Soil decomposition activity sensors

Abstract

An exemplary system and method are disclosed that employs a soil sensor device comprising multiple soil sensors configured in an array on a single substrate that can sense soil at multiple depths and / or locations to provide an aggregated sensor output that is representative of a heterogenous soil area or volume, e.g., to monitor microbial activity.

Description

SOIL DECOMPOSITION ACTIVITY SENSORSCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. provisional patent application No. 63 / 634,317, filed on April 14, 2024, and titled “SOIL DECOMPOSITION ACTIVITY SENSORS,” the disclosure of which is expressly incorporated herein by reference in its entirety7.STATEMENT REGARDING FEDERALLY FUNDED RESEARCH

[0002] This invention was made with government support under grant number DE- AR0001602 awarded by the U.S. Dept, of Energy7, and grant number 2019-05291 awarded by USDA - NIFA. The government has certain rights in the invention.BACKGROUNDThere are benefits to sensing soil microbes and their enzymatic and other decomposition activities. Current techniques are typically employed in a laboratory7and are time and labor- intensive. Such offsite assessments are not necessarily reflective of in situ bio-chemical-physical processes and may not capture temporal variations that can be important.

[0003] Evaluations of microbial activity' in incubated soils and with microbial biomass carbon (MBC) in field trials employ large numbers of devices to improve measurement variability7in field measurements due to soil stochasticity.SUMMARYAn exemplary' system and method are disclosed that employs a soil sensor device comprising multiple soil sensors configured in an array on a single sensor substrate that can sense soil at multiple depths and / or locations to provide an aggregated sensor output that is representative of a heterogenous soil area or volume, e.g., to monitor microbial activity7, as a function of soil decomposition. The exemplary system can be deployed in scale over a field to evaluate soil microbial and enzymatic activities using low-cost soil sensor devices that provide higher accuracy of measurements, via system approach design, for subsequent analysis.

[0004] In some aspects, implementations of the present disclosure include a soil sensor including: a circuit substrate; and a plurality of soil sensing elements formed on the circuit substrate, including a first sensing element and a second sensing element arranged in a parallel circuit configuration, wherein the soil sensor is configured to measure a combined response of the plurality of soil sensing elements.

[0005] In some aspects, implementations of the present disclosure include a soil sensor, wherein at least one of the soil sensing elements includes a decomposable conductive ink, and wherein the plurality of soil sensing elements includes the decomposable conductive ink.

[0006] In some aspects, implementations of the present disclosure include a soil sensor, wherein the plurality of soil sensing elements are arranged in a grid.

[0007] In some aspects, implementations of the present disclosure include a soil sensor, wherein the grid includes at least two rows of adjacent soil sensing elements of the plurality of soil sensing elements.

[0008] In some aspects, implementations of the present disclosure include a soil sensor further including a computing device operably coupled to the plurality of soil sensing elements, wherein the computing device is configured to measure a property of the soil using the soil sensor.

[0009] In some aspects, implementations of the present disclosure include a soil sensor, wherein the computing device includes persistent memory’ storage.

[0010] In some aspects, implementations of the present disclosure include a soil sensor, w herein the computing device includes a wireless network interface.

[0011] In some aspects, implementations of the present disclosure include a soil sensor, wherein the parallel circuit configuration is coupled to a multiplexer, and wherein the multiplexer is configured to measure the combined response of the plurality of soil sensing elements by determining that at least one soil sensing element is an outlier, and rejecting measurements of the outlier.

[0012] In some aspects, implementations of the present disclosure include a soil sensor, wherein the multiplexer is configured to record a plurality of separate responses corresponding to the plurality of soil sensing elements.

[0013] In some aspects, implementations of the present disclosure include a soil sensor, wherein the computing device is configured to estimate a distribution of microbial activity based on the plurality of separate responses corresponding to the plurality of soil sensing elements.

[0014] In some aspects, implementations of the present disclosure include a method including: positioning a plurality' of soil sensors in a field, including a first soil sensor at a first field location and a second soil sensor at a second field location; and measuring, at periodic interval, the response of the soil sensing elements of the plurality of determining, via the measured resistance value, microbial activity for the field.

[0015] In some aspects, implementations of the present disclosure include a method, wherein the microbial activity is determined as a slope of the measured resistance versus time.

[0016] In some aspects, implementations of the present disclosure include a method, wherein the slope is positive when the microbial activity increases, and negative when the microbial activity decreases.

[0017] In some aspects, implementations of the present disclosure include a system including: a plurality of soil sensors, wherein each soil sensor of the plurality of soil sensors includes: a circuit substrate, a plurality of soil sensing elements formed on the circuit substrate, including a first sensing element and a second sensing element arranged in a parallel circuit configuration, wherein the soil sensor is configured to measure a combined response of the plurality of soil sensing elements; and a computing device operably coupled to the plurality of soil sensing elements, wherein the computing device is configured to measure a property of the soil using the soil sensor, a controller operably coupled to the computing devices of the plurality of soil sensors by a network, wherein the controller is configured to: measure, at a periodic interval, a plurality of resistance values corresponding to the plurality of soil sensing elements; determine, a soil property for a field the plurality of soil sensors are deployed in.

[0018] In some aspects, implementations of the present disclosure include a system, wherein the soil property is microbial activity.

[0019] In some aspects, implementations of the present disclosure include a system or claim 15, wherein the network is a wireless network, and wherein the computing device of each of the plurality of soil sensors includes a wireless network interface.

[0020] In some aspects, implementations of the present disclosure include a soil sensor, wherein the parallel circuit configuration of each of the plurality of soil sensors is coupled to a multiplexer of each soil sensor, and wherein the multiplexer of each soil sensor is configured to separately measure the responses of soil sensing elements.

[0021] In some aspects, implementations of the present disclosure include a soil sensor, wherein the multiplexer is configured to record a plurality of separate responses corresponding to the plurality of soil sensing elements.

[0022] In some aspects, implementations of the present disclosure include a system, wherein the computing device is configured to transmit the plurality' of responses to the controller.

[0023] In some aspects, implementations of the present disclosure include a system, wherein the controller is configured to determine a distribution of microbial activity based on the plurality of separate responses corresponding to the plurality of soil sensing elements.

[0024] It should be understood that the above-described subject matter may also be implemented as a computer-controlled apparatus, a computer process, a computing system, or an article of manufacture, such as a computer-readable storage medium.

[0025] Other systems, methods, features and / or advantages will be or may become apparent to one with skill in the art upon examination of the following drawings and detailed description. It is intended that all such additional systems, methods, features and / or advantages be included within this description and be protected by the accompanying claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The components in the drawings are not necessarily to scale relative to each other. Like reference numerals designate corresponding parts throughout the several views.

[0027] FIG. 1 A illustrates an example soil sensor, according to implementations of the present disclosure.

[0028] FIG. IB illustrates an example system including the example soil sensor of FIG. 1 A, according to implementations of the present disclosure.

[0029] FIG. 2 illustrates a multiplexer circuit configured to address sensing elements, according to implementations of the present disclosure.

[0030] Figs. 3 A and 3B show configurations for the soil sensor coupled to a controller as a data logger.

[0031] FIGS. 4A-4H illustrate example configurations of the soil sensor having a sensing element array, according to various implementations of the present disclosure.

[0032] FIGS. 5A-5C illustrate example axial placements of sensing elements around a soil sensor, according to various implementations of the present disclosure.

[0033] FIG. 6 illustrates an example computing device.

[0034] FIG. 7 illustrates results of a study of four segment PCB sensors in controlled outdoor environments.

[0035] FIGS. 8A-8B illustrate soil sensors used in a study of an example implementation of the present disclosure.

[0036] FIGS. 8C-8D illustrate results from the sensors shown in FIGS. 8A-8B, according to a study of an example implementation of the present disclosure.

[0037] FIGS. 9A-9H illustrate results of a study of example soil sensor topologies, according to a study of an example implementation of the present disclosure.

[0038] FIG. 10A illustrates an example sensor being buried in a field.

[0039] FIG. 10B illustrates an example smoothed response of glass and PCB sensors, according to implementations of the present disclosure.

[0040] FIG. 10C illustrates CO2 readings for an example experiment, according to a study of an example implementation of the present disclosure.

[0041] FIG. 11 illustrates an estimate of Leaf Area Index, according to a study of an example implementation of the present disclosure.

[0042] FIG. 12 illustrates example dimensions that can be used for the traces of soil sensors, according to implementations of the present disclosure.

[0043] FIGS. 13A-13D illustrate example initial resistances of soil sensors, according to implementations of the present disclosure.

[0044] FIG. 14 illustrates an experimental setup and example geometries of sensors, according to a study of an example implementation of the present disclosure.

[0045] FIGS. 15A-15C illustrate experimental results for drought and flooded conditions, according to a study of an example implementation of the present disclosure.

[0046] FIG. 16 illustrates laboratory measurements collected to characterize the soils and recovery7of example soil plots, according to a study of an example implementation of the present disclosure.

[0047] FIG. 17A illustrates soil volumetric moisture content, according to a study of an example implementation of the present disclosure.

[0048] FIG. 17B illustrates soil temperature, according to a study of an example implementation of the present disclosure.

[0049] FIG. 17C illustrates CChflux, according to a study of an example implementation of the present disclosure.

[0050] FIG. 17D illustrates air temperature, according to a study of an example implementation of the present disclosure.

[0051] FIG. 18 illustrates sensor signals over the course of a study of an example implementation of the present disclosure.

[0052] FIG. 19 illustrates a comparison of the CO2 flux and the decomposition sensor slope in a study of an example implementation of the present disclosure.

[0053] FIG. 20 illustrates a comparison of Treatment and Control conditions evaluated on CO2 accumulation slope or integral in a study of an example implementation of the present disclosure.DETAILED DESCRIPTION

[0054] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. Methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present disclosure. As used in the specification, and in the appended claims, the singular forms “a,” "an." "the" include plural referents unless the context clearly dictates otherwise. The term ■‘comprising” and variations thereof as used herein is used synonymously with the term “including” and variations thereof and are open, non-limiting terms. The terms “optional” or “optionally” used herein mean that the subsequently described feature, event or circumstance may or may not occur, and that the description includes instances where said feature, event or circumstance occurs and instances where it does not. Ranges may be expressed herein as from "about" one particular value, and / or to "about" another particular value. When such a range is expressed, an aspect includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent "about," it will be understood that the particular value forms another aspect. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint. While implementations will be described for measuring soil, it will become evident to those skilled in the art that the implementations are not limited thereto, but are applicable for any other type of sensing.

[0055] Example Device

[0056] Fig. 1 A shows an example soil sensor 130 (e.g., a decomposition activity sensor) configured with multiple parallel sensing elements 110 (e g., resistive elements) configured to provide a single aggregated resistance measurement of soil at a given location. The example sensor is 140 mm by 31 mm and employs decomposable ink, poly(hydroxybutyrate-co-3- hydroxyvalerate) (PHBV) and flake carbon, printed on a circuit board that can couple to a wireless transceiver (e.g., through traces 120) to provide continuous microbial and enzymatic activity monitoring for a plot of farmland or other body of soil. The sensors can be configured with other dimensions. The exemplary system can be deployed in scale over a field to evaluate soil microbes and enzymatic activities using low-cost soil sensor devices that provide higher accuracy of measurements, via system design, for subsequent analysis.

[0057] In having multiple sensors co-located at different positions on a single sensor substrate and in a parallel topology’, the configuration can provide more accurate measurementsin having a larger sensing area. In addition, the parallel topology of the sensors, via resistive measurements, beneficially provided outlier exclusion functions of an individual faulty sensor (e.g., due to low cost, low-grade material, and high variability associated with low-cost in manufacturing) without the need for filters or processing that would otherwise would need to be implemented to have sufficient or reasonable quality in the measurement (i.e., for the acquired data to be meaningful from an engineering or scientific perspective).

[0058] The outlier exclusion functions can be attributed to the characteristic of resistive elements being added in a parallel topology7as an inverse of one another (e.g., 1 / Rsensori + 1 / Rsensor2 + 1 / Rsensor3 + etc), which would reduce the effect of a failed sensor in the set in a resistance measurement. For example, if ”sensor2“ of the set was damaged during installation or is positioned in the ground at a position that has an open soil pocket (which tends to happen with heterogenous complex soil measurement), the measurement can be measured an open having a high resistance value in the mega-ohm whereas the other sensor measurement that are likely operating as intended are measuring in the k-ohms. In this scenario, l / (large mega ohm) value would have little effect to the l / (k-ohms) measurements for the other sensors as they are added in parallel. To this end, low-cost fabrication / manufacturing of the device can be implemented for high-end measurements without the need for additional circuitries or control components that could increase the cost of the sensor fabrication. The design allows for rigorous engineering and scientific work in application domain that are very price sensitive.

[0059] FIG. IB illustrates an example system for monitoring a field that can use any of the soil sensors described herein. The system can include any number of soil sensors 130a, 130b, 130c, 130d, 130e, 130f distributed across a field. The soil sensors 130a-130f can be coupled to a controller 140 (e.g., by a wired or wireless network), so that the controller 140 can measure the response of all the soil sensors 130a-130f and measure a soil property of the field (e.g., soil decomposition activity). The soil sensors 130a-130f can optionally include multiplexers and / or computing devices, as described further below.

[0060] Multiplexer

[0061] The soil sensor can include a multiplexer circuit that can address one or more of the sensing elements, as shown in FIG. 2. The multiplexer 122 can be employed to perform measurements of each sensing element 110 of the soil sensor. In the example show n in FIG. 2, the array of sensing elements 110 terminates at a common node for the sensing element. The multiplexer provides for the selection of 1 of n pins to allow the selection of individual sensingelements for the measurement. Selection pins (Sei 0 and Sell) can be employed to select the individual sensing element.

[0062] In some embodiments, the sensing may be performed in a sequence to acquire measurements for each of the sensing elements individually.

[0063] Remote Datalogger

[0064] In some embodiments, the circuit board sensor can couple to a datalogger, e.g., configured with data storage, e.g., a low-cost microcontroller coupled to an SD card.Figs. 3A and 3B show configurations for the soil sensor coupled to a controller as a data logger. In Fig. 3A, the microcontroller is connected to the sensing elements via a multiplexer. In some embodiments, the microcontroller is connected to the sensing elements, e.g.. as shown and described in relation to FIGS. 4A - 4F. The microcontroller can optionally evaluate the resistance or electrical response of the sensing elements.In Fig. 3A, the microcontroller is coupled to a removable data storage device, e.g., an SD card. In other embodiments, the microcontroller is coupled to a persistent memory device, e.g., EEPROM, to store the measurement. The microcontroller may include a serial interface to allow writing and reading of the persistent memory device. In Fig. 3B, the microcontroller is coupled to a wireless network interface for wireless communication. One example protocol that can be used for the wireless network interface is the LoRa protocol. In such embodiments, to conserve power for longer-term data logging, the microcontroller may be triggered to enable the wireless network interface to transmit the data to a remote computing device.

[0065] Sensing Element Array

[0066] The soil sensors can be positioned in a number of different arrangements and / or a number of different combinations. Figs. 4A-4F show example configurations of the soil sensor 130 having a sensing element array in accordance with various embodiments.

[0067] Fig. 4A shows four sensing elements 110 arranged in parallel configuration, where the sensing elements are coupled in parallel between traces 120. The sensing elements 110 are arranged uniformly apart from one other.

[0068] Fig. 4B shows four sensing elements 1 10 arranged in a parallel configuration. The sensing elements 1 10 are arranged with different spacing apart from one other.

[0069] Fig. 4C shows another configuration having n sensing elements 110 arranged in a parallel configuration. The number of n may be 3, 4, 5, 6, 7, 8. It can be similarly arranged with different spacings as shown in relation to Fig. 4B.

[0070] FIGS. 4D-4F show different numbers and arrangements of traces 120 joining the sensing elements 1 10 together. Any combination of parallel and / or series connections is possible through any number of traces 120. Fig. 4D shows an array of 2 x 4 sensing elements 110 arranged as two sets of parallel circuits having n number of sensing elements. The number of n may be 3, 4, 5, 6. 7, 8.

[0071] Fig. 4E shows an array of 3 x 3 sensing elements 110 arranged as three sets of parallel circuits having n number of sensing elements. The number of n may be 3, 4, 5, 6, 7, 8.

[0072] Fig. 4F shows an array of 2 x 2 sensing elements 110 arranged with an offset from one other. With more sensing elements, the array can be configured as a mesh.

[0073] FIG. 4G shows an array of sensing elements 110 coupled by three traces 120. As shown in FIG. 4G, the traces 120 can be different lengths.

[0074] FIG. 4H shows an additional example array with two traces 120 coupling eight sensing elements 110.

[0075] In some embodiments, the sensing element array is arranged on two sides (front and back) of the soil sensor. Alternatively or additionally, sensing elements 110 can be positioned axially around the soil sensor 130. As shown in FIGS. 5A-5C, both symmetrical and asymmetrical configurations are possible, and both symmetrical and asymmetrical configurations can be configured with any number of sensing elements 110

[0076] It should be appreciated that other geometric configurations for the soil sensor may be employed.

[0077] Experimental Results and Additional Examples

[0078] Example 1

[0079] A study was conducted to develop a soil decomposition activity sensor with degradable conductors or components. The sensing of soil microbial and enzymatic activity continues to be a challenge, as current techniques are typically limited to the laboratory and are time and labor-intensive. In addition, such offsite assessments are not necessarily reflective of in situ bio-chemical-physical processes.

[0080] The study developed printed decomposition sensors using poly(hydroxybutyrate-co- 3 -hydroxy valerate) (PHBV) and carbon composite material that can measure the microbial activity in incubated soils and with microbial biomass carbon (MBC) in field trials based on sensor response. These sensors included a single fuse-like resistive element. Field measurements by the sensor were observed to be subject to stochastic effects in soil.

[0081] To address the stochastic heterogeneous features of the soil, the study further developed a field soil sensor having multiple numbers of parallelizing sensing elements. An example second device included multiple printed sensing elements on a custom printed circuit board (PCB).

[0082] The study observed the second device to be effective at smoothing out sensor response at a field demonstration conducted at a potato farm in the Upper Midwest region of the United States. To further shape the signal response of these decomposition sensors, different parallel topologies were studied by vary ing the number of sensing elements, element width, and element length. FIG. 1 A illustrates a parallel decomposition sensor printed on a PCB, including example overall dimensions of sensor package and an illustration of a stencil printing process where the cut stencil is applied with transfer tape, the tape is peeled off, the ink is manually applied with a razor blade, and the stencil is peeled off.

[0083] Materials and methods

[0084] Sensor fabrication and data logger design

[0085] Example printed circuit boards (PCBs) were designed as a 130 x 25 mm “stake” with a tapered end. The connection points include two vias. Contact points for printed traces include pairs of exposed square pads. Pads were 2 mm squares for 4-segment PCBs and 1 mm square for all other PCBs, e g., as shown in FIG. 1 A. For segment counts greater than 1. pairs of pads were connected in parallel. All pads and vias were immersion gold-plated by the fabricator to prevent corrosion in soil and liquid environments.

[0086] For 4 segment sensors, stencils were built up by placing two layers of polyimide tape forming an average stencil thickness of 104 pm. For 1, 3, 5, and 7 segment sensors, stencils were fabricated from adhesive vinyl using a computer-controlled cutter, forming an average thickness of 85 pm. These stencils were then applied with the aid of transfer tape. Decomposition sensing inks were formulated and stencil -printed with a razor blade (Atreya, et al., 2023). Wires were soldered to the vias of the PCB. The PCB was then placed inside a custom carriage which was fabricated from FDM-printed poly(lactic acid) (PLA). The potting dam portion of the carriage was filled with liquid rubber sealant.

[0087] U-shaped single conductor sensors for outdoor deployment (“single-trace glass sensors” or “glass sensors”) were printed on 25 x 75 mm glass slides using a laser-cut poly imide tape stencil forming an average print profile height of 45.4 ± 10.0 pm. Wires were attached to the trace with conductive silver epoxy. The slide was then placed inside an FDM-printed PLAcarriage. The potting dam portion of the carriage was filled with the same liquid rubber sealant mentioned above.

[0088] Initial resistance measurements of sensors were taken using a Keithley 2110 DMM.Resistance measurements during all laboratory and field trials were taken approximately every 30 minutes with a custom data logger circuit comprising a SD card-enabled microcontroller, a power timer, real-time clock (DS3231), and a voltage divider for every sensor with a 2.2 kQ reference resistor.

[0089] Sensor testing in laboratory. The study used soil in all laboratory experiments that had been sieved with # sieve and stored at room temperature. Sand was sterilized at 120 °C. 4 segment sensors were tested at 35°C vertically in soil or sand inside 1 liter plastic containers with a single drainage hole pierced at the base.

[0090] Laboratory tests of 1, 3, 5, and 7 segment PCB sensors were conducted at 30°C with sensors placed horizontally in the soil in 5 7-liter plastic containers, which had 5 drainage holes pierced at the base. For each sensor topology, each replicate was placed in a separate container with 3-4 different sensors in each container. In all laboratory experiments, the soil was fully saturated with tap water and re-saturated three times a week. Deionized water was used for the sand test in place of tap water.

[0091] Sensor deployment in outdoor pot. Outdoor tests conducted in a pot were conducted in unsieved potting soil in a large pot placed outdoors in Boulder. CO, USA, (40.009716, - 105.241149). Soil moisture was measured with a commercial TDR moisture sensor. Three glass sensors and three 4 segment PCB sensors were buried in the pot. 48-point moving average filter was applied to smooth sensor response data.

[0092] Sensor deployment in agricultural fields. Sets of three 4 segment PCB and three single trace glass sensors were deployed in two commercial potato fields in the Upper Midwest region of the United State of America. For all sensors, installation occurred at 10 cm below the surface with reference to the tops of furrows. A 10 cm hole was excavated with a trowel. At the base of the hole, a sensor installation cavity was created with a custom installation tool in the case of the glass and PCB sensors, and with an auger in the case of the CO2 sensors. Installations were arrayed around the logger enclosure, maximizing coverage by installing each pair of sensors 5.5m from the logger and co-locating a glass and PCB sensor at each maximum distance.

[0093] At the central location defined by the logger, the study installed a sub-surface CO2 sensor in each field along with associated logging and power equipment, also at a depth of 10 cm from the top of the furrow. These custom systems use an off-the-shelf CO2 sensor weatherizedwith marine epoxy and a polycarbonate enclosure with a waterproof-breathable membrane to enable soil deployment. The CO2 sensor uses an LED light source for extreme energy efficiency, and supports reading ranges between 400-5, OOOppm (high accuracy, ± 3%) and 5,000- 10,000ppm (low accuracy, ± 10%). A weatherized enclosure houses readout electronics which write time-stamped data to an SD card, connected to the CO2 sensor via a Im cable. The enclosure is mounted beneath a 2.5 W solar panel for autonomous logging. The two fields were selected for similar attributes, including location and crop type. The field labeled Field A was fumigated in the previous fall with a soil injection application of 40gal / A 42% Metam sodium product, while the field labeled "Field B” was not fumigated. Fungicide (azoxystrobin) and insecticide (thiamethoxam) were applied in-furrow in both fields. Both fields were also fertilized (350 lbs N, 20 lbs, P, 470 lbs L) and irrigated similarly (1 1.5”). Crops were planted in both fields on April 26, 2023. Decomposition sensors were deployed July 31, 2023. For plotting, decomposition sensor response was averaged for each treatment and smoothed using a 48 point moving average filter. Linear regression was conducted for decomposition ranges on unsmoothed data (Atreya, 2023)(Sharpe, 2024).

[0094] Processing remote sensing data. Leaf area index (LAI) for the two crop fields was computed from data from Sentinel-2 Earth observation satellite using Google Earth Engine. The image collection was filtered for images with a maximum 50% cloud coverage. Clouds and cloud shadows were masked using the s2cloudless algorithm. Instances where multiple values were found for a given day, values were averaged. From the resulting data set, Enhanced Vegetation Index (EVI) was calculated using the following equation (Boegh, et al, 2022)(Liu & Huete, 1995).PNIR “I- 6 * prea 7.5 * pbiue+ 1

[0095] Where pN[Ris the near-infrared reflectance, predis the red reflectance, and Pbiueis the blue reflectance.

[0096] The Leaf Area Index was then calculated using the following equation (Boegh, et al, 2022).LAI = 3.618 * EVI - 0.118

[0097] Results and Discussion

[0098] Parallelizing. Soil is a complex, heterogenous, and dynamic environment, making many forms of in situ sensing challenging. Moisture, temperature, soil average particle size, and contact mechanics at the interface of the soil medium and the sensor can affect sensor readings.In addition, drastic changes in those factors, caused by changepoint events such as sudden weather shifts, can cause false sensor readings. Single-trace decomposition sensors are susceptible to such stochastic effects (Atreya, 2023) (Sharpe, 2024). The slopes of resistance versus time curves were taken at certain time intervals and correlated with microbial activity. Then, to account for variability’, slopes from multiple sensors in similar soil conditions were often averaged together. FIG. 7 illustrates results of a study of four segment PCB sensors in controlled outdoor environments.

[0099] The example implementation mitigated the effects of soil and sensor variability' by parallelizing multiple printed sensing elements together on a single printed circuit board (PCB). Initially, 4-segment PCB sensors were fabricated by printing PHBV-C inks on custom PCBs with the aid of polyimide tape stencils (see FIG. 8A), and then integrating these PCBs into custom protective carriages (FIG. 8B). FIG. 8C show s parallelized PCB decomposition sensors were observed to yield expected responses in laboratory-incubated soil and sterilized sand, with an increasing resistance in microbial active soil and no response in sand.

[0100] Single-trace sensors printed on glass and encased in protective carriages ("‘glass sensors”), and 4 segment PCB sensors w ere deployed in a large container of potting soil outdoors in Boulder, Colorado, USA, in late summer / early autumn. FIG. 8D shows the results from this experiment, wherein glass sensors (indicated as lines Gl, G2, and G3), demonstrated varied behavior from early on. Sensors Gl and G2 both began to show a significant increase in resistance at about 5 days with Gl showing a greater slope. Sensor G3 is generally unresponsive throughout the span of the experiment, likely resulting from poor soil contact or other heterogeneities within the soil container. The technique first described by Atreya, et al. and Sharpe, et al. to analyze these sensor responses is to select a certain date range (the ■‘decomposition window”) and use a linear regression to obtain a slope, which should in turn correlate with the biological processes in the soil. In the case of the single-trace glass sensors shown here, slopes taken between 5 and 10 days w ould result in a high error rate, while slopes taken past 14 days would not yield a physically meaningful result as the slopes of sensors Gl and G2 in this range imply open circuits. In contrast, the responses of the 4 segment PCB sensors (indicated as lines Pl, P2, and P3) are clustered together with slower increases in resistance until about 15 days when they began to diverge.

[0101] Example Parallel Decomposition Sensor Topologies. As demonstrated above, parallelizing decomposition sensors with the aid of a custom PCB substrate is effective in reducing the stochastic effects commonly experienced by soil sensors and smoothing out theresponse such that better decomposition windows can be selected for linear regression. A combination of such PCB platforms and stencil printing also allows for rapid iteration of different topologies. Here, we explore various decomposition sensor topologies by varying segment count, segment width, segment length, and carbon loading, and observing the effects on decomposition behavior in laboratory incubated soil. FIG. 12 illustrates example trace dimensions. Additional custom PCBs were fabricated with capability for 3, 5, and 7 parallel printed traces. PCBs for a single traces were also fabricated for characterizing various trace geometries and to act as control groups in experiments. Plots of initial resistances are be found in FIG. 13A-13D. All the example PCB sensors in the study were tested in fully saturated sifted potting soil incubated at 30 °C.

[0102] FIGS. 9A, 9C, and 9E, show the normalized sensor response curves for different topologies printed from a 3: 1 carbon to PHBV ratio by mass ink. FIGS. 9B, 9D, and 9F show the slopes resulting from calculating a linear regression on the 12-20 day portions of the lines from their corresponding response plots with standard deviation errors bars included. As discussed previously, sensor-to-sensor errors could have resulted from slightly varied soil conditions in each container, the interface that develops between the soil and each sensor, and manufacturing defects.

[0103] FIGS. 9A and 9B show the effect of increasing the number of segments, with segment width held constant at 1mm and length held at 17 mm between pads. Simply increasing the number of segments from 1 to 3 drastically flattens the response curve, with a further decrease in response for 5 and 7 segments. Increasing segment count also tends to decrease error between replicates. The slopes for 1 segment sensors were intentionally omitted from FIG. 8D since they reached open circuit resistance before 12 days.

[0104] FIGS. 9C and 9D show the responses resulting from increasing segment length by increasing its tortuosity between pads, with segment count held at 3 and segment width held at 1mm. This change yields a similar trend, with longer segment sensors showing a flatter response curve. However, while increasing length from 17 to 38 mm does decrease sensor-to-sensor error, further increasing segment length to 55 mm does not show a significant decrease in error.

[0105] FIGS. 9E and 9F show the effect of increasing segment width in the region between pads, with segment count held at 3 and segment length between pads held at 17 mm. Here, 2 mm wide sensors show a flatter response curve than sensors with 1 mm wide segments. However, there is an increase in the average slope in sensors with 3 mm wide segments. One explanation for this observation is that pad portion of the prints were 3 mm wide in all cases,therefore the 3 mm wide segments were rectangular, rather than the dog-bone shape of 1 and 2 mm segments, inherently differentiating them. Future studies with wider segments could elucidate a trend that might suggest that a local minimum could exist with 2 mm wide traces.

[0106] Effect of reduced carbon loading. Increasing segment count increases initial sensor conductivity, which could be detrimental to efforts towards low-cost and low-energy data acquisition, as it imputes a higher current drain during sensor readings (cite?) Typically, reducing particle loading in composites increases interparticle distance which in turn leads to a decrease in initial conductivity' (cite all the articles). Decreasing the loading of conductive particles is also another method of tuning the desired decomposition sensor response, as it increases the rate of degradation and resulting rate of increase in resistance (Tao. Tsang, et al.. 2019). Reducing the loading of C:PHBV from 3: 1 to 5:2 and 2: 1 increases the sensor response as show n in FIG. 9G. FIG. 9H, showing the slopes of the sensor response curves taken from 7-15 days, shows that reducing carbon loading can re-claim some of the sensitivity lost to parallelizing sensors.

[0107] Exploring parallel decomposition sensor topologies. As demonstrated above, parallelizing decomposition sensors with the aid of a custom PCB substrate is effective in reducing the stochastic effects commonly experienced by soil sensors and smoothing out the response such that better regions can be selected for linear regression. A combination of such PCB platforms and stencil printing can also allow for rapid iteration of different topologies. Here, the study' explores various decomposition sensor topologies by varying segment count, segment width, segment length, and carbon loading and observing the effects on decomposition behavior in laboratory -incubated soil. Additional custom PCBs were fabricated with the capability for 3, 5, and 7 parallel printed traces. PCBs for a single traces were also fabricated to characterize various trace geometries and to act as control groups in experiments. All instantiations of PCB sensors were tested in fully saturated sifted potting soil incubated at 30 °C.

[0108] Effect of reduced carbon loading. Increasing segment count increases initial sensor conductivity, which could be detrimental to efforts towards low-cost and low-energy data acquisition, as it imputes a higher current drain during sensor readings. Typically, reducing particle loading in composites increases interparticle distance which in turn leads to a decrease in initial conductivity7. Decreasing the loading of conductive particles is also another method of tuning the desired decomposition sensor response, as it increases the rate of degradation and resulting rate of increase in resistance (Tao, Tsang, et al., 2019). As expected, reducing theloading of C:PHBV from 3: 1 to 5:2 and 2: 1 increases the sensor response as shown in FIGS. 9G and 9H.

[0109] Four segment PCB sensors in agricultural fields. In order to determine fieldreadiness of parallel PCB sensors, and to further verify their efficacy against single trace sensors, sets of 4 segment PCB sensors and single trace glass sensors similar to the ones described in herein were deployed in two separate potato fields in the Upper Midwest region of the United States (see FIG. 10A for an illustration of PCB sensor burial and verification with a multimeter that burial did not damage the sensor). Potato farms are suitable candidates for microbial activity' sensing as potatoes are prone to numerous soil-bome diseases (Fiers, Edel-Hermann, et al.. 2012). In addition, although no-till and reduced till practices are being explored and adopted, the high soil disturbance required for potato production leads to erosion, water quality issues, and release of soil carbon to the atmosphere (Djaman, et al., 2022). Custom-built gaseous CO2 sensors were also deployed in each field to monitor soil microbial respiration. Treatments and conditions for both fields, Field B and Field A were similar except that Field A was fumigated before planting.

[0110] FIG. 10B shows the smoothed average response of both glass sensors and PCB sensors. Glass sensors experienced a rapid failure to open circuit within 20 days, with this failure occurring more rapidly in Field B than Field A. As expected, parallel PCB sensors showed a more muted and interpretable response with sensors from both fields showing a relatively flat response. Conducting a linear regression from 5-23 days on the unsmoothed PCB sensor readings in Field B yielded a slope of 0.001 ± 0.002 days"1while sensors in Field A yielded a slope of 0.002 ± 0.002 days"1.

[0111] Approximately 18 days into the experiment, sensors in Field B showed an increase in slope, deviating from sensors in Field A. Linear regression was conducted on the unsmoothed PCB sensor readings from 18-27 days, and the sensors in Field B yield an average slope of 0.022 ± 0.013 days’1and those in Field A yield an average slope of 0.000 ± 0.002 days’1. Interestingly, at around 18 days after deployment, the concentration readings from the CO2 sensor in Field B reached the maximum sensor reading value of 10,000 ppm, implying an increase in microbial respiration from this point until the end of the experiment as shown in FIG. 10C. The CO2 sensor in Field A did not show such a response. A correlation between sensor response and CO2 efflux has been previously shown in laboratory experiments with single trace PHBV decomposition sensors (Atreya, 2023), and the correlation between CO2 flux and CO2 concentration is generally positive, although other factors such as soil porosity may change that correlation

[0112] The Leaf Area Index (LAI) of Field B at the time of increased decomposition sensor slope and increased CO2 reading was about 44% of the maximum value seen in the growing season while the LAI of Field A was about 84%, as show n in FIG. 11 and Table 1. This implies a significantly greater amount of leaf litter in Field B from senescing potato vines, which typically occurs earlier in non-fumigated fields. Leaf litter is typically responsible for a significant increase in production microbial extracellular enzymes in soil (Tian & Shi, 2014) (Dorbunsh, 2007). With single trace PHBV sensors in controlled laboratory environments, Atreya (2023) previously proposed that such sensors could possibly detect a changepoint, specifically an increase, in microbial activity. In the field study presented here, the slower and smoother response of parallel configuration PCB sensors, as compared with single trace glass sensors, allowed for detection of increases in microbial activity occurring later in the growing period.

[0113] Table 1: Enhanced Vegetation Index (EVI) and Leaf Area Index (LAI) for Fields 1 and 2.

[0114] As Field B was not fumigated before planting, the soil microbiome was likely fully present and effective at decomposing the PHBV binder. Prior research has shown that, in many cases, soil microbial communities do recover soon after fumigation (Sederholm, et al., 2018). However, the effect of fumigation treatments on the soil microbiome is likely related to long term management practices (Li, et al., 2022). While fumigation may not affect soil microbial richness, it has been shown to reduce soil bacterial diversity (Li, et al. 2022), and soil microbial diversity has been shown to be directly correlated to the production of poly (hy droxy butyrates) PHB depoly merase enzy me (Dey & Tribedi, 2018), w hich is responsible for the degradation of PHBV (Atreya, 2023)(Roohi, et al.. 2018)(Altaee, et al. 2016). Future studies, where sensors are kept in fumigated and non-fumigated fields, with plant senescing occurring in both fields, could elucidate the effect of fumigation on the rate of decomposition of leaf litter.

[0115] Example !

[0116] Implementations of the present disclosure can improve measurement of soil decomposition rates at a higher temporal resolution than existing indirect methods. An example implementation of the present disclosure was designed and tested as a non-destructive method of assessing microbial activity, including tracking decomposition of soil organic matter in real time. The example implementation included a non-destructive way of monitoring microbially-derived decomposition through a series of environmental perturbations, showing both stress and recovery. The results herein can be used to show the dynamic nature of soil microbial processes that would otherwise have gone unobserved, such as the continued decomposition activity of the flooded wheat.

[0117] The example implementation addresses deficiencies with existing measurements of soil decomposition. Many indicators vary spatially and temporally, so techniques based on sampling and lab analysis have limited temporal resolution (e.g., the time-point when the samples were taken). Additionally, measurements of enzymatic activity, soil nutrient concentrations and soil texture usually need to be taken through destructive soil collection, which causes problems for repeated sampling. Implementations of the present disclosure overcome these deficiencies through methods that are both non-destructive and have much higher temporal resolution through continuous monitoring / sampling. Additionally, the simple structure of the example implementation makes it suitable for large scale deployment, dramatically increasing spatial resolution.

[0118] The example implementation includes a decomposition sensor based on low-cost printed electronics. The active sensing surface comprises a biodegradable polymer with suspended carbon particles, which transduces physical changes in the polymer into a resistance signal. As microbial activity degrades the polymer, its swelling behavior changes in the presence of soil moisture, driving the carbon particles apart and increasing the resistance over time. These sensors provide a method by which the degradation activity of microbial communities in soil can be more directly measured. These sensors are extremely low cost compared with CO2 chamber systems, require no moving parts to control gas flux or accumulation, and can provide continuous in-situ data at high frequency (e.g., seconds or minutes per sample) and in high spatial resolution (e.g., mm scale).

[0119] Methods

[0120] Experimental Setup. Intact mesocosms were collected from two sites near Penrith, north-west England. The first was planted wi th a monoculture of Winter Wheat Triticumaestivum L ). and the soil type was a sandy loam (54°45’17"H 2" 47'28" W). The second was a species-rich grassland (54f'43'23"N, 54O43’23"N)., which had not been subjected to any chemical inputs in living memory and, given the plant biodiversity, have probably have never received fertilizer inputs The species mix included Carex flacca, Plantago lanceolata, Lolium perenne, Lotus corniculatus and Anthoxanthum odoratum. The study collected the mesocosms by inserting sections of pipe into the soil and removing the intact core, to reduce disturbance to the soil. The sections were 15 cm deep with a 10 cm diameter. The study collected 18 mesocosms from each site on 26th March and a further 18 on 10th April 2024, and set them up in a polytunnel at the Hazelrigg Research Station at Lancaster University. Each pot was assigned a random number, and one of three treatments: drought, flood, or control. The final experimental design was 2 vegetation types x 3 climate treatments x 6 replicates = 36 pots. The study began the climate treatment on 11th April, and it continued for 32 days, with recovery' beginning on the 13th May. For the flood treatment, mesocosms were placed in a tank with rainwater collected from the site, maintained level with the top of the soil throughout the study. For the control, watering with rainwater was administered three times a week, as needed. The drought was left to dry down through the treatment phase. When the treatments were alleviated, the flooded mesocosms were allowed to drain, and the drought treatment received one liter of fresh rainwater per pot. For the recovery period, all pots received 500ml twice per week and were allowed to drain freely. Mesocosms were harvested on the 18th June after 36 days of recovery.

[0121] Decomposition Sensors. Sensors were fabricated through a hybrid printing process comprising a 3D-printed deployment carriage, a printed circuit board (PCB), and stencil-printed ink traces composed of a mixture of Poly(3-hydroxybutyrate-co-3-hydroxy valerate) (PHBV) and carbon flake. The experimental setup and specific geometries of the sensors deployed are shown in FIG. 14. FIG. 14 illustrates six replicates of six soil type and treatment combinations that were instrumented with decomposition sensors comprising a biodegradable ink printed on a printed circuit board substrate. The sensing surface is assembled within a 3D-printed deployment carriage which facilitates installation in soil and protects electrical connections with an epoxy dam. Suspended carbon flake makes the ink conductive, and as the PHBV matrix is degraded by soil microbes, the conductivity of the ink changes and can be read as a resistance change by low- cost electronics.

[0122] The ink was printed directly onto exposed pads on the PCB in a set of three conductive strips, using transfer tape and a razor blade. Each strip was printed in a "dogbone” geometry with a primary trace width of 1 mm and a length of 25mm. All conductive strips werelaid out in parallel, a configuration that controls for stochastic variability or mechanical damage impacting a single trace. Following ink printing, signal conductors were soldered to the PCB before installation in the deployment carriage. The carriage was 3D-printed on a consumer FDM machine using a non-died PLA filament. The carriage serves to protect the PCB and ink traces from mechanical damage during installation and provides a potting dam to encapsulate electrical connections. A weatherizing adhesive was poured into the potting dam and allowed to cure for at least 48 hours. Between fabrication and deployment, the ink traces were protected using a 3D- printed travel sheath.

[0123] Sensors were deployed in each sample pot using a 3D-printed nylon installation spade in the case of all treatments except the drought pots. The installation spade creates a pre-formed cavity into which the sensor is press-fit to encourage good surface contact with the surrounding soil. In the case of the drought pots, the soil was too dry for this installation method. Instead, soil was manually excavated to form a cavity large enough for the sensor to be installed, then packed back down around the sensor. During each sensor’s installation, its resistance was monitored during insertion to ensure that the traces were not mechanically damaged prior to the experiment.

[0124] Sensor resistances were collected using custom data loggers which use the voltagedivider principle to calculate the resistances of sensors installed across 6 channels per logger.Resistance readings were gathered every 30 minutes and logged to an SD card. In this experiment, one channel from each logger was used to monitor each treatment to avoid any systematic error introduced by logger electronics and to prevent data loss if a logger were to fail mid-experiment.

[0125] Soil Respiration and Covariate . Soil moisture and temperature were measured twice per week throughout the treatment and recovery time using an HH2 Moisture Meter with attached WET-2 sensor. Soil respiration was measured weekly using an EGM-5 infra-red gas analyser (IRGA) with attached SRC-2 soil respiration sensor. The study modified the chamber using a piece of pipe 15 cm long to extend the chamber's height and prevent crushing of vegetation. Data were collected in the form of mg CO2 m-2 hr-1.

[0126] Plant Analyses. Aboveground biomass was collected by cutting all vegetation at the soil surface in each mesocosm. The biomass was placed in paper bags and dried at 105 C for 48 hours before accurately weighing. The root biomass was carefully shaken free of the soil and washed to remove all stones and soil particles. Wheat roots were stored fresh at 5 C before excision of small cross sections for imaging. The Species Rich biomass and remaining wheat biomass was then dried and weighed as for the aboveground portion.

[0127] Soil Analyses. The soil was collected and sieved through a 2 mm sieve to remove stones and other debris, then stored at 5 C until analysis. Gravimetric soil moisture was measured by drying 5g of fresh soil at 105 C for 24 hours and calculating the proportion of water lost. For the microbial biomass carbon and nitrogen, we carried out the chloroform-fumigation method. Briefly. 5 g of soil from each mesocosm was weighed into a glass vial, and 5 g into a 50 ml Falcon tube. The glass vials were added to a desiccator with a beaker containing 50 ml of chloroform, and a vacuum was applied. The sealed desiccator was left for 24 hours to lyse all microbial cells in the soil samples. At the end of this period, the soil was added to 50 ml Falcon tubes and both fumigated and unfumigated samples were extracted for carbon and nitrogen using 25 ml 0.5 M K2SO4 per sample. These were shaken thoroughly on an orbital shaker for 30 minutes at 150 rpm. Samples were then fdtered through Whatman 42 filter papers and analyzed for microbial biomass carbon (MBC) using a Shimadzu TOC analyzer, and for microbial biomass nitrogen using an autoanalyzer. The final values were calculated by adjusting for soil moisture, then subtracting the unfumigated from the fumigated portion, and finally adjusting by KEC factor for carbon, and KEN for nitrogen, to account for extraction efficiency (Joergensen,1996). The dried fraction of soil was ground in a ball mill and analyzed for total carbon and nitrogen using an elemental combustion analyzer.

[0128] Soil extracellular enzymes were assayed using the methods of Jackson et al. (2013), with modifications outlined in Broadbent et al. (2022), which added artificial p-nitrophenyl (pNP) linked substrates to induce a color reaction through p-nitrophenyl production. The enzymes assayed using this approach were cellobiohydrolase (CBH), (3-glucosidase (GLC), N- acetylglucosaminidase (NAG). [3-xylosidase (XYL) and acid phosphatase (PHO). CBH and GLC are key enzymes for decomposing cellulose into glucose for microbial uptake. NAG is a step in chitin decomposition, making carbon and nitrogen available for microbes. XYL is part of the decomposition pathway for lignocelluose, and PHO breaks down organic phosphate forms into inorganic forms. The study carried out further assays to measure phenol oxidase (POX) and peroxidase (PER), which are instrumental in breaking down complex polyphenols such as lignin and tannins.

[0129] Statistical Analysis. To remove spurious readings, a simple filter was applied to remove any resistance readings above lOkQ prior to any other data processing. To account for differing initial sensor resistances due to the manufacturing process, all sensor signals were normalized: an initial resistance (Ro) was calculated on a per-sensor basis by taking an average of readings within an initial-resistance time window following initial settling in the soil. A soilsetling period began after installation on April 27th at midnight. Next, Ro was calculated as the sensor-specific resistance between April 28th at midnight and 12 hours later. Each following instantaneous resistance reading was divided by this initial resistance, yielding a normalized resistance (Rnorm) which we took to be the primary decomposition sensor signal, as we have in previous research (Atreya 2022, 2023).

[0130] Most data analysis and visualizations are represented as treatment-specific Rnorm averages taken at a 24-hour frequency to remove diurnal effects. Sensor standard deviations within each treatment type are show n using error clouds to represent individual sensor variability within each treatment. For all analyses regarding treatment windows, the Treatment Window was taken to be all readings between May 1st and May 13th. A 3 -day lag was introduced between physical intervention and the Recovery Window, which was taken to be all readings between May 16th and June 17th.

[0131] Results

[0132] Decomposition sensor response to floods and droughts. Resistance, as a proxy for microbial activity, showed contrasts according to catering treatments over time, as shown in FIGS. 15A and 15B. FIGS. 15A-15C shows that over the course of the experiment, sensor responses w ere characterized in two decomposition windows: while treatment effects w ere applied, and during recovery. FIG. 15A shows normalized resistance means and standard deviations are shown across the sensor fleet, where visual examination reveals dramatic sensor signal changes following the removal of drought and flooding conditions. FIG. 15B shows an analysis of slope distributions within the treatment and recovery windows. FIG. 15C shows a strong response between window s for most treatments.

[0133] In the control treatments, where watering was consistent throughout, resistance did not change between the treatment and recovery periods. This was true of both the species rich and winter wheat, although species rich pots became more variable over time as shown in FIG. 15C. As expected, sensor decomposition increased sharply with the alleviation of the drought treatment through reweting in both the Species Rich and Winter Wheat cores FIG. 15 A and FIG. 15B) with the rate of increase in normalized resistance significantly higher as shown in FIG. 15C in both treatments. When the flood was removed, but allowing the plots to drain, as expected the recovery7in the Species Rich cores was clear, but at a significantly slower rate as shown in FIG. 15C when compared to the recovery from drought shown in FIGS. 15A and 15B). The flooded Winter Wheat cores behaved differently. Here the flooding did not slow the rates of decomposition as it had in the Species Rich cores, with no significant difference found betweenthe treatment and recovery window. In all. apart from the wheat control treatment, variation between replicates increased after the recover}' period started, reaching a maximum after 30 days.

[0134] Soil Characteristics. Comparing the Winter Wheat and the Species Rich soils, pH and activity of the carbon degrading enzy mes CBH and GLC were all significantly higher in the Winter Wheat Soil, and soil carbon, total nitrogen, POX, BGB were significantly lower. Few significant effects of the watering treatments were observed between the measured soil parameters, indicating successful recovery', with the exception of AGB, where in both Species Rich and Winter Wheat soils, aboveground biomass was lower in drought soils than control as shown in FIG. 16. Flooding caused a reduction in AGB in winter wheat but not species rich plants. FIG. 1 shows laboratory' measurements taken at the end of the experiment to characterize differences between the two soil t pes and to examine the recovery' of pots which had been subjected to flood or drought conditions.

[0135] Environmental parameters. Soil volumetric moisture content FIG. 17A showed clear differences during the treatment period with the treatments converging during the recovery period. The measured CO2 flux shown in FIG. 17C surprisingly showed no response in the recovery' period, with respiration in each of the treatments remaining similar throughout the experiment.

[0136] FIG. 18 shows individual sensor signals over the course of the experiment, with the Treatment to Control changepoint indicated by a dashed line. No relationship was found between the measured CO2 flux and the decomposition sensor slope as shown in FIG. 19. No clear signal is visible between Treatment and Control conditions when evaluated purely on CO2 accumulation slope or integral as shown in FIG. 20. Soil temperatures shown in FIG. 17B were on average higher during the drought treatment (average 29.5°C) than during the flooding treatment (average 24.3°C), 'hich follows the expected trend compared to the control treatment (27.3°C on average). Average air temperature captured by the loggers shown in FIG. 17D showed a strong diurnal pattern with temperatures reaching a daily mean maximum of 37.3°C and a daily mean minimum of 10.7°C. Air temperature was collected by on-board sensors in the data loggers, where soil moisture and temperature were discrete measurements. CO2 efflux was collected on particular days using a collar fitted around each pot.

[0137] Discussion

[0138] The example implementations of sensors embedded in different types of soils, under varying moisture stresses, can offer detailed information on soil decomposition rates in real time.The data captures nuanced information on changes in microbial conditions in the soil in a nondestructive manner. Gathering data destructively at the end of a period of stress and recovery' leaves knowledge gaps on the exact impacts of the stress on microbially-driven activity' through the stress. As shown here, a recovery' period can obfuscate the overall decomposition rate of soil organic matter by showing apparently similar values in one snapshot of time.

[0139] The study included deploying these sensors in contrasting soil and vegetation settings which have been exposed to either a period of drought or waterlogging, to demonstrate in real time how the decomposition rate changes following the removal of that stress.

[0140] The example soil decomposition sensors were able to respond with a rapid and strong positive shift in resistance, indicating more decomposition of the sensor, following alleviation of environmental stress, with both the species rich and winter wheat treatments showing a similar rate of sensor decomposition throughout the experiment. These responses follow' the expected pattern as it is well known that microbial activity' sharply increases when osmotic stress is removed (Birch, 1958). The main reasons for this are thought to be associated with the release of metabolites due to cell lysis upon rewetting of the soil, the death of microbes which then provide and easily decomposable food sources for surviving organisms and the release to protected organic matter (Singh et al, 2023).

[0141] In the case of flooding the picture is more complicated. It was expected that flooding would cause significant impacts on soil functioning (Tate. 1979; Huang et al.. 2023; Sanchez- Rodriguez, 2019). We expected sensor decomposition rates in the flooded soils to be lower than the control, due to anaerobic conditions being established at the sensor, thus slowing or stopping the decomposition activity of aerobic and facultative organisms and that there would steady recovery following the pots being allowed to drain under gravity. For the species rich treatment, as expected, sensor decomposition is reduced by the flood, with slow rate of recovery' in the sensor decomposition rate once the treatment is alleviated. This may be due to the slower rate of alleviation of flooding compared to the alleviation of the drought, since it takes longer for water to drain from a soil than it takes for a soil to rewet under experimental conditions. However, in the case of winter wheat the flooding had no negative impact on sensor decomposition rates, in fact decomposition rates w ere higher throughout the experiment in this treatment combination than in any other treatment including the controls. This may be due to the actively growing wheat plants that maintained an oxygenated rhizosphere which promoted biological activity and thus, sensor decomposition. Support for this hypothesis is found in the aerenchyma that found inthe thin sections of the wheat roots. Aerenchyma forms in some species of plants in response to low oxygen availability, allowing the plant to transport oxygen to the root surface (REF).

[0142] The soil analysis conducted at the end of the experiments showed differences between the species rich and the winter wheat cores, but unlike the decomposition sensors they were not able to show differences between the drought, flood and control treatments. The behavior of the soil chemical parameters and enzymes largely conformed to expectations. The pH was higher in the winter wheat soil as the soil will have been limed to maintain a pH suitable for arable cropping. The pH of 5.5 for the Species Rich soil is close to the mean of XX for UK grasslands (REF), the higher soil carbon and nitrogen contents in the grassland soil are also inline with expectations. The regular disturbance of the arable soil increased aeration and availability of organic carbon for decomposition. The higher nitrogen content in the grassland soil is perhaps a little surprising, given that the arable soil had regularly received inorganic and organic sources of fertilizer, however offtake by the crop and through leaching is likely to be higher in the arable soil than in the species rich grassland which is essentially a closed system. pH, CBH and GLC were all significantly higher (padj <0.01), and soil carbon, total nitrogen, POX, BGB were significantly lower (padj < 0.01) in the Winter Wheat Soil.

[0143] Conclusion

[0144] It should be appreciated that the logical operations described herein with respect to the various figures may be implemented (1) as a sequence of computer implemented acts or program modules (i. e. , software) running on a computing device (e.g., the computing device described in FIG. 6), (2) as interconnected machine logic circuits or circuit modules (i.e., hardware) within the computing device and / or (3) a combination of software and hardware of the computing device. Thus, the logical operations discussed herein are not limited to any specific combination of hardware and software. The implementation is a matter of choice dependent on the performance and other requirements of the computing device. Accordingly, the logical operations described herein are referred to variously as operations, structural devices, acts, or modules. These operations, structural devices, acts and modules may be implemented in software, in firmware, in special purpose digital logic, and any combination thereof. It should also be appreciated that more or fewer operations may be performed than shown in the figures and described herein. These operations may also be performed in a different order than those described herein.

[0145] Referring to FIG. 6, an example computing device 600 upon which the methods described herein may be implemented is illustrated. It should be understood that the examplecomputing device 600 is only one example of a suitable computing environment upon which the methods described herein may be implemented. Optionally, the computing device 600 can be a handheld or laptop devices, multiprocessor systems, microprocessor-based systems, network personal computers (PCs), minicomputers, embedded systems, and / or distributed computing environments including a plurality of any of the above systems or devices. Distributed computing environments enable remote computing devices, which are connected to a communication network or other data transmission medium, to perform various tasks. In the distributed computing environment, the program modules, applications, and other data may be stored on local and / or remote computer storage media.

[0146] In its most basic configuration, computing device 600 typically includes at least one processing unit 606 and system memory 604. Depending on the exact configuration and type of computing device, system memon 604 may be volatile (such as random access memory (RAM)), non-volatile (such as read-only memory (ROM), flash memory, etc.), or some combination of the two. This most basic configuration is illustrated in FIG. 6 by dashed line 602. The processing unit 606 may be a standard programmable processor that performs arithmetic and logic operations necessary for operation of the computing device 600. The computing device 600 may also include a bus or other communication mechanism for communicating information among various components of the computing device 600.

[0147] Computing device 600 may have additional features / functionality. For example, computing device 600 may include additional storage such as removable storage 608 and nonremovable storage 610 including, but not limited to, magnetic or optical disks or tapes. Computing device 600 may also contain network connect on(s) 616 that allow the device to communicate with other devices. Computing device 600 may also have input device(s) 614 such as a keyboard, mouse, touch screen, etc. Output device(s) 612 such as a display, speakers, printer, etc. may also be included. The additional devices may be connected to the bus in order to facilitate communication of data among the components of the computing device 600. All these devices are well known in the art and need not be discussed at length here.

[0148] The processing unit 606 may be configured to execute program code encoded in tangible, computer-readable media. Tangible, computer-readable media refers to any media that is capable of providing data that causes the computing device 600 (i.e., a machine) to operate in a particular fashion. Various computer-readable media may be utilized to provide instructions to the processing unit 606 for execution. Example tangible, computer-readable media may include, but is not limited to, volatile media, non-volatile media, removable media and non-removablemedia implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. System memory 604, removable storage 608, and non-removable storage 610 are all examples of tangible, computer storage media. Example tangible, computer-readable recording media include, but are not limited to, an integrated circuit (e.g.. field-programmable gate array or application-specific IC), a hard disk, an optical disk, a magneto-optical disk, a floppy disk, a magnetic tape, a holographic storage medium, a solid-state device, RAM, ROM, electrically erasable program read-only memory' (EEPROM), flash memory' or other memory' technology', or magnetic storage devices.

[0149] In an example implementation, the processing unit 606 may execute program code stored in the system memory 604. For example, the bus may carry’ data to the system memory 604, from which the processing unit 606 receives and executes instructions. The data received by the system memory 604 may optionally be stored on the removable storage 608 or the nonremovable storage 610 before or after execution by the processing unit 606.

[0150] It should be understood that the various techniques described herein may be implemented in connection with hardware or software or, where appropriate, with a combination thereof. Thus, the methods and apparatuses of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, hard drives, or any other machine-readable storage medium where, when the program code is loaded into and executed by a machine, such as a computing device, the machine becomes an apparatus for practicing the presently disclosed subject matter. In the case of program code execution on programmable computers, the computing device generally includes a processor, a storage medium readable by the processor (including volatile and nonvolatile memory and / or storage elements), at least one input device, and at least one output device. One or more programs may implement or utilize the processes described in connection with the presently disclosed subject matter, e.g., through the use of an application programming interface (API), reusable controls, or the like. Such programs may be implemented in a high level procedural or object-oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language and it may be combined with hardware implementations.

[0151] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above.Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

WHAT IS CLAIMED:1 . A soil sensor comprising: a circuit substrate; and a plurality of soil sensing elements formed on the circuit substrate, including a first sensing element and a second sensing element arranged in a parallel circuit configuration, wherein the soil sensor is configured to measure a combined response of the plurality of soil sensing elements.

2. The soil sensor of claim 1, wherein at least one of the soil sensing elements comprises a decomposable conductive ink, and wherein the plurality of soil sensing elements comprises the decomposable conductive ink.

3. The soil sensor of claim 1 or 2, wherein the plurality of soil sensing elements are arranged in a grid.

4. The soil sensor of claim 3, wherein the grid comprises at least two rows of adjacent soil sensing elements of the plurality of soil sensing elements.

5. The soil sensor of any one of claims 1-4 further comprising a computing device operably coupled to the plurality' of soil sensing elements, wherein the computing device is configured to measure a property' of the soil using the soil sensor.

6. The soil sensor of claim 5, wherein the computing device comprises persistent memory storage.

7. The soil sensor of claim 5 or claim 6, wherein the computing device comprises a wireless network interface.

8. The soil sensor of any of claims 5-7, wherein the parallel circuit configuration is coupled to a multiplexer, and wherein the multiplexer is configured to measure the combined response of the plurality of soil sensing elements by determining that at least one soil sensing element is an outlier, and rejecting measurements of the outlier.

9. The soil sensor of claim 8, wherein the multiplexer is configured to record a plurality of separate responses corresponding to the plurality of soil sensing elements.

10. The soil sensor of claim 9, wherein the computing device is configured to estimate a distribution of microbial activity based on the plurality of separate responses corresponding to the plurality of soil sensing elements.

11. A method comprising: positioning a plurality of soil sensors according to any one of claims 1-10 in a field, including a first soil sensor at a first field location and a second soil sensor at a second field location; and measuring, at periodic interval, the response of the soil sensing elements of the plurality of soil sensors; and determining, via the response, a microbial activity of the field.

12. The method of claim 11, wherein the microbial activity is determined as a slope of the response versus time.

13. The method of claim 12, wherein the slope is positive when the microbial activityincreases, and negative when the microbial activity decreases.

14. A system comprising: a plurality of soil sensors, wherein each soil sensor of the plurality of soil sensors comprises: a circuit substrate, a plurality of soil sensing elements formed on the circuit substrate, including a first sensing element and a second sensing element arranged in a parallel circuit configuration, wherein the soil sensor is configured to measure a combined response of the plurality of soil sensing elements; and a computing device operably coupled to the plurality of soil sensing elements, wherein the computing device is configured to measure a property of the soil using the soil sensor.a controller operably coupled to the computing devices of the plurality of soil sensors by a network, wherein the controller is configured to: measure, at a periodic interval, a plurality of resistance values corresponding to the plurality' of soil sensing elements; determine, a soil property for a field the plurality of soil sensors are deployed in.

15. The system of claim 14, wherein the soil property is microbial activity.

16. The system of claim 14 or claim 15, wherein the network is a wireless network, and wherein the computing device of each of the plurality of soil sensors comprises a wireless network interface.

17. The system of any of claims 14-16, wherein the parallel circuit configuration of each of the plurality of soil sensors is coupled to a multiplexer of each soil sensor, and wherein the multiplexer of each soil sensor is configured to separately measure the responses of soil sensing elements.

18. The system of claim 17, wherein the multiplexer is configured to record a plurality of separate responses corresponding to the plurality of soil sensing elements.

19. The system of claim 18, wherein the computing device is configured to transmit the plurality of responses to the controller.

20. The system of claim 19, wherein the controller is configured to determine a distribution of microbial activity based on the plurality of separate responses corresponding to the plurality of soil sensing elements.

Images