is less than the predicted first pressure, the computer system can update the annual model accordingly and / or recommend the new treatment in the future to treat pressures of the stressor.

[0103] In some embodiments, the computer system can extract insights from a single sensor plant (e.g., in a crop of non-sensor plants, in a greenhouse, in crop of sensor plants) to: monitor pressures of stressors in plants in an agricultural environment; develop models for predicting plant behavior over time; develop models for predicting plant response to various stressors present at the sensor plant; develop models for interpreting pressures of stressors at the sensor plant from measurements; testing efficacy of treatments for various stressors present at the single sensor plant; and / or develop models for plant response to these treatments.

[0104] In some embodiments, a single sensor plant or a single cluster of sensor plants can be grown in a crop of non-sensor plants. In some embodiments, a single sensor plant is monitored in an agricultural environment comprising sensor plants. This single sensor plant (or single cluster of sensor plants) can be monitored for presence of stressors at the sensor plant. For example, the computer system can access data (e.g., images) recorded by a sensor (e.g., a smartphone) and interpret a first pressure of a particular stressor at the sensor plant based on features extracted from this data. Based on the interpreted first pressure at the single sensor plant, the computer system can extract insights into plants proximal the single sensor plant and / or within the crop of non-sensor plants or other sensor plants. Further, the computer system can suggest a particular treatment for plants in the crop based on the interpreted first pressure. Upon application of the particular treatment by a user, the computer system can interpret a second pressure to confirm efficacy of the particular treatment.

[0105] In another example, a sensor plant may be grown in a greenhouse. The computer system can access data (e.g., hyperspectral images) recorded by an optical sensor in the greenhouse to extract a first set of measurements (e.g., intensities of wavelengths) indicative of plant health. A user (e.g., associated with the greenhouse)may collect a sample from the sensor plant to confirm health of the sensor plant and / or presence of any stressors at the sensor plant. In this example, if the user interprets the sensor plant as healthy and interprets no pressures of a particular stressor present at the sensor plant based on the collected sample, the computer system can link the first set of measurements to a healthy plant exhibiting no pressures of the particular stressor and store this information into a model. Later, the user may subject the sensor plant to a pressure of the particular stressor (e.g., a fungal stressor). The computer system can again access data recorded by the optical sensor in the greenhouse to extract a second set of measurements (e.g., intensities of wavelengths) corresponding to the sensor plant. The computer system can then link the second set of measurements of the sensor plant to the pressure of the particular stressor introduced by the user at the sensor plant and store this information into the model. Thus, over time, the computer system can develop the model linking measurements extracted from data recorded by the optical sensor in the greenhouse to pressures of the particular stressor at the sensor plant.

[0106] In some embodiments, the computer system can extract insights related to plant treatment efficacy over time. For example, a sensor plant can be grown in a greenhouse of plants arranged in vertical stacks (e.g., via vertical farming). The computer system can extract measurements from data (e.g., images) recorded by a sensor in the greenhouse to extract insights into plant health. The computer system can interpret a first pressure of a particular stressor at the sensor plant based on a first set of measurements extracted from data recorded by the sensor at a first time. The computer system can then notify a user associated with the greenhouse of the first pressure. The user may then apply a particular treatment to plants proximal the sensor plant in the greenhouse to mitigate the first pressure. Later, the computer system caninterpret a second pressure of the particular stressor at the sensor plant based on a second set of measurements extracted from data recorded by the sensor at a second time (e.g., 24 hours after application of the particular treatment). Based on the first and second pressure, the computer system can derive a model representing pressures of the particular stressor over time in response to application of the particular treatment. The computer system can therefore derive models for predicting plant responses to various treatments and / or agricultural techniques. In some embodiments, a user may proactively apply a particular stressor to a sensor plant in order to assess the efficacy of a treatment for that particular stressor.

[0107] It is appreciated that the methods of analysis and detection of sensor plant signals described herein are applicable in different agricultural environments. For example, in some embodiments, an agricultural environment exclusively comprises sensor plants. In some embodiments, an agricultural environment comprises one or more different sensor plants (i.e., detecting different stressors). In some embodiments, an agricultural environment comprises clusters of sensor plants. In some embodiments, an agricultural environment comprises sensor plants interspersed with non-sensor plants.

[0108] The computer systems and methods described herein can be embodied and / or implemented at least in part as a machine configured to receive a computer- readable medium storing computer-readable instructions. The instructions can be executed by computer-executable components integrated with the application, applet, host, server, network, website, communication service, communication interface, hardware / firmware / software elements of a user computer or mobile device, wristband, smartphone, or any suitable combination thereof. Other computer systems and methods of the embodiment can be embodied and / or implemented at least in part asa machine configured to receive a co puter-readable medium storing computer- readable instructions. The instructions can be executed by computer-executable components integrated by computer-executable components integrated with apparatuses and networks of the type described above. The computer-readable medium can be stored on any suitable computer readable media such as RAMs, ROMs, flash memory, EEPROMs, optical devices (CD or DVD), hard drives, floppy drives, or any suitable device. The computer-executable component can be a processor, but any suitable dedicated hardware device can (alternatively or additionally) execute the instructions.Use of Sensor Plants

[0109] A network or crop consisting entirely of sensor plants can be deployed to an agricultural environment to communicate (e.g., visually, thermally, chemically) to a user biotic and abiotic stressors in the sensor plants and in an agricultural environment. In particular, a sensor plant can experience, react, and deteriorate in presence of certain plant stressors in the same or similar measures as comparable non-sensor plants planted in the crop when exposed to these plant stressors and stressors. Sensor plants can also be deployed in an agricultural environment such that the use of sensor plants in the agricultural environment allows for the monitoring of individual plants to better monitor the health of all plants within an agricultural environment. Therefore, the sensor plant may function as an accurate sensor and predictor of disease and / or stressors in an agricultural environment. In some embodiments, an agricultural environment comprises only sensor plants — such that the monitoring of each individual sensor plant in an agricultural environment allows for the rapid and effective detection and treatment of stressors in an agricultural environment. In some embodiments, sensor plants are deployed to an agriculturalenvironment and planted with other non-sensor plants — such as in clusters of sensor plants surrounded by non-sensor plants or sensor plants of a different type (e.g., detecting a different stressor) — in order to detect, measure, and communicate certain stressors in the sensor plants, which may then be interpolated or extrapolated to stressors in nearby plants.

[0110] In some embodiments, an agricultural environment comprises only sensor plants. In some embodiments, the sensor plants are monitored to identify the presence of stressors in individual sensor plants in an agricultural environment. The presence of a stressor in a sensor plant can provide valuable information on the health of individual plants, a region of an agricultural environment, and an agricultural environment as a whole. In some embodiments, sensor plants are used to screen for the presence of a particular stressor, so that a treatment can be selectively applied only where necessary. The screening of plants for selective treatment reduces the unnecessary use of treatments, limits untargeted exposure to treatments, can reduce the environmental impact associated with treatment, reduces the chance for the development of tolerance resulting from overexposure to a treatment, and can improve the overall efficiency of treatment and the health of sensor plants within an agricultural environment. Thus, the present disclosure includes methods of curtailing excessive exposure of treatments (e.g., chemical treatments) in an agricultural environment.

[0111] In some embodiments, the agricultural environment comprises sensor plants and other plants (e.g., non-sensor plants and sensor plants detecting a different stressor). In some embodiments, sensor plants are monitored for the presence of stressors (e.g., pests, diseases, dehydration) in the sensor plants in the agricultural environment. In some embodiments, sensor plants are monitored for the presence of stressors (e.g., pests, diseases, dehydration) in the sensor plant and / or other plants.Generally, a small quantity of sensor plants can be monitored to extract insights into a larger population of plants (e.g., in crops). For example, a cluster of sensor plants can be planted along an outside edge of a crop of plants and monitored for the presence of pests to inform a user (e.g., a farmer, an agronomist, a botanist) associated with the crop if and / or when a population of pests has entered the crop along this outside edge. In some embodiments, individual sensor plants are monitored in an agricultural environment exclusively comprising sensor plants. In another example, a sensor plant of a first plant type (e.g., tomatoes) can be grown in a greenhouse setting (e.g., glass roof orfactory farm), located in a particular region, and monitored for the presence of stressors (e.g., dehydration, disease, pest) indicative of plant health. A user associated with the greenhouse setting may extract insights from stressors present at the sensor plant to inform planting and / or treatment of other plants (e.g., in a crop) of the same plant type when grown in the particular region. The user may also screen different treatments for their efficacy in addressing a particular stressor before a treatment is selected for broader application.

[0112] In some embodiments, the sensor plants described herein are used to screen various treatments fortheir efficacy against a particular stressor. Sensor plants in an agricultural environment can be exposed to a particular stressor, which results in the production of a detectable signal in the sensor plant. One or more treatments can be applied to the sensor plant that have been exposed to the stressor. The sensor plant can then be monitored to determine the change in signal as a result of the treatment, in some embodiments, a treatment that is efficacious at treating a particular stressor, may, for example, change the intensity of a signal produced by the sensor plant in response to the stressor. The computer system can detect and analyze the changes in intensity of the signal being produced by the sensor plant in order to extractinsights into the efficacy of a treatment. Thus, novel treatments for a plant stressor can be evaluated in the sensor plants in a rapid and objective manner. The system may also evaluate different forms of treatment application, and be used to compare the efficacy of different treatments that are already known, but where efficacy may be impacted by other factors that can be simulated by the sensor plants, agricultural environment, and / or detection system (e.g., co-application with other treatments, soil composition, environmental conditions, etc.).

[0113] In some embodiments, the present disclosure provides a method for curtailing excessive exposure of an agricultural environment to a plant treatment, comprising: (a) capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; (b) detecting a signal in the spectral images of the sensor plants in response to a stressor; and (c) selectively applying a treatment targeting the stressor to a subgroup of plants in the agricultural environment based on the detected signal. In some embodiments, the treatment is a chemical treatment. Plant treatments are generally known to those skilled in the art, and can include, but are not limited to a fungicide, a biofungicide, an insecticide, a herbicide, and a fertilizer, as described herein.

[0114] \r\ some embodiments, the present disclosure provides a method for improving plant treatment efficiency, comprising: (a) capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; (b) detecting a signal in the spectral images of the sensor plants in response to a stressor; and (c) selectively applying a chemical treatment targeting the stressor to a subgroup of plants in the agricultural environment based on the detected signal. Insome embodiments, the treatment is a chemical treatment. Plant treatments are generally known to those skilled in the art, and can include, but are not limited to a fungicide, a biofungicide, an insecticide, a herbicide, and a fertilizer, as described herein. In some embodiments, the signal produced by the one or more sensor plants identifies the presence of a stressor faster than the visible signs of disease that a plant may normally exhibit (e.g., discoloration, wilting, etc.). Therefore, the sensor plant and systems described herein allow for early identification of stressors, which could allow for early mitigation of a stressor prior to visible symptoms start to develop in a plant. As such, the sensor plants described herein can improve plant health, yield, and provide for numerous benefits and efficiencies associated with early intervention to protected against plant stressors

[0115] In some embodiments, the present disclosure provides a method for screening plants for treatment, comprising: (a) capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; (b) detecting a signal in the spectral images of the sensor plants in response to a stressor; and (c) selectively applying a chemical treatment targeting the stressor to a subgroup of plants in the agricultural environment based on the detected signal. In some embodiments, the treatment is a chemical treatment. Plant treatments are generally known to those skilled in the art, and can include, but are not limited to a fungicide, a biofungicide, an insecticide, a herbicide, and a fertilizer, as described herein.

[0116] In some embodiments, the present disclosure provides a method for testing the efficacy of a treatment in plants, comprising: (a) applying a stressor to one or more sensor plants in an agricultural environment, wherein the sensor plants havebeen genetically engineered to produce a signal in response to a stressor; (b) capturing spectral images of the sensor plants in the agricultural environment; (c) detecting a signal in the spectral images of the sensor plants in response to the stressor; (d) selectively applying a treatment to the plants in the agricultural environment based on the detected signal; and (e) recording the change in signal after the treatment has been applied. In some embodiments, the treatment is a chemical treatment. Plant treatments are generally known to those skilled in the art, and can include, but are not limited to a fungicide, a biofungicide, an insecticide, a herbicide, and a fertilizer, as described herein.Treatments for Plants

[0117] In some embodiments, the present disclosure provides sensor plants as described herein to screen plants for treatment in response to a stressor. In some embodiments, the present disclosure provides sensor plants as described herein to screen an agricultural environment for treatment in response to the presence of a stressor. In some embodiments, the present disclosure provides sensor plants as described herein to screen treatments for efficacy against a particular stressor. In some embodiments, the computer system as herein described, upon the detection of a stressor in a sensor plant, provides a prompt to a user to apply a minimal treatment to mitigate a pressure in the sensor plant. In some embodiments, the treatment is a chemical treatment. In some embodiments, the treatment is an organic treatment or an inorganic treatment. In some embodiments, the treatment is selected from a fungicide, an insecticide, a biofungicide, a herbicide, and / or a fertilizer.

[0118] In some embodiments, the treatment being applied to sensor plants and / or an agricultural environment as described herein is a fungicide. Fungicidesinclude, but are not limited to, triazoles, strobilurins (quinone outside inhibitors), and succinate dehydrogenase inhibitors (SDHIs).

[0119] Triazoles are fungicides used to prevent and treat fungal disease in agricultural environments. Triazole fungicides are stable and possess long chemical and photochemical half-lives. Because of their increased stability and persistence, triazole fungicides readily accumulate and spread into environmental soil and water. Because triazoles are persistent and difficult to degrade, they are considered organic pollutants. Triazole fungicides that are useful in the present disclosure include, but are not limited to, tebuconazole and cyproconazole.

[0120] Strobilurin have been used around the world to combat fungal disease for decades Strobilurin fungicides are types of quinone outside inhibitors, which inhibit mitochondrial respiration in fungi by binding to the quinol oxidation site of cytochrome b. Because strobilurin inhibit mitochondrial respiration, they exhibit a broad spectrum of activity and are non-specific, allowing for widespread use and application. Similar to triazoles, strobilurin fungicides have prolonged persistence after application, and often contaminate environmental water, soil, and ecosystems. Strobilurin fungicides that are useful in the present disclosure include, but are not limited to, azoxystrobin, trifloxystrobin, pyraclostrobin, and picoxystrobin.

[0121] SDHI fungicides are also widely used to control turfgrass diseases, and also serve as effective alternatives to fungicides having other mechanisms of action due to their non-specificity. SDHI fungicides function by inhibiting mitochondrial respiration in fungi through the inhibition of succinate dehydrogenase complex. SDHI fungicides block the transport of electrons that is mediated by the succinate dehydrogenase enzyme, therefore arresting fungal growth. SDHI fungicides that are useful in the present disclosure include, but are not limited to, bixafen and flutoanil.

[0122] It is appreciated that those skilled in the art will know which fungicides that are effective in treating various fungal infections, and are appropriate for application in a given plant species. While fungal treatments are known and commonly used, overexposure to fungicides and the persistence of fungicides in the environment leads to increased pollution, contamination, and risk for the development of tolerance. The present disclosure provides for sensor plant systems that address these, and other, risks associated with the use of fungicides in an agricultural environment by allowing for the testing of novel fungicide treatments as well as the screening of sensor plants in an agricultural environment for the selective application of fungicides to sensor plants in need of treatment.

[0123] Rather than relying on chemicals to treat fungi and / or other pathogens, biofungicides comprise living organisms as the active ingredient, wherein the living organisms have specific activity against plant pathogens (e g., a fungi). Different biofungicides will have different mechanisms of action, including, but not limited to competition (e.g., living organism in biofungicide outcompetes the plant pathogen), antibiosis (i.e. , an antibiotic or toxin that targets the pathogen), predation or parasitism, and induction of resistance in the plant. In some embodiments, a biofungicide comprises multiple mechanisms of action. Because biofungicides do not comprise chemicals that may persist in an agricultural environment, biofungicides pose a decreased environmental risk. Because biofungicides are comprise naturally occurring active ingredients, they have limited efficacy across different species of pathogens, and therefore the sensor plant system as described herein enables the quick and effective testing of biofungicides prior to an application to ensure efficacy of a treatment. Biofungicides contemplated by and useful in the present disclosure include,but are not limited to, Trichoderma-based biofungicides, Bacillus subtilis-based biofungicides, and Bacillus amyloliquefaciens-based biofungicides.

[0124] Plants in agricultural environments may also be treated with insecticides in response to insect stressors. Insecticides pose significant environmental risks through toxicity and their potential impact on an environmental system as a whole. Insecticides can be systemic in nature, where they are incorporated and distributed throughout a plant. Insecticides may also be contact insecticides, so that physical contact with an insecticide is toxic to the insect. Synthetic insecticides contemplated by and useful in the sensor plants and sensor plant systems described herein include, but are not limited to, organochlorides, organophosphates, carbamates, pyrethroids, neonicotinoids, phenylpyrazoles, butenolides, and ryanoids / diamides. Insecticides may also comprise insect growth regulators (e.g., hormone mimics and benzoylphenyl ureas) and biological pesticides, which include, but are not limited to, Bacillus thuringiensis-based and other bacterial-based insecticides, interfering oligonucleotides (e.g., RNAi-based insecticides), venom-based insecticides, and enzyme-based insecticides.

[0125] Herbicides are common treatments that are applied to plants in an agricultural environment to control the growth of unwanted plants. Herbicides can be broad-spectrum and more selective, and can be applied prior to planting, prior to emergence, and after emergence of the undesired plant. Herbicides, like the other treatments described herein, can be applied to the soil of an agricultural environment or to the plants in an agricultural environment. Herbicides contemplated by and useful in the sensor plants and sensor plant systems as described herein include, but are not limited to, herbicides that inhibit acetyl coenzyme A carboxylase (ACCase), herbicides that inhibit acetolactase synthase (ALS), herbicides that inhibit enolpyruvylshikimate3-phosphate synthase enzyme (EPSPS), auxin-like herbicides, herbicides that inhibit photosystem II, herbicides that inhibit photosystem I, and herbicides that inhibit 4- hydroxyphenylpyruvate dioxygenase.

[0126] Fertilizers are also regularly applied to agricultural environments in order to promote plant growth and health. The sensor plants and sensor plant systems described herein can be used to test and screen for fertilizers, as well as to signal the need for selective application of fertilizers to sensor plants and / or an agricultural environment. Because fertilizers add specific nutrients to an agricultural environment, excessive application of fertilizers to an agricultural environment can negatively impact growth of plants and water and soil quality, among other consequences. Fertilizers can be organic or inorganic, and can also be applied as a liquid and a solid. Fertilizers contemplated by and useful in the sensor plants and sensor plant systems described herein include, but are not limited to, single nutrient fertilizers (e.g., ammonium nitrate, urea, etc.), multinutrient fertilizers (e.g., two-component fertilizers, NPK fertilizers), and micronutrients (e.g., boron, zinc, manganese, etc.).

[0127] In some embodiments, the sensor plants described herein screen newly developed treatments and / or treatments that are being contemplated for application to plants in an agricultural environment in response to the presence of a stressor. In some embodiments, sensor plants of a particular plant species of interest having been genetically engineered as described herein to produce a signal in response to a stressor are exposed to the stressor. In some embodiments, the sensor plants are able to screen treatments in a controlled agricultural environment (e.g., a greenhouse or indoor farm). Exposure of sensor plants to a stressor induces the signal (e.g., phenotypic change) associated with the stressor, and one or more treatments may be applied to the sensor plants to treat the stressor. As a particular treatment is effectiveat treating a stressor and the pressure associated with the stressor falls under a predetermined threshold level, in some embodiments, the signal is changed in response to the stressor. In some embodiments, one or more optical sensors monitoring sensor plants in the agricultural environment captures images of the sensor plants at predetermined intervals, and the computer system is able to analyze the images to determine the efficacy of a treatment as a result of the change in the detectable signal produced by the sensor plants.

[0128] In some embodiments, the sensor plants described herein further provide for effective screening of treatment application methods in order to assess efficacy of different treatment delivery methods.EXAMPLES

[0129] Example 1 : Drought Sensing in Tomato PlantsThe ability of sensor plants to respond to a water stressor (i.e., drought) was assessed using tomato plants (variety M82) that were genetically engineered with a drought-sensing promoterreporter pair. M82 tomato plants were genetically engineered using Agro bacterium- mediated transformation to insert a promoter-reporter pair comprising the Arabidopsis RD29a promoter operably linked to a gene encoding a TdTomato fluorescent protein (V298-27-8). Control plants for this experiment included unmodified M82 tomato plants, as well as M82 tomato plants that were genetically engineered with a Ubiquitin promoter operably linked to a DsRed fluorescent protein, which would constitutively express the DsRed fluorescent protein.

[0130] Tomato plants (n=5 plants per group) were grown in a growth chamber under a long day condition (16 / 8 h I ight / d ark cycles) at a constant temperature of 24°C for three weeks. Prior to initiation of the drought experiments, all plants were weighed and brought to the same water content level. Water was then withheld from all plantsto turn on the drought stressor. Once visible wilting of the plants was observed (96 hours after withholding of water was started), the plants were re-watered to turn off the drought stressor. Throughout the experiment, fluorescence measurements were collected every day using a handheld spectrometer on three leaves per plant.

[0131] FIG. 3A shows the fluorescent signal produced by plants genetically engineered with the drought stress sensor (V298-27-8) and with the constitutive fluorescent signal (DsRed) prior to the withholding of water. As FIG. 3A shows, before the drought stressor was turned on, whereas the constitutive DsRed signal showed a strong fluorescent signal in all leaves, the fluorescent signal was noticeably absent in the leaves of the drought sensor plant (V298-27-8).

[0132] FIG. 3B shows that once the drought stressor was turned on, leaves in the drought sensor plant (V298-27-8) produced a strong fluorescent signal, indicating the presence of the drought stressor in the plant. By comparison, the constitutive DsRed signal plant maintained its strong fluorescent signal in all leaves before and after the drought stressor was applied.

[0133] A time course of the mean fluorescent signal measured for the three different groups of plants (i.e., V298-27-8, DsRed, and unmodified M82) is depicted in FIG. 3C. As the graph in FIG. 3C shows, shortly after water began being withheld, drought sensor plants (V298-27-8), in response to the drought stressor, began producing a distinct fluorescent signal relative to the untransformed M82 plants. This fluorescent signal from the drought sensor plants increased in intensity as time progressed in the study, up to the point where symptoms of the drought stressor were visibly present, at 96 hours after water was initially withheld. Once plants were rewatered at the 96-hour timepoint, the fluorescent signal from the drought sensor plants decreased in intensity as the stressor was mitigated. Notably, the drought stressorsensor plant allowed for early identification of the drought stress, which could allow for early mitigation of a stressor prior to visible symptoms starting to develop in a plant. As such, the sensor plants described herein can improve plant health, yield, and provide for numerous benefits and efficiencies associated with early intervention to protected against plant stressors.Example 2: Soybean Sensor Plants.

[0134] The ability of sensor plants to respond to a fungal stressor was assessed using soybean plants (Thorne background) that were genetically engineered with a fungal-sensing promoter-reporter pair. The fungal stressor promoter-reporter pair comprised a promoter element from a tomato chitinase gene operably connected to a bfloGFP reporter gene. A total of 28 soybean plants grown to the second trifolate stage (V2) in a growth chamber under a long day condition (16 / 8 h light / dark cycles) at a constant temperature of 24°C. The plants were then divided (n=7 per group) into four different treatment groups (mock, fungal, fungal + tebuconazole fungicide, and fungal + triticonazole fungicide). At timepoint zero, the mock group was treated with water and a surfactant (0.05% Tween-20), whereas the other three groups were exposed to fungus by application of Cercospora spores prepared in water + surfactant (0.05% Tween-20) at a concentration of approximately 150,000 spores / mL. For groups receiving a fungicide, the respective fungicide was applied by foliar treatment using a Preval sprayer at 24 hours after exposure to the fungal spores. Fluorescence was measured by intensity from fluorescent images that were collected with a camera in a custom imaging box. Disease progression was monitored visually, with scoring ranging from 0-10 (a score of 0 for no symptoms present and a score of 10 for the most severe symptoms).

[0135] FIG. 4A shows a time course of visual disease scoring for fungal sensor plants exposed to the Cercospora spores. Plants first demonstrated visual disease at 192 hours post-exposure to the fungus. Plants that were exposed to the fungus with no fungicide treatment exhibiting high visual disease scores at 192 hours relative to the other groups of plants where fungicides were applied 24 hours after exposure to the fungus.

[0136] Figure 4B shows the rapid increase in fluorescent signal observed in all plants exposed to the fungus. Notably, within 24 hours of the fungus exposure, a marked increase in fluorescent signal was measured in all fungal sensor plants. As such, FIG. 4B shows that, compared to FIG. 4A, the fungal sensor plants provide an indication of a fungal stressor approximately seven days prior to the onset of visual disease symptoms. The fluorescent signal from the fungal sensor plants increased after application of both tested fungicides (tebuconazole or triticonazole). Without being bound to any particular theory, it is believed that the fungicides interacted with the promoter-reporter pair to keep the signal on and potentially amplified the induced fluorescent signal. Nonetheless, the fungal stressor sensor plant allowed for a significantly earlier identification of the fungal infection relatively to reliance on visual symptoms of disease, which could allow for earlier mitigation of a stressor prior to visible symptoms detection in a plant. As such, the sensor plants described herein can improve plant health, yield, and provide for numerous benefits and efficiencies associated with early intervention to protect against stressors.

[0137] As a person skilled in the art will recognize from the previous detailed description and from the examples, figures, and claims, modifications and changes can be made to the embodiments of the invention without departing from the scope of this invention as defined in the following claims.

Claims

CLAIMSWhat is claimed is:1 . A method for curtailing excessive exposure of an agricultural environment to a chemical plant treatment, comprising: a. capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; b. detecting a signal in the spectral images of the sensor plants in response to a stressor; and c. selectively applying a chemical treatment targeting the stressor to a subgroup of plants in the agricultural environment based on the detected signal.

2. The method of claim 1 , wherein the stressor is a fungal stressor.

3. The method of claim 2, wherein the treatment is a fungicide.

4. The method of claim 3, wherein the fungicide is selected from a triazole, a strobilurin, and a succinate dehydrogenase inhibitor.

5. The method of claim 1 , wherein the stressor is an insect stressor.

6. The method of claim 5, wherein the treatment is an insecticide.

7. The method of claim 6, wherein the insecticide is selected from an organochlorine, an organophosphate, an organosulfur, a carbamate, a formamidine, a dinitrophenol, an organotin, a pyrethroid, a nicotinoid, a spinosyn, a pyrazole, a pyridazinone, a quinazoline, a botanical, a synergist / activator, an antibiotic, a fumigant, an inorganic, a biorational, and a benzoylurea.

8. The method of claim 1 , wherein the chemical plant treatment is a fertilizer.

9. The method of claim 8, wherein the fertilizer is selected from a single nutrient fertilizer, a multinutrient fertilizer, and a micronutrient.

10. The method of claim 1 , wherein the chemical treatment is a herbicide.11 . The method of claim 10 wherein the herbicide is selected from a herbicide that inhibits acetyl coenzyme A carboxylase (ACCase), a herbicide that inhibits acetolactase synthase (ALS), a herbicide that inhibits enolpyruvylshikimate 3- phosphate synthase enzyme (EPSPS), an auxin-like herbicide, a herbicide that inhibits photosystem II, a herbicide that inhibits photosystem I, and a herbicide that inhibits 4-hydroxyphenylpyruvate dioxygenase.

12. The method of claim 1 , wherein stress levels are mapped to plants in the agricultural environment based on the detected signal.

13. A method for improving plant treatment efficiency, comprising: a. capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; b. detecting a signal in the spectral images of the sensor plants in response to a stressor; and c. selectively applying a treatment targeting the stressor to a subgroup of plants in the agricultural environment based on the detected signal.

14. The method of claim 13, wherein the stressor is a fungal stressor.

15. The method of claim 14, wherein the treatment is a fungicide.

16. The method of claim 15, wherein the fungicide is selected from a triazole, a strobilurin, and a succinate dehydrogenase inhibitor.

17. The method of claim 13, wherein the stressor is an insect stressor.

18. The method of claim 17, wherein the treatment is an insecticide.

19. The method of claim 18, wherein the insecticide is selected from an organochlorine, an organophosphate, an organosulfur, a carbamate, a formamidine, a dinitrophenol, an organotin, a pyrethroid, a nicotinoid, a spinosyn, a pyrazole, a pyridazinone, a quinazoline, a botanical, a synergist / activator, an antibiotic, a fumigant, an inorganic, a biorational, and a benzoylurea.

20. The method of claim 13, wherein the chemical plant treatment is a fertilizer.21 . The method of claim 20, wherein the fertilizer is selected from a single nutrient fertilizer, a multinutrient fertilizer, and a micronutrient.

22. The method of claim 13, wherein the chemical treatment is a herbicide.

23. The method of claim 22 wherein the herbicide is selected from a herbicide that inhibits acetyl coenzyme A carboxylase (ACCase), a herbicide that inhibits acetolactase synthase (ALS), a herbicide that inhibits enolpyruvylshikimate 3- phosphate synthase enzyme (EPSPS), an auxin-like herbicide, a herbicide that inhibits photosystem II, a herbicide that inhibits photosystem I, and a herbicide that inhibits 4-hydroxyphenylpyruvate dioxygenase.

24. The method of claim 13, wherein stress levels are mapped to plants in the agricultural environment based on the detected signal.

25. A method of screening plants for treatment, comprising: a. capturing spectral images of one or more sensor plants in an agricultural environment, wherein the one or more sensor plants have been genetically engineered to produce a signal in response to a stressor; b. detecting a signal in the spectral images of the sensor plants in response to a stressor; andc. selecting a subgroup of plants in the agricultural environment for treatment based on the detected signal, wherein the treatment is achieved by a chemical targeting the stressor.

26. The method of claim 25, wherein the stressor is a fungal stressor.

27. The method of claim 26, wherein the treatment is a fungicide.

28. The method of claim 27 wherein the fungicide is selected from a triazole, a strobilurin, and a succinate dehydrogenase inhibitor.

29. The method of claim 25, wherein the stressor is an insect stressor.

30. The method of claim 29, wherein the treatment is an insecticide.

31. The method of claim 30, wherein the insecticide is selected from an organochlorine, an organophosphate, an organosulfur, a carbamate, a formamidine, a dinitrophenol, an organotin, a pyrethroid, a nicotinoid, a spinosyn, a pyrazole, a pyridazinone, a quinazoline, a botanical, a synergist / activator, an antibiotic, a fumigant, an inorganic, a biorational, and a benzoylurea.

32. The method of claim 25, wherein the chemical plant treatment is a fertilizer.

33. The method of claim 32, wherein the fertilizer is selected from a single nutrient fertilizer, a multinutrient fertilizer, and a micronutrient.

34. The method of claim 25, wherein the chemical treatment is a herbicide.

35. The method of claim 34, wherein the herbicide is selected from a herbicide that inhibits acetyl coenzyme A carboxylase (ACCase), a herbicide that inhibits acetolactase synthase (ALS), a herbicide that inhibits enolpyruvylshikimate 3- phosphate synthase enzyme (EPSPS), an auxin-like herbicide, a herbicide that inhibits photosystem II, a herbicide that inhibits photosystem I, and a herbicide that inhibits 4-hydroxyphenylpyruvate dioxygenase.

36. The method of claim 25, wherein stress levels are mapped to plants in the agricultural environment based on the detected signal.

37. A method of testing the efficacy of a treatment for plants, comprising: a. applying a stressor to one or more sensor plants in an agricultural environment, wherein the sensor plants have been genetically engineered to produce a signal in response to a stressor; b. capturing spectral images of the sensor plants in the agricultural environment; c. detecting a signal in the spectral images of the sensor plants in response to the stressor; d. selectively applying a treatment to the plants in the agricultural environment based on the detected signal; and e. recording the change in signal after the treatment has been applied.

38. The method of claim 37, wherein the stressor is a fungal stressor.

39. The method of claim 38, wherein the treatment is a fungicide.

40. The method of claim 39, wherein the fungicide is selected from a triazole, a strobilurin, and a succinate dehydrogenase inhibitor.41 .The method of claim 37, wherein the stressor is an insect stressor.

42. The method of claim 41 , wherein the treatment is an insecticide.

43. The method of claim 42, wherein the insecticide is selected from an organochlorine, an organophosphate, an organosulfur, a carbamate, a formamidine, a dinitrophenol, an organotin, a pyrethroid, a nicotinoid, a spinosyn, a pyrazole, a pyridazinone, a quinazoline, a botanical, a synergist / activator, an antibiotic, a fumigant, an inorganic, a biorational, and a benzoylurea.

44. The method of claim 37, wherein the chemical plant treatment is a fertilizer.

45. The method of claim 44, wherein the fertilizer is selected from a single nutrient fertilizer, a multinutrient fertilizer, and a micronutrient.

46. The method of claim 37, wherein the chemical treatment is a herbicide.

47. The method of claim 46, wherein the herbicide is selected from a herbicide that inhibits acetyl coenzyme A carboxylase (ACCase), a herbicide that inhibits acetolactase synthase (ALS), a herbicide that inhibits enolpyruvylshikimate 3- phosphate synthase enzyme (EPSPS), an auxin-like herbicide, a herbicide that inhibits photosystem II, a herbicide that inhibits photosystem I, and a herbicide that inhibits 4-hydroxyphenylpyruvate dioxygenase.

48. The method of claim 37, wherein stress levels are mapped to plants in the agricultural environment based on the detected signal.

49. A method of testing the efficacy of a treatment for plants, comprising a. capturing spectral images of one or more sensor plants in an agricultural environment, wherein the sensor plants have been genetically engineered to produce a signal in response to a stressor; b. detecting a signal in the spectral images of the sensor plants in response to the stressor; c. selectively applying a treatment to the plants in the agricultural environment based on the detected signal; and d. recording the change in signal after the treatment has been applied.

50. The method of claim 49, wherein the stressor is a fungal stressor.51 . The method of claim 50, wherein the treatment is a fungicide.

52. The method of claim 51 , wherein the fungicide is selected from a triazole, a strobilurin, and a succinate dehydrogenase inhibitor.

53. The method of claim 49, wherein the stressor is an insect stressor.

54. The method of claim 53, wherein the treatment is an insecticide.

55. The method of claim 54, wherein the insecticide is selected from an organochlorine, an organophosphate, an organosulfur, a carbamate, a formamidine, a dinitrophenol, an organotin, a pyrethroid, a nicotinoid, a spinosyn, a pyrazole, a pyridazinone, a quinazoline, a botanical, a synergist / activator, an antibiotic, a fumigant, an inorganic, a biorational, and a benzoylurea.

56. The method of claim 49, wherein the chemical plant treatment is a fertilizer.

57. The method of claim 56, wherein the fertilizer is selected from a single nutrient fertilizer, a multinutrient fertilizer, and a micronutrient.

58. The method of claim 49, wherein the chemical treatment is a herbicide.

59. The method of claim 58, wherein the herbicide is selected from a herbicide that inhibits acetyl coenzyme A carboxylase (ACCase), a herbicide that inhibits acetolactase synthase (ALS), a herbicide that inhibits enolpyruvylshikimate 3- phosphate synthase enzyme (EPSPS), an auxin-like herbicide, a herbicide that inhibits photosystem II, a herbicide that inhibits photosystem I, and a herbicide that inhibits 4-hydroxyphenylpyruvate dioxygenase.

60. The method of claim 49, wherein stress levels are mapped to plants in the agricultural environment based on the detected signal.61 . The method of any one of claims 1 -60, wherein the one or more sensor plants is a dicot.

Images