Progress 09/01/23 to 08/31/24

Outputs
Target Audience:We identified two broad stakeholder groups that will benefit from the developed technology. The first is direct users, such as farmers and organizations involved in land management. The second group of stakeholders includes the researchers working on problems dealing with soil gas emissions as applied to climate change, agriculture land management, and efficient land management. No activities occurred targeting the direct user of the technology. The CSM PI, UCB PI, and a UCB graduate student contributed and participated in the Signals in the Soil (SitS) Hybrid Workshop in Washington, DC, held in June 2024. A letter of interest (LOI) has been submitted to NIFA to request funding for a workshop that researchers and stakeholders, including farmers and farming groups, will attend. Changes/Problems:A no-cost extension was approved until August 2025. We plan to use the leftover funds for a limited workshop. If an additional finds request is granted, a more expanded workshop will be conducted at UCB. A LOI will be submitted to seek funding for the workshop. What opportunities for training and professional development has the project provided?The project supported a full-time postdoctoral research associate from March 2020 to February 2023. The postdoc's background was in soil science, and the project allowed her to develop skills in laboratory testing and modeling. The postdoc also worked with a Ph.D. student at UCB (supported by NSF) during the optical fiber testing in the intermediate-scale test systems. The UCB student also visited Rothamsted Research to assist in the fieldwork. How have the results been disseminated to communities of interest?Results and findings were presented at national and international conferences attended by multidisciplinary researchers interested in porous media, soil physics, and sensor technologies. The CSM PI, UCB PI, and a UCB graduate student contributed and participated at the Signals in the Soil (SitS) Hybrid Workshop in Washington DC held in June, 2024. What do you plan to do during the next reporting period to accomplish the goals?Even though this project's funding started on September 1, 2019, the spending and research started in March 2020. A no-cost extension was granted for the project to end in August 2025. Because of an error in the indirect cost calculations, the CSM budget ended up with leftover funding that would be used to conduct a workshop. 1. Complete the final report. A draft report was prepared by the post-doctoral research associate who completed her three-year position in February 2023. The PI will edit and complete this report to prepare the project's final report, which is due in August 2025. The PIs will also work with the UCB collaborators to incorporet the thie reseach into the final report and develop research articles based on the tank data generated at CSM. 2. Planning a workshop The team is planning a workshop on "Sensing for Soil Processes Monitoring for Agricultural and Hydrologic Applications" at the Center for Smart Infrastructure (CSI) located at UCB, of which the UCB PI is the director, and the CSM collaborator has an affiliate. The workshop will bring together scientists working on soil sensing technologies for agricultural and hydrologic applications, post-doctoral researchers, and graduate students interested in soil sensing, sensing networks, and modeling. The objective is to share results and findings from past and ongoing research, including our own, while identifying existing knowledge gaps, challenges, opportunities, and potential future collaborations. The strategies and potential use of testing facilities that could be shared will also discussed. We are requesting additional funding. If this is not successful, we propose to have a meeting of the research team and three invited researchers using the leftover finds at CSM (close to $13K).

Impacts
What was accomplished under these goals? In previous years, a large-scale field test occurred at Rothamsted Research, UK. Multiple DFOS sensors (up to 200m) were buried under the ground where plants grew, accompanied by natural weather variations. Data collection continued during this report period. The fiber optics-based research continued at UBC. The first set of results examined the feasibility of Distributed Fiber Optic Sensors (DFOS) in detecting the early signs of crack generation in soil. The hypothesis was that cracking due to thermal and mechanical effects affects the spatial and temporal distribution of gas emanating from the subsurface. This hypothesis was validated using the test systems at CSM. These tests are complementary cases to illustrate soil cracking behavior under varying saturation levels and dry density conditions. Cracks were intentionally induced through four-point bending. An increase in the bending moment led to an elevation in tensile strain within the pure bending area, ultimately resulting in crack development. These cracks induced localized strain concentrations and the strain profile acquired through the OFDR system demonstrated the capability of DFOS to detect crack propagation from these localized strain concentrations. Multiple layers of fiber optics identified the evolution of crack propagation. The width of crack openings was estimated by integrating the strain profile near the crack, and Digital Image Correlation (DIC) was employed for validation. Observations revealed slippage between the fiber optic sensor and clay, showing the importance of selecting fiber optic cables for effective soil crack detection using DFOS. Two criteria, maximum strain and de-bonding strain, were selected for categorizing the test cases. An experimental chart was generated to evaluate the performance of this specific strain fiber-optic sensor across Kaolinite with varying water content and dry unit weight. The quality of the DFOS can be predicted for future applications based on soil water content and dry density, especially for Kaolin clay. The second research task completed by the UCB was to gain a deeper understanding of the operational performance of DFOS in field scenarios; four test soil boxes were fabricated at Rothamsted Research in the United Kingdom. The experimental setup controlled environmental factors such as temperature and air moisture content, ensuring that the embedded sensors could acquire data from the soil without being affected by irrelevant variables. This experiment was designed for long-term measurements, including plant growth within the tank, to simulate natural soil-cracking conditions in the field. In this context, cracks are not only induced by desiccation but also by the presence of plant roots. The test duration was planned to be at least two months, with the potential for extension if the plants continue to thrive until March 2024. Upon sprouting, precise gas measurement instruments will be employed to analyze CO2 and CH4 emissions resulting from the root system's biochemical reactions. Previously a large-scale field test took place at Rothamsted Research, UK. Multiple DFOS sensors (up to 200m) were buried under the ground where plants grew, accompanied by natural weather variations. A BOTDA analyzer collected a significant amount of field data with a spatial resolution of 1m. Meanwhile, photos were taken in the field, which was used to track the development of soil cracks. However, the performance of the old analyzing method is not ideal for detecting localized strain distribution caused by the cracks. Since the spatial resolution is 1m, the acquired data represents an average of all strain profiles within the resolution. Therefore, the strain concentration was not obvious enough to identify the crack. Another research task aims to utilize a novel analyzing scheme for BOTDA data to develop a new method for analyzing the field data acquired by the BOTDA unit. With the help of the scheme proposed in Ou, R., Luo, L., & Soga, K. (2023), a probabilistic optimization approach based on the spectrum data was designed for the field data acquired. As shown in Figure 7, this approach captures the contribution, or in another sense - probability, of all reasonable strains within one spatial resolution. Then the probability distribution was recovered by optimization to capture the 'second peak' of the spectrum, which is related to the high localized strain concentration induced by the soil crack. BOTDA spectrum data taken in Warren field in 2011 was used to test this approach. Then the result was interpreted by inverse transforming sampling. Figure 8 shows a box chart of the sample, where the box covers 25%-75% of the sample, while the whisker covers 2%-98% of the sample. A straightforward judgment of the cracking location is a rather large variation, leading to a longer box or whisker. Locations that suit this characteristic were picked up and compared with the field photo taken in 2021. The match between the DFOS data and the photo is impressive, showing that this approach is applicable for BOTDA field data analysis. In addition to distrurbance resulting from thermal and mechanically induivced cracking, the soil disturbance from plant rootes were investigated. Whether the fiber-optics can detect plant roots penetrating the layer and distinguish between roots and desiccation-induced cracks constitutes another research question of interest. However, in the 3D tank experiment, the systematic development of the root system remains invisible without a reliable reference. To address this, a series of 2D tank experiments were conducted in the year 2023, serving as a pilot for future experiments. The 2D tank measures 400mm x 400mm x 20mm and is constructed from transparent acrylic, allowing for Digital Image Correlation (DIC) from both the front and back. Three layers of fiber optics were embedded in the soil to monitor crack propagation. One tank was planted with lettuce, while another tank without plants (but with a similar setup) served as a reference (Figure 9). Additionally, a carefully controlled tank filled with kaolin clay was set up to test the feasibility of DIC, as depicted in Figure 10. Strain data from the fiber optics was compared with the DIC strain field. The fiber-optics cable inside the tank planted with lettuce exhibited a higher strain concentration than the reference group. The Distributed Fiber Optic Sensing (DFOS) system successfully captured the propagation of the crack. Furthermore, the water content of the two tanks indicated that the plant tended to retain water during the early-stage desiccation. However, as the plant grew larger and the roots penetrated deeper, the plant absorbed more water compared to the reference group. Consequently, the water content dropped abruptly in the late stage of the experiment. The DFOS results corroborate this observation. In summary, the primary conclusions of this experiment are: a. The fiber optic sensor successfully captured the growth of plant roots and should be capable of identifying root locations. b. Plants may preserve moisture and prevent soil fracturing if the soil is relatively wet. However, in dry soil conditions, transpiration results in drier soil, increasing the likelihood of crack development.

Publications

Progress 09/01/22 to 08/31/23

Outputs
Target Audience:In the proposal submitted, we identified two broad stakeholder groups that will benefit from the developed technology. The first is direct users, such as farmers and organizations involved in land management. The second group of stakeholders includes the researchers working on problems dealing with soil gas emissions as applied to climate change, agriculture land management, and efficient land management. Colorado Scholl of Mines team focused on completing the intermediate-scale experiments. The papers, reports, and documentation will be developed in collaboration to initiate some of the proposed activities targeting the stakeholders. Changes/Problems:In the no-cost extension request, it was stated that we plan to use the porous media/wind tunnel facility that was relocated from CSM to the US Army Corps of Engineers ERDC in Vicksburg, MS. The leftover funds were proposed to be used for the PI's time and travel to Vicksburg to work with ERDC and UCB teams. This plan was based on the expectation that the UCB PhD student, who had worked with CSM collaborators, would lead the fiber optic testing. Soil cracking experiments were completed during the previous reporting periods in the wind tunnel when the facility was at CSM. Another set of experiments integrating fiber optics and gas measurements was completed during this reporting period. We changed the plans by not conducting additional wind tunnel experiments. Instead, the PI and PhD students will visit the Rothamsted Research site in the United Kingdom, where the fiber optic systems have been installed in the field. The UCB collaborators will support the travel and expenses for the field visit of the student. Funds are not available to conduct testing at ERDC. Since a set of wind tunnel experiments have already been completed and COVID restrictions limit field visits to the UK, using the leftover funding to conduct fieldwork as proposed would be more beneficial in meeting the project's stated goals. What opportunities for training and professional development has the project provided?The project supported a full-time post-doctoral research associate from September 2022 to February 2023. The post-doc's background was in soil science. The project allowed her to develop skills in laboratory testing and modeling. The post-doc also worked with a Ph.D. student at UCB (supported by NSF) during the optical fiber testing in the intermediate-scale test systems. How have the results been disseminated to communities of interest?Results and findings were presented at national and international conferences attended by multidisciplinary researchers interested in porous media, soil physics, and sensor technologies. What do you plan to do during the next reporting period to accomplish the goals?Even though this project funding started on September 1, 2019, the spending and the research started on March 12th, 2020 A no-cost extension was granted for the project to end in August 2024. The three-year funding to support the post-doctoral research associate expired in February 2023. The experiments in the 3D tank were completed. A balance of $11,545 will be expended during the next report period, ending with the project's completion on August 31, 2024. Two activities are planned: 1. Complete the final report. A draft report was prepared by the post-doctoral research associate who completed her three-year position in February 2023. The PI will edit and complete this report to prepare the project's final report, which is due in August 2024. The PIs will also work with the UCB collaborators to develop research articles based on the tank data generated at CSM. 2. Visit the field site The work plan included visiting the Rothamsted Research site in the United Kingdom. The fieldwork by the collaborators in the UK involved the installation of the fiber optics soil sensing systems in the test agricultural fields. The UK collaborators who installed these systems visited CSM to work on the 3D test tank, where some gas collection systems were developed and tested. The same researcher visited UCB to participate in testing the fiber optics sensor systems. This field visit aims to work with the UK collaborators to determine the functioning of the laboratory-developed systems in the field. During the trip, the PI will work with the modelers in the UK team to use the data generated in the 3D tank in model testing and validation.

Impacts
What was accomplished under these goals? The experiments conducted using the intermediate scale test systems at CSM focused on the 3-D testing of the integrated sensing using fiber optics for crack detection, soil moisture, and gas migrations. A soil moisture sensing system was deployed in the 3D tank with 40 sensors, to acquire data on soil moisture response to changing water table and rainfall infiltrations. Soil moisture patterns were observed in the sandy soil. The 3-D tank, Experimental Setup for Soil Gas Migration and Soil strain - cracks To mimic a gas leakage system, eight gas injection pipes were located 12.5 cm from the bottom of the tank within the #30 layer. Each pipe is 60 cm long with an internal diameter of 2 cm, with a porous filter acting as a gas injection port. The soil moisture sensors were placed 20 cm apart from each pipe to prevent any of them from influencing or disturbing the gas migration pathway. The methane gas (98%) was injected into the porous media (#30 sand) at a rate of 10 mL/min over a 1-min period. The 3D tank was packed with a silt field layer and medium #30/40 sand setup for the fiber optics strain and temperature measurements. The three chambers were used to measure the mass flow rate of the methane soil gas emissions. Fiber optics were used for a few gas injections to try to find the feasibility of DFOS to detect crack generation in soil. The crack propagation in a solid usually brings a localized strain concentration on the crack tip. With the help of high-resolution OFDR strain sensing, crack generation can be captured based on the strain concentration shown by the DFOS measurements. In addition, soil cracking is governed by the change in effective stress during the tensile process. To analyze the mechanism of soil cracking, measuring negative pore pressure during soil tensile progress is vital to evaluate the effective stress. In this test, we utilize miniature fiber-optic-based pore pressure transducers to measure the pore pressure at the location in front of the soil fracture. To capture the desiccation of the soil near the crack, 4 MPPTs are installed in the center area accompanied by the central notch. The locations of these MPPTs are installed either on the left or right sections, providing convenience for gas collection. Each MPPT has an EC5 moisture sensor as a reference, such that the acquisitions shall verify the soil-water characteristic. the advance of either a wetting or a non-wetting front (Fransje Praagman & Femke Rambags, 2008). It observed that the course-textured layers were relatively low in saturation than the fine-textured layers under steady-state conditions. The capillary tension limits the water's downward movement from finer soil into coarser soil. A coarser soil layer underlying a finer soil layer causes temporary water storage in the upper (finer) soil layer that can ultimately be removed by evaporation, lateral drainage, or percolation into the lower layer (Stormont and Anderson, 1999). The degree of saturation is a key controller for gas migration and invariably saturated porous media since both air permeability and gas diffusivity are strongly moisture-dependent parameters (Sallam et al., 1984; Springer et al., 1998). Previous laboratory-based methane transport studies highlighted the effect of saturation on steady-state methane migration (e.g., changes in the dispersion coefficient; Batterman et al., 1995). However, the saturation effect was apparent in partially saturated layered systems, especially for the bottom layers where high CH4 concentrations were observed compared to dry conditions. This is due to the increased water-induced tortuosity for methane diffusion. The soil methane gas migration was tracked in this study using 14 Figaro metal-oxide methane gas sensors with a high resolution of 1 ppm. In both heterogeneous and homogeneous experiments, gas diffused discontinuously, consisting of incoherent channels, which migrated vertically per diffusion, with some lateral movement as well. The methane continued to migrate per diffusion vertically until it reached the silt field soil at the top tank. Based on methane concentrations, it looks like the gas pattern was narrow near the gas injection point, then expanding vertically and laterally as gas diffused further from the source. In both homogeneous and heterogeneous setups, the gas pattern had a maximum width of 20 cm distance between the S3 and S1 soil gas monitoring locations. The data collected by the Figaro sensors had similar patterns to the previous data collected in 2021, by the infrared methane gas sensors resolution of 100 ppm. Methane concentration profiles methane concentrations presented similar patterns both for heterogeneous and homogeneous system in completely dry sandy soil. The homogeneous system presented higher methane concentrations for partially saturated soil than the heterogeneous system setup. Past experimental and numerical studies have observed similar methane concentration profiles in variable permeability soil systems (e.g., Jiránek, 2010). The relatively low methane concentrations in the heterogenous system might be due to improved mixing and migration of CH4 within the coarse-textured sand layers compared to the fine-textured layers. The average pore size in the coarse-textured (#12/20) sand is almost twice as large as that in the medium-textured (#30/40) sand. This essentially implies the presence of large, well-connected, and less tortuous pore space in the coarse-textured (# 12/20) sand compared to the fine-textured sand, as noted by Chamindu Deepagoda and Elberling, 2015. Consequently, rapid mixing and migration (both in lateral and vertical directions) of CH4 within the coarse-textured layer is expected, causing decreased CH4 concentrations within the layer. In summary, figures 16 - 20 show the soil methane gas migration from source point G. The methane gas concentrations were much higher for the partially saturated soil than the dry conditions. The soil gas concentrations were much higher close to the gas injection port, and lower at those farther locations. Based on several studies, at proximity to the point of leakage, both advective (pressure-driven) and diffusive (concentration-driven) transport occurs, however, the flow becomes more and more diffusion-controlled as methane moves away from the source (Okamoto and Gomi, 2011). In the homogeneous system under dry conditions, diffusion-induced upward methane movement is readily supported by lateral diffusion (due to a high diffusion coefficient under dry conditions). In the heterogeneous systems under the same conditions, both the diffusion-driven upward movement and the diffusion-driven lateral movement are more pronounced (due to the presence of the high permeability coarse-textured layer. In partially saturated sandy soils, the gaseous methane movement is influenced by limited air-filled pore space (science part of the total pore space is occupied by water), simultaneous movement of water vapor in the gaseous phase, and increased (water-induced) gaseous phase tortuosity and discontinuity. Consequently, under partially saturated conditions, the diffusion-controlled lateral movement is more hindered than the diffusion-induced upward movement of methane. There is a dominant effect of saturation over the texture in a predominantly diffusion-controlled gas transport system, as observed in many past studies (e.g., Moldrupet al., 2000; Rolston and Moldrup, 2002). The experimental data is currently being analyzed and will be used in the modeling conducted by the UK collaborators. The PI will present this data to the collaborators during the planned visit in 2024.

Publications

• Type: Journal Articles Status: Published Year Published: 2021 Citation: 1. Xiaoxian Zhang, Andrew S. Gregory, W. Richard Whalley, Kevin Coleman, Andrew L. Neal, Aurelie Bacq-Labreuil, Sacha J. Mooney, John W. Crawford, Kenichi Soga, Tissa H. Illangasekare, 2021. Relationship between soil carbon sequestration and the ability of soil aggregates to transport dissolved oxygen, Geoderma, Volume 403, 2021, 115370, ISSN 0016-7061
• Type: Journal Articles Status: Published Year Published: 2022 Citation: 2. Zhang Xiaoxian, Whalley Peter A., Gregory Andrew S., Whalley W. Richard, Coleman Kevin, Neal Andrew L., Mooney Sacha J., Soga Kenichi, and Illangasekare Tissa H. 2022An overlooked mechanism underlying the attenuated temperature response of soil heterotrophic respiration J. R.Soc. Interface.192022027620220276, http://doi.org/10.1098/rsif.2022.0276

Progress 09/01/21 to 08/31/22

Outputs
Target Audience:We identified two broad stakeholder groups that will benefit from the developed technology. The first is direct users, such as farmers and organizations involved in land management. The second group of stakeholders includes the researchers working on problems dealing with soil gas emissions as applied to climate change, agriculture land management, and efficient land management. The papers, reports, and documentation that were developed during the reporting period will be used to initiate some of the activities that were proposed. Changes/Problems:Colorado School of Mines requested a no-cost extension on the project until August 31, 2023. Even though the funding started in September 2019, we were able to recruit a post-doctoral research associate to start in March 2020. As this was the start of the pandemic, with many safety restrictions, the experimental research was delayed. The collaborators from the University of California could not visit as planned to use the test systems at CSM. The extension will give us the time to complete the testing at CSM and at the wind tunnel facility that was relocated to the US Army Corp of Engineers ERDC in Vicksburg, MS. What opportunities for training and professional development has the project provided?The project supported a full-time post-doctoral research associate. The post-doc's background was in soil science. The project allowed her to develop skills in laboratory testing and modeling. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?Even though this project funding started on September 1, 2019, the spending and the research started on March 12th, 2020. With the original funding, we will be able to continue this research until the end of February 2023 (3 years). From the start of the new report period from September 1, 2022, we will have six months to complete the project. A no-cost extension was requested to have more time to execute more experiments and submit the Final report by the end of August 2023. In the work plan submitted to NIFA at the project initiation, three primary tasks under laboratory testing that will be conducted at CSM were identified: (1) test method development, (2) preliminary proof of experiments, and (3) laboratory testing of sensing systems. The progress up to date has been made on the first and second tasks. In the next reporting period, we will task 2 and 3. Under task 2, we plan to complete the experiments in the 3-D tank using methane as the test gas. Methane was selected as the test gas because of the equipment's unavailability to measure CO2. The feasibility of conducting CO2 experiments with available sensors will be pursued. Low-cost methane Figaro sensors will be used for methane detection, lab calibrations will be conducted. Based on the experiments conducted in the 2-D and 3-D tanks and the new upcoming gas injection experiments, we plan to complete journal articles focusing on modeling the methane gas migration related to the soil moisture content. Different climate conditions (rainfall and high temperatures) will be created onto the 3D tank; we will focus primarily on observing how much methane was emitted from dry, cracked soil and how much methane will be emitted from partially saturated soil. Few experiments will be conducted in the climate-controlled Wind Tunnel facility as well, where more than hundreds of soil moisture sensors, wind speed/direction, temperature and rainfall, and fiber optics (UC Berkeley) will be used, and we will focus on CO2 gas migration.

Impacts
What was accomplished under these goals? Before initiating any methane gas injection experiments, one of the objectives was to evaluate the soil moisture sensors' accuracy in a laboratory setting under different boundary conditions. The soil moisture sensors were calibrated before the experiments for a wide range of soil water content at known temperature conditions expected during the tests. The experiments focused on observing soil moisture patterns and their performance under various imbibition and drainage scenarios. The sensors were able to successfully capture the complex spatial and temporal variations of the soil moisture in the tank. Based on the time-series data, the variations in the specific properties of the sand in the packing led to different saturation levels in the tank. A combined laboratory and field study is in progress to develop and apply a spatially distributed sensing technology to be used in agricultural and other plant environments. A part of this study is to test integrated sensing technology that uses embedded and fiber optics-based sensing and wireless sensor technologies under various soil and climatic conditions. Testing was done in an intermediate-scale three-dimensional porous media tank (4.4m x 2.4 m x 0.45), where the temperature and water mass flux boundary conditions can be controlled. It is our intention to upscale these experiments to a longer test length of ~8 m in a coupled porous media and climate-controlled wind tunnel facility where the climate conditions in the near-surface boundary layer can be controlled. The goals of the tank experiments were to develop test methods and to test the hypotheses that fractures in soils will enhance greenhouse loading from the subsurface generated from geo-biochemical reactions. The optical fiber-based sensing system will be able to capture the soil fractures or fractures zones. The tank was packed with a heterogeneous configuration using five well-characterized uniform silica sands with effective sieve numbers #110, #70, #30, #40/50, #20/30, and #12/20, respectively. Spatial and temporal variations of soil moisture were monitored using embedded soil moisture sensors. A layer of silt with a high clay content was placed on the topsoil surface to create thermal fractures using heat lamps. The silt layer was instrumented with fiber optical cables at two depths to monitor the fracture propagation from the soil surface into the silt layer. Methane was injected at the bottom of the tank using a distributed set of ports and allowed to migrate through the heterogeneous formation. The gas concentrations at the soil surface were measured using three soil-gas flux chambers developed and tested as a part of the research. The gas concentrations were also measured within the soil profile. In addition to the soil moisture sensors within the tank, the soil water tension within the silt was recorded continuously using precision micro tensiometers with the least disturbance to detect moisture variations during crack development from drying. The preliminary findings suggest that the testing methods were successful, and the data will allow testing the hypotheses on whether the fractures affect the gas loading to the atmosphere. The generated data will be used to validate numerical models to further testing of this hypothesis under possible field conditions and climate scenarios. The main conclusion derived from gas injection experiments is related to how soil gas migration pathways develop through the soil profile. Under dry conditions, these patways are completely defined by the gas soil permeabilities. Pathways with high permeabilities will move the gas preferentially and appear at the soil surface. Wet conditions are created either due to precipitation or irrigation water applications or water table fluctuations. In this case, the gas permeability will depend on the water content that dynamically changes with precipitation and drainage. Further research is needed to provide accurate estimates of soil gas emissions related to soil moisture content and soil properties. The experimental data demonstrate the importance of monitoring the soil moisture dynamics to interpret gas concentration measurements within the soil profile and the land-atmospheric interface.

Publications

• Type: Journal Articles Status: Published Year Published: 2022 Citation: Zhang Xiaoxian, Whalley Peter A., Gregory Andrew S., Whalley W. Richard, Coleman Kevin, Neal Andrew L., Mooney Sacha J., Soga Kenichi, and Illangasekare Tissa H. 2022An overlooked mechanism underlying the attenuated temperature response of soil heterotrophic respiration. R. Soc. Interface.192022027620220276, http://doi.org/10.1098/rsif.2022.0276

Progress 09/01/20 to 08/31/21

Outputs
Target Audience:In the original proposal submitted to NSF, under the broad impacts, we identified our target audience as the scientific community and industry. We identified two broad stakeholder groups that will benefit from the developed technology. The first is direct users, such as farmers and organizations involved in land management. The second group of stakeholders includes the researchers working on problems dealing with soil gas emission as applied to climate change, agriculture land management, and efficient land management. In addition, we stated that the funding would allow us to support and train students and implement programs to support K-12 education. The educational material that will be prepared at the conclusion of the project targeting K-12 students includes discussing the different processes occurring underground and how they interact to change the environment. The materials will be suitable for the general public as well. We will develop a series of activities to raise students' awareness of soil degradation processes at different levels. Even though the above activities were proposed, it was not possible to implement them because of the constraints imposed by the pandemic. The papers, reports, and documentation that were developed during the reporting period will be used to initiate some of the activities that were proposed. Changes/Problems:The project started in mid-March of 2020 after a post-doctoral research associate was recruited. The university closed in March 4th and access was limited to the laboratories. A limited number of days we were able to access the lab, restrictions we place on the number of people that were allowed to work at the same time. It was not possible for the collaborators at UCB to visit the CSM labs to install and test the fiber optical methods. The time we were not using the lab, we made progress on the modeling. When we got full access to continue the laboratory work, the experiments were expedited Advancing modeling work we would have done after the experiments will allow us to stay still close to the schedule. We will keep the program director informed if we cannot get back on track. In June 2020, the wind tunnel facility was moved to the Army Corp of Engineer's ERDC laboratory in Vicksburg, MS. The facility was fully assembled and will be available from August 2021. The research team plan to initiate the experiments in fall 2021. The project will definitely need an extension beyond the end date of August 2022. The PI will keep the PO updated. What opportunities for training and professional development has the project provided?The project supported a full-time post-doctoral research associate and an hourly undergraduate student. The post-doc's background was in soil science. The project allowed her to develop skills in laboratory testing and modeling. The undergraduate student is in the environmental engineering program. He was able to develop skills in experimental methods and the use of sensor technologies. The student was also got the opportunity to co-author a paper with the PI and the post-doc. How have the results been disseminated to communities of interest?Because of the Corona constraints, nothing significant was accomplished but the scientific progress was shared at online meetings and conferences. What do you plan to do during the next reporting period to accomplish the goals?Even though this project funding started on September 1, 2019, the spending and the research started in March 2020. With the original funding, we will be able to continue this research until March of 2023 (3 years). From the start of the new report period from September 1, 2021 we will have 1 year and 6 months to complete the project. The collaborators at CSM, UC Berkeley, and Rothamsted will continue to meet regularly every other week to share progress at the three institutions and coordination. The next reporting period from September 1, 2020 to August 31, 2021, we plan to continue the following tasks related to laboratory testing of the sensing systems. 1. 2-D tank experiments ( target completion December 2021) o Continue the develop of the experimental methods and tuning of the experimental methods to be used in 3-D tank. o Continue the generate a data set to study gas migration and the resulting signal at the soil surface from subsurface gas injection. o Repeat the homogenous dry sand and heterogenous wet sand (unsaturated conditions) using the high accuracy sensors within the soil profile (the previous sensors were found not to have the needed resolution to measure very low concentrations) o The past experiments had issues with the ability to repeat the data. Focus on the details of measurements to achieve acceptable repeatability o Work with Lawrence Berkeley National Laboratory to model the experiments o Share the generated data with Rothamsted modeling group. o As methane is lighter than air, consider a gas that is heavier (as CO2) to use with FID. o Currently experiments use point injection-- consider distributed sources o Getting the GC to work to look at other organic gases o Crate thermal boundary conditions at the soil surface to study the effects of soil drying on the gas signals. 2. Three-dimensional tank experiments (Target completion July 2022) o Repeat similar experiments as in the 2-D tank with the current packing with three heterogenous layers and top layer of homogeneous sand. o Replace the top layer with silt that will crack due to thermal boundary conditions. o Conduct gas injection experiments with with single and multiple injection points. The distributed injection will simulate the gas generation from bio-geochemical processes. o Create wetting and drainage cycles using rainfall and water table drop o Heat boundary condition at the soil surface o Silt layer with artificial cracks o Silt with vegetation generated cracks o Modeling at LBNL and Rothamsted o Coordinate with UCB to install fiber optics to measure cracks and gas signals. ( o Set of papers on soil moisture dynamics affecting gas migration and loading 3. Wind tunnel tank experiments (completion in March 2023) o Repeat 3-D experiments in the wind tunnel under both thermal and mass flux boundary conditions o Use the rainmaker to simulate mass flux at the soil surface o Thermal and air flow boundary conditions o Use Picaro to measure CO2 concentration o Install fiber optics o Working with models at ERDC o Set of papers on integration

Impacts
What was accomplished under these goals? Understanding carbon loading from soil-vegetation systems to the atmosphere is of critical importance to assess climate change drivers. In general, soil-vegetation systems include all forests, grassland, tundra, and desserts. Long-term experiments show that the content of soil organic carbon (SOC) is the result of a balance between the inputs and outputs of organic C. In these systems, an obstacle to progress our understanding of soil is a lack of spatio-temporal data measured at high resolution at field scales. This project attempts to fill this gap by integrating spatially distributed fiber optic sensing technology and in-ground wireless sensor network (WSN). Successful completion of the research will contribute to improving our understanding of how gases generated in the subsurface migrate through different pathways in the soil that are dynamic due to the forcing conditions of climate and water application (in irrigated land) on the land surface. The sensing systems will be able to make measurements of gas flux loading to the atmosphere that is useful in developing strategies for agricultural land management that need to factor in issues of climate change and adaptability. This project is a collaboration between three institutions: (1) University of California at Berkeley (UCB), (2) Colorado School of Mines (CSM), and (3) Rothamsted Research (RR), UK. The research is organized under six work packages. The project is planned under four broad tasks with specific objectives: (1) design and development of the integrated sensing systems, (2) testing the system under highly controlled conditions in laboratory test systems, (3) field deployment and modeling. The research tasks 1 and 3 that are primarily led by the two collaborating PIs at UCB and RR. The USDA/NIFA component of the funding assigned to the CSM PI primarily supports the second task. A detailed report with experimental design and results has been submitted to the NIFA program officer. The equipment available for the measurement of gas concentration was a flame ionizing detector (FID) that can only be used with combustible gases. The capabilities for CO2 measurements exist in the climate-controlled couples soil tank and wind tunnel facility that becomes available after August 2021. To develop and validate the experimental methods, before using the wind tunnel facility, methane that can be measured with the FID was used as the test gas. 1. The 2-D tank, Intermediate-scale experiments - soil gas migration An intermediate-scale 2-D system was developed to test methane gas leaks in sandy soils along with a distributed soil moisture, temperature, and electrical conductivity sensors under controlled conditions. The test tank was used to mimic methane leaks into the unsaturated and partially saturated sandy soil layers. The system was packed with a homogeneous and heterogeneous configuration using five uniform silica sands with texture range varying from fine gravel to very fine sand. The layered packing configuration was chosen to create zones with different permeability and soil moisture content. These setups had the advantage of being able to accurately characterize the heterogeneity through packing with test soils with well-defined properties and to control the boundary and initial conditions that are not possible in the field. The experiments focused on observing soil gas migration in comparison to soil moisture patterns in each packing configuration, as mentioned earlier. Multiple trials were conducted in the experiment to ensure the consistency of the measurements. 2. The 3-D, Experimental setup for the soil moisture variation The development of the 3-D test systems was completed. The primary goal of the experiments was to test the ability of a soil sensing system to acquire data of soil moisture response to changing water table and rainfall infiltration. The experiments focused on observing soil moisture patterns in the sandy soil and investigate the performance of soil moisture sensors. Multiple trials were conducted in the experiment to ensure the consistency of the measurements. The water table was raised and lowered using the upstream and the downstream gated reservoirs. The wetting and drainage phase simulated complex transient soil moisture distribution in the tank to test the ability of the sensors to capture the dynamics. The 3D tank was packed with a different sand layers setup for the soil methane gas migration. Eight gas injection pipes were located 12.5 cm from the bottom of the tank within the #30 layer to mimic a gas leakage system. Each pipe is 60 cm long with an internal diameter of 2 cm, with a porous filter acting as a gas injection port. The soil moisture sensors were placed 20 cm apart from each pipe to prevent any of them from influencing or disturbing the gas migration pathway. The methane gas (98%) was injected into the porous media (#30 sand) above the water table level, at a rate of 10 mL/min over a 1-min period. Since the saturation has not reached steady-state conditions, methane concentrations were measured only at the soil-atmosphere interface using the combustion flame ionization detector Fast FID. The key findings contributing to meeting the project goals: • The degree of saturation is a key controller for gas migration in variably saturated porous media since both air permeability and gas diffusivity are strongly soil-water content-dependent parameters. As the degree of saturation varies dynamically with the mass (precipitation, evapotranspiration, and irrigation water application) and thermal flux (climate factors) boundary conditions, the gas loading signal at the soil surface will be spatially non-uniform and temporary variable. Hence, The sensing system needs to be distributed and adaptive. • Both advective (pressure-driven) and diffusive (concentration-driven) transport occurs, however, the flow becomes more and more diffusion-controlled as gas moves away in the case of point sources. When the gas reaches the soil surface, the migration will be fully controlled by diffusion and get diluted after reaching the atmospheric boundary layer. Any surface measurements of these low concentrations will be affected by the atmospheric boundary layer dynamics and climate. This suggests the importance of testing the systems in the wind tunnels facility where boundary layer and climate can be controlled. • When the data from the homogeneous and heterogeneous packing configurations were compared, it is very clear that the heterogeneities significantly change the gas migration pathways, and hence the gas loading signal observed at the soil surface. This has to be factored in when designing the soil sensing system and interpretation of the data. • It was found that the gas preferentially migrates through connected pathways in the heterogeneous systems. These pathways will change depending on the dynamics of the water saturations controlled by the water table fluctuations and water flux at the soil surface. Further research is needed to provide accurate estimates of soil gas emissions related to soil moisture content and soil properties. • Preliminary testing done in the wind tunnel test systems with fiber optics showed that the optical fibers were able to detect thermal cracks on the soil surface. We hypothesize that the fractures in the soil surface created due to soil drying from thermal boundary conditions and plant growth will affect the gas signals. • Preliminary modeling done in collaboration with the Lawrence Berkeley National Laboratory showed that the models are able to capture the soil water dynamics in heterogenous packing in the 3-D tank. ?

Publications

• Type: Conference Papers and Presentations Status: Awaiting Publication Year Published: 2020 Citation: Ilie, A.C,Goebel, C.,and Illangasekare, T.H., 2021. Performance assessment of soil moisture sensors under controlled conditions in a laboratory setting and recommendations for field deployment, 5th International Conference on Water Science. DOI: 10.3390/ECWS-5-08041 International Electronic Conference on Water Sciences, conference paper in press to be published by Environmental Sciences Proceedings MDPI.
• Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Ilie, A.M.C, Illangasekare,T.H.,Soga, K., Whalley, R., Trautz, A., and h Cihan, A.,2021. Intermediate-scale testing of a spatially distributed sensing technology for monitoring gas emission from soils as applied to climate change, International Society of Porous Media, Annua Meeting
• Type: Conference Papers and Presentations Status: Awaiting Publication Year Published: 2021 Citation: 1. Ana Ilie, Cody Goebel, and Tissa Illangasekare. Design and Implementation of an Intermediate-Scale Test to Study Effects of Soil Heterogeneity on Methane Gas Migration. June 2021, 3rd Euro-Mediterranean Conference for Environmental Integration to be published in proceedings by Springer ASTI.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2021 Citation: Ana Ilie, Cody Goebel, and Tissa Illangasekare. Design and Implementation of an Intermediate-Scale Test to Study Effects of Soil Heterogeneity on Methane Gas Migration. June 2021, 3rd Euro-Mediterranean Conference for Environmental Integration, abstract oral presentation.
• Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Ana Ilie, Tissa Illangasekare, Kenichi Soga, William Whalley. A laboratory study of the effect of soil-water dynamics on the migration of gases from subsurface sources. April 2021, EGU General Assembly 2021, abstract oral presentation.
• Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: Ana Ilie, Cody Goebel, Tissa Illangasekare, Kenichi Soga, Richard Whalley and Abdullah Cihan. Intermediate Scale Test System to Study Soil Moisture Dynamics and Gas Migration Using Sensors  Preliminary Tests. September 2020, Signal in the Soil (Sits) Workshop Principal Investigators, abstract poster presentation.

Progress 09/01/19 to 08/31/20

Outputs
Target Audience:In the original proposal submitted to NSF, under the broad impacts, we identified our target audience as the scientific community and industry. We identified two broad stakeholder groups that will benefit from the developed technology. The first is direct users, such as farmers and organizations involved in land management. The second group of stakeholders includes the researchers working on problems dealing with soil gas emission as applied to climate change, agriculture land management, and efficient land management. In addition, we stated that the funding would allow us to support and train students and implement programs to support K-12 education. The educational material that will be prepared at the conclusion of the project targeting K-12 students includes discussing the different processes occurring underground and how they interact to change the environment. The materials will be suitable for the general public as well. We will develop a series of activities to raise students' awareness of soil degradation processes at different levels. Even though this reporting period covers one year starting from September 2019, the funding was received in February 2020. The project started in mid-March after a post-doctoral research associate was recruited. Changes/Problems:Two factors impacted the progress of the project. Even though this reporting period covers one year starting from September 2019, the funding was received in February 2020. The project started in mid-March after a post-doctoral research associate was recruited. This delayed the start of the project for six months. As finds were not expended as planned, it will be possible to accomplish the proposed task with a six-month no-cost extension. The second factor was the COVID-19 pandemic that restricted access to the laboratory. In addition to the limited number of days we were able to access the lab, restrictions we place on the number of people that were allowed to work at the same time. The time we were not using the lab, we made progress on the modeling. When we get full access to continue the laboratory work, having the model will be useful to expedite the experiments. Advancing modeling work we would have done after the experiments will allow us to stay still close to the schedule. This is assuming that we will have full access soon.We will keep the program director informed if we cannot get back on track. What opportunities for training and professional development has the project provided?An undergraduate hourly student supported through PI's endowment funds assisted in setting up test systems for UCB collaborators to conduct preliminary sensor tests. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?2.1 Test method development The objective of this research task is to develop testing methods specifically applicable to the soil sensing application. We will continue to develop the sensing system to make measurements of soil moisture and gases. The experimental methods for the soil moisture measurements that include calibration, installation, and data transfer have been completed. The development of sensing methods for gases is still in progress. The gas that is of interest is CO2. The UCB collaborators are developing optical fiber-based sensing systems for gas measurements. CSM currently has a gas chromatograph (GC) and ionizing flame detector (FID) for measuring volatile gases and methane. As we don't have a way of benchmarking the sensors for CO2, we will continue to develop sensors to measure inert gases and methane. The data will be used in the modeling until the fiber optic-based sensor are ready to be installed. 2.2 Preliminary proof of concept experiments The objective of this task is to conduct a preliminary set of experiments under scenarios that are expected in the field. The experiments that were initiated during the last report period on water table fluctuations and rainfall will be continued. During this reporting period, we will complete the Phase I and Phase II experiments. We will use the data generated in these experiments to continue the model development effort in collaboration with the Lawrence Berkeley National Laboratory. 2.3 Distributed sensor installation The objective of this task is to complete a step vise installation and testing process of the sensing systems that are under development at UCB. The progress of this task will depend on the ability of the UCB collaborators to travel to Golden. Currently, there are restrictions for them to travel. The research groups remotely meet with the UCB and UKRI every other week where the progress is evaluated, and the development activities are coordinated. 2.4 Laboratory testing of the integrated sensing system The objective of this task is the installation of the fully integrated sensing system in the laboratory testbed and conduct all the necessary tests before field deployment in the site in the UK. We will completer the Phase, and Phase II experiments will start the Phase III experiments. Until the travel restrictions are relaxed, it will not be possible to make any progress on this task. We continue the integration of the sensor systems that are installed in our laboratory test system.

Impacts
What was accomplished under these goals? A. Development of the three-dimensional test facility The 3D tank has dimensions of 4.87 m (long) × 2.43 m (wide) × 0.76 m (depth). For the proposed tests, the tank needs to have the capability to control the water table and the soil moisture. The water table is controlled using six independently gated constant-head reservoirs, which allows for the creation of various flow and dynamic water table configurations within the tank. The tank contains 44 wells with sensors to monitor the dynamic response of the water table subjected to changes in the boundary heads and rain infiltration. The other sensors installed in the tanks for the measurement of soil moisture, temperature, and gas concentrations will be presented under the design of experiments. Figure 1: Three-dimensional test tank set up at the Center for Experimental Study of Subsurface Environmental Processes (CESEP at the Colorado School of Mines The system was packed with a heterogeneous configuration using five uniform silica sands with the effective sieve numbers #70, #16, #8, #12/30, and #20/30 (Accusands, Unimin Corp., Ottawa, MN. The sands have porosity ranging in values from 0.31 to 0.42. The tank configuration presents a homogeneous zone and a heterogeneous zone with four different sand columns 17in x 17in, as shown in Figure 1. The water that outflows from the system during water table fluctuations and rainfall will be collected at the downstream end of the tank using a 5-gallon tank. B. Design of rainfall simulator A rainfall simulator was developed to control the water flux on the soil surface. We designed and installed a rainfall simulator made of porous hoses for irrigation systems (Figure 2). These porous hoses were fixed into a PVC pipe and placed in alignment with each other for uniform water distribution. The rainfall simulator is connected to a pump with a known flow rate and a constant head device. The size of the drops and their fall velocity is held constant, while the distribution of rainfall across the test area is uniform and can be achieved with low water pressures. The amount of simulated rainfall during the test will be measured. One approach is to cover the whole test area with a water container and measure the rain caught in a fixed period. This will be done before and after the experiments to make sure there is no change during the test. Alternatively, measurements can be made during the test either by small rain-gauges installed on the test area or by collecting channels across the area. Figure 2: Details of the rainfall simulator C. Installation of WSN, soil moisture sensors and pressure sensors Wireless sensors network (WSN) experiments were done in collaboration with UC Berkeley, to check the ability of WSN's communication in the tank, looking into the signal strength at different depths and hop between nodes. Depth and intervals of buried wireless nodes were changed several times to better understand the differences resulting from signal strength as illustrated in Figure 3. Figure 3: Initial wireless sensor networks tests by UCB Figure 4: Pressure sensors connected to the wells through white hoses and Arduino microcontroller Thirty-four sensors (ECH2O 5TE and EC5 Meter Environment) were available for Phase1 experiment and each one was calibrated to provide unique fitting parameters relating the sensor's measured voltage to known water content. Each sensor has its calibration slope and zero offset intercept, therefore requiring a unique calibration be conducted. Each sensor has been calibrated for each type of sand #70, #8, #16, #20/30, #12/20. The calibration process included for each sensor went through the following steps : - 1000 cubic centimeters of corresponding sand weighed and put in 8in. x 8in. x 2in. metal bin - sensor buried in a sand sample and packed down moderately by hand - measurements taken for 10 minutes - 100 ml of water added to sand, thoroughly mixed - sensor replaced and sand packed down again - repeated until saturation reached, measurements were taken for the final time - measured Volumetric Water Content vs sensor avg. reading charted The sensors calibration data was acquired by directly connecting each sensor to a data logger and then to a computer for real-time sampling. Based on linear regression analysis, coefficient of determination (R2) was calculated for each calibration curve fit. For all 34 sensors, the minimum R2 value obtained was 0.98, further indicating that a linear calibration relationship exists for all sensors used. The Decagon EM50 data loggers are capable of reading five sensors each; therefore seven data loggers were required for all 34 sensors, as shown in Figure 5. Figure 5: To measure soil moisture, seven datalogger with 34 sensors (EC5 and 5TE) were placed in the 3D tank. Sensor readings were taken once every minute and recorded with Decagon EM50 dataloggers. The 3D tank with soil moisture and dataloggers is shown in Figure 6. Each sensor was individually calibrated and sensor raw data was checked before the calibration, looked into the noisy sensors to be replaced with new sensors. After the calibration process, all sensors were checked and validated against each other. Few noisy sensors resulted after the calibration process were replaced. Data processing, data cleaning, and visualization were done through Tableau Software. Figure 6: Sensor layout in the 3D tank D. Design of experiments The experiments initiated and progressed during this report period are done in two phases: (1) monitoring of soil moisture dynamics, and (2) monitoring gas migration. A third phase involving effects of soil fractures on the gas migration will be initiated during the second year of the project. The progress made in Phase I experiments and the setup of the Phase II experiments are presented. Phase I experiments on soil moisture monitoring: The primary goal of this experiment is to test the ability of a soil sensing system to generate data on soil moisture response to changing water table and rainfall infiltration. Both imbibition and drainage will be simulated. The experimental setup is schematically shown in Figure 7. The experiments will be focusing on observing soil moisture patterns in the sandy soil and investigate the soil moisture sensors performance. We plan to raise the water table using both the end reservoirs and application of rainfall. Our expectation is that both these excitations will creates complex transient soil moisture distribution in the tank and we will be able to capture the dynamics. After the fiber optics based and the wireless sensor network hardware are developed by the collaborators at UCB, they will be installed in conjunction with the current system. These two systems will allow us to evaluate the integrated distributed soil sensing system before field testing and demonstration at UKRI. In addition to the sensors in the unsaturated soil, the 44 wells integrated into the tank are connected to pressure sensors that allows or the monitoring of the water table response to the changes in the boundary heads and the rainfall infiltration. The lowering of the water table will simulate drainage conditions. Even though the main focus is not gas migration, few gas sensors will be installed to help in designing the Phase II. In parallel to the experiments, a modeling effort that will help to interpret data and eventually develop modeling strategies is underway. Part of this effort is to evaluate whether traditional Richards Equation-based numerical models can capture the transients in soil moisture. The modeling effort and progress are presented later in the report. The software was used to model the soil organic carbon data from three long-term field experiments (Urumqi, Yangling and Zhengzhou) in China. The long-term goal is to couple RothC with the unsaturated zone flow and gas migration model developed at the Lawrence Berkeley National Laboratory (LBL)

Publications