The Automated Bulk Sampling System (ABSS), a low-cost solution for integrated nitrous oxide emission quantification in field studies

The Automated Bulk Sampling System (ABSS), a low-cost solution for integrated nitrous oxide emission quantification in field studies | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Method Article The Automated Bulk Sampling System (ABSS), a low-cost solution for integrated nitrous oxide emission quantification in field studies Randy Clark, Nick Friedenberg, Dan Chamberlain, Chris Parry, Timothy Hart, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9013846/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Agricultural soils are a major source of nitrous oxide (N 2 O), a potent greenhouse gas, but challenges in measuring its highly dynamic flux in field settings hamper modeling and mitigation efforts. We developed a remotely-deployable, high-frequency sampling system that lowers the cost of N 2 O flux measurement and minimizes laboratory analysis. The Automated Bulk Sampling System (ABSS) is self-powered and accumulates hourly open- and closed-chamber headspace gas samples into two separate gas collection bags. The system produces two bulked gas samples at the end of the measurement period, allowing average hourly N 2 O flux over 2-week collection periods to be estimated. Lab-based validation experiments showed high agreement between real-time analyzer and accumulated ABSS concentration readings (r 2 : 0.998, bias: -0.009 ± 0.002). The system also showed high precision, or repeatability (r 2 : 0.791) in field validation experiments, but an underestimation bias of 25% for N 2 O fluxes was observed when compared to 2-week average real-time analyzer results. In exploring sources of error, we found overestimation of ambient, open-chamber samples by the ABSS to be the largest source of error (15%), augmented by underestimation of closed-chamber sample concentrations (5%). Loss of information from meteorological variation and two-point flux calculation contributed slightly to underestimation bias (6%). We used historic weather data from the U.S. Corn Belt to simulate the potential error contribution from air density variation, and found an average error of 0.049%, with the largest range in error occurring at lower fluxes. Our results demonstrate that ABSS is a valuable low-cost and low-labor solution for integrated estimates of soil N 2 O flux in large-footprint, replicated plot experimental contexts and can help resolve critical questions in managing soil N 2 O emissions in agricultural systems. Greenhouse Gas Nitrous Oxide Trace Gas Flux Agriculture Chamber Soil Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction Nitrous oxide (N 2 O) is a potent greenhouse gas (GHG) and is the largest direct component of on-farm GHG emissions in row-crop agriculture (Mbow et al., 2017 ; Reay et al., 2012 ; Tian et al., 2020 ). Emissions of N 2 O from agricultural soils are a byproduct of nitrification and denitrification by native soil microbes and are driven by environmental and management factors including temperature, water, oxygen, pH, substrate availability, and energy resources (Butterbach-Bahl et al., 2013 ). Considering the large role of agriculture in anthropogenic N 2 O emissions, there is a critical need to quantify management impacts on soil N 2 O emissions (Olander et al., 2014 ). Despite decades of research, mechanistic models of N 2 O flux often struggle to replicate empirical measurements, emphasizing gaps in our understanding that require additional data collection efforts to address (Fuchs et al., 2020 ; Reay et al., 2012 ). A major challenge is the high spatiotemporal variation in N 2 O emissions influenced by micro-climate and micro-site factors, often called “hot spots” and “hot moments” (Wagner-Riddle et al., 2020 ). Relatively infrequent and low-density gas flux measurements often miss key locations or events in the flux landscape, leaving the links between agricultural management with N 2 O emissions uncertain (Charteris et al., 2020 ). Some agricultural practices have relatively well-known effects on emissions (i.e. fertilizer reduction, the use of nitrification inhibitors or other enhanced efficiency fertilizers), but other potentially mitigative factors such as tillage, cover crop use, biological and organic amendments, and cropping system and crop variety are less understood (Charles et al., 2017 ; Della Chiesa et al., 2024 ; Olander et al., 2014 ; Storer et al., 2018 ). The large cost, complexity, and effort required for N 2 O data collection limits the frequency and replication of measurements, while also creating both resource and technical barriers to understanding agriculture management impacts on GHG emissions. Chamber headspace sampling methods are the primary approach to measuring N 2 O and other GHG emissions from plot-scale agricultural experiments. Manual non-steady-state (i.e. closed) chamber collection methods are the most common and inexpensive and are generally the most time-consuming and laborious. In these approaches, chambers are installed and sampled multiple times while fully closed for a period, usually less than an hour, then the samples are transported back to a laboratory for analysis (Charteris et al., 2020 ). Portable, “fast analyzers” may also be employed to measure gas fluxes directly in the field by transporting a flow-through analyzer to each collar site manually, which saves some time and effort (Maier et al., 2022 ). These manually measured chamber site locations are typically visited at weekly to monthly intervals and allow for flexibility in spatial layouts (Charteris et al., 2020 ) and the ability to measure multiple plots or locations with a single portable chamber. The use of automated chamber methods, by contrast, provides greater temporal resolution through higher-frequency measurements throughout the experimental period. These methods either use automatic samplers, which store individual samples in separate vials, or utilize real-time gas analyzer systems to record trace gas concentrations at the time of sampling with no lab analysis required (Davis et al., 2018 ; Lawrence & Hall, 2020 ). Both automated chamber approaches have drawbacks, however. Automatic samplers may offer greater flexibility for deployment and scaling, but they can generate a cumbersome number of samples, leading to relatively high labor and analysis costs. Real-time analyzers coupled to automated chambers are costly and require significant expertise to deploy, power, and maintain, while also limiting the experimental footprint and the number of chambers that can be monitored. Bulk sampling of gas, wherein multiple samples are accumulated into a single container over time, provides an intermediate solution, leading to a reduced lab analysis workload while capturing cumulative emissions with complete, high-frequency temporal coverage (Ambus et al., 2010 ). Like automatic samplers, bulk sampling minimizes the in-field labor requirements and separates the sampling apparatus from the more expensive analytical instrument, allowing scalability and flexibility of deployment. An example, the SIGMA bulk sampler (Ambus et al. 2010 ), was used in a 2009 field experiment (Juszczak and Augustin, 2013) but has not seen wider use, perhaps due to cost or lack of commercial models. Sample pooling has also been applied spatially, where gas samples from multiple chambers are pooled to account for spatial heterogeneity and reduce analytical effort (Arias-Navarro et al., 2013 ). With relatively simple engineering, bulk sampling systems could provide an affordable and flexible route to broader N 2 O mitigation investigations. However, potential biases and sources of uncertainty must be properly considered. In consideration of the need for low-cost, scalable N 2 O measurement systems and the potential of the “sample bulking” approach, we sought to design and develop a low-cost bulk sampling system to measure N 2 O (and other trace GHGs) in remote field settings compatible with agricultural small plot and on-farm experimental designs. Our second objective was to validate this system and assess sources of measurement uncertainty, including systematic and mechanical contributions, as well as those introduced by natural variation in atmospheric conditions over the sample collection period. To address the first objective, we designed, fabricated, and validated the Automated Bulk Sampling System (ABSS). The system collects and composites high-frequency samples of chamber headspace gas into air-tight gas collection bags that can be retrieved, replaced, and analyzed every two weeks. To validate the system and the general sample bulking approach, we conducted lab dosing experiments and a field test in corn ( Zea mays L.) in Iowa, USA, where we directly compared soil N 2 O flux estimates generated from the ABSS system with fluxes measured with a real-time in-field analyzer. We investigated different sources of error within our field-generated dataset, including estimating data loss due to the “bulking” approach. We also used historic meteorological data to estimate the likely error contribution from weather variation. 2. Materials and Methods 2.1 Design and fabrication The ABSS system was designed and fabricated using low-cost materials and is powered by a solar panel and battery for remote deployment. It consists of cylindrical chamber with an automatic, air-tight lid connected to an insulated enclosure housing the electrical components, valves, and the two gas sample collections bags (Fig. 1 ). The system collects specific gas volume samples from the headspace of an automated chamber at pre-defined periods (i.e. hourly for two weeks) from two different chamber closure states: open-chamber (OC) and closed-chamber (CC). All samples are combined into single gas sample bags for each closure state. Details on the chamber components, construction and design are described in the Supplemental Methods. 2.2 Lab validation 2.2.1 Lab experimental setup Our initial lab validation experiments used gas injection to simulate soil fluxes and compared composite N 2 O measurements from the ABSS with concentration measurements taken in real-time from automatic chambers. In this approach, we focused on comparing direct N 2 O concentrations of open-chamber air and closed-chamber air that was injected with N 2 O calibration gas, as these two values underpin the estimation of flux. A set of four automated chambers were programmed to cycle through open-chamber and closed-chamber states 336 times over 2.8 days. A set amount of N 2 O calibration gas (Gasco Affiliates LLC, Oldsmar, FL) was injected into the chamber headspace during the closed-chamber state to simulate a target flux rate. Individual experiments were run with consistent, non-varying target flux rates, as well as with “spikes” representative of short term, higher flux moments. 2.2.2 Real-time analyzer measurements in lab validation All automated chambers were connected in parallel to a Picarro G2580, with a valve controller system that allowed each individual chamber to be measured separately. The analyzer system cycled through the four chambers so that the real-time measurements were taken for 80 seconds with 10 seconds of ambient air reading between chambers. The chamber closure timing was staggered so that the real-time analyzer cycled through all four chambers in an open state and then again when all four chambers were in a closed state at the same phase of their cycle. The chamber headspace concentration reading was taken as the average the last 30 seconds of readings from the 80 second measurement time. The average concentration over all the cycles in the experiment was then calculated to compare to the ABSS bulk measurement, described below. 2.2.3 ABSS measurement in lab validation Each automated chamber was also connected to an ABSS gas sampling system so that headspace gas samples were collected during the open-chamber and closed-chamber phases of the cycles. The ABSS gas samples were synchronized to be collected in the last 10 seconds of the real-time analyzer measurement time. All samples taken during the open-chamber phase of the cycles were collected in the same open-chamber sample bag, while all samples taken from the closed-chamber phase were collected into the same closed-chamber bag, resulting in two sample bags at the end of each experiment. After completion, the bag N 2 O concentrations were measured on the same Picarro G2580 used for the real-time measurements. See Figure S4 for experimental set-up and example data plots. 2.2.4 Statistical Analysis for Lab Validation The agreement between ABSS bulk sample concentrations and those measured by the real-time analyzer were assessed by direct regression to assess overall precision (r 2 ) and bias (deviation from the 1:1 line). We further explored the bias by regressing the difference (or error) in concentration readings from the two methods against the real-time readings. Analysis of the specific concentrations from the open-chamber and closed-chambers phases, instead of the overall concentration change (flux), allowed us to better examine sources of error between the two measurement approaches. The distribution of gas concentrations investigated was non-normally distributed. Analysis employed iteratively weighted robust regression to reduce the influence of high measured N 2 O fluxes on the fit while preserving interpretation in natural rather transformed units. All analyses were performed using the statsmodels v 0.14.4 Python package and Python 3.11 (Seabold & Perktold, 2010 ). Weights were obtained from the rlm (robust linear model) function and used as inputs to the wls (weighted least squares) function to obtain traditional model summary statistics. 2.3 Field Validation Experiment The robustness of the ABSS and 2-point bulk sampling method under field conditions was empirically tested during the spring and summer of 2023. In this evaluation, the average soil N 2 O fluxes captured by the ABSS were compared to average soil N 2 O flux computed from averaging real-time flux measurements taken over 2-week sample collection periods. 2.3.1 Field Experiment Design Soil gas measurements were collected from a corn field in Johnston, IA at the Corteva Agriscience Research Farm from September to October 2023. Average temperature during the sampling period was 18.9°C and rainfall totaled 6.2 cm in September. The soil is classified as a Wiota silty clay loam. Airtight PVC chambers with automated lids were mounted on top of stainless-steel collars that were inserted 10 cm into the soil and secured with gaffer’s tape. Once installed, the inside dimensions of the chamber headspace above the soil measured 30 cm tall by 25.45 cm diameter. Field validation testing consisted of two separate two-week sample collection periods, where 8 ABSS were connected to separate GHG chambers that were part of an ongoing 12 chamber real-time N 2 O measurement experiment. In this design, ABSS gas samples were collected from the same chambers that were being monitored by a real-time N 2 O analyzer system. 2.3.2 Real-Time Analyzer Measurements in Field Validation Real-time field measurements of N 2 O concentration in chamber headspaces were made via a custom multiplex valve system that allowed each chamber to be isolated in a closed loop with an LGR-ICOS™ GLA151- N 2 O M1 portable N 2 O /CH 4 analyzer (LGR/ABB Inc., Zurich, Switzerland). At the beginning of each measurement cycle, all 12 automated chamber lids closed simultaneously. The real-time analyzer system then cycled through each chamber, measuring headspace gas for 76 seconds each time, and cycling through all chambers 4 times before all chambers opened again. As a result, each chamber has four headspace measurements, but all at different times within the chamber closure period, which lasted about 60 minutes. This closed period was followed by a 60-minute open chamber period to allow the concentrated gases in the chamber headspaces to vent and equilibrate with the surrounding environment. The rate of chamber concentration change was calculated by regression using best-fit model selection between quadratic or linear models. Quadratic fits are a simple and performant way to capture possible concentration saturation over the course of a chamber measurement (Parkin et al., 2012 ; Wagner et al., 1997 ). The quadratic model was rejected when its quadratic term was positive, when the model was not statistically significant at the α = 0.05 level, or when the quadratic model had a Bayesian Information Criterion (BIC) score less than 2 units lower than that of a linear fit. The rate of concentration change (ppm/min), \(\frac{dC}{dt}\) , was taken as the linear term of the selected model or 0 if neither model was statistically significant, equivalent to the first derivative of the regression model evaluated at t = 0. Flux of N 2 O was estimated using air pressure, P (atmospheres), and temperature, T (degrees Kelvin) from on-site weather station with 15-minute resolution, using the measurement most coincident with chamber closure time. The formula for flux, F , estimated from real-time analyzer data was $$F=\frac{dC}{dt}\frac{P}{T}\frac{V{M}_{N}}{RA}\times60\text{min/h}$$ 1 , where \(\frac{dC}{dt}\) is the rate of concentration change (ppm/min) extracted from the best-fit regression line, \(\frac{P}{V}\) is the air density coefficient extracted from the weather station, V is the headspace volume of the GHG chamber (L), M N is the molecular weight of N 2 O-N (28 µ g/ µ mol), R is the ideal gas constant (0.082057 atm L mol − 1 K − 1 ), and A is soil surface area inside the GHG chamber (m 2 ). 2.3.3 ABSS Measurements in Field Validation Similar to the laboratory validation experiment, ABSS system headspace samples were collected from the same chambers as the real-time analyzer. However, while real-time analyzer measurements were taken at varying times in the chamber closure cycle, all ABSS samples were coordinated to be taken simultaneously to correspond with the synchronized chamber closure. Thus, all ABSS closed-chamber samples were collected 10 minutes after chamber closure to avoid saturation effects. The open-chamber samples were collected at the end of the 60-minute open-chamber phase, just before chamber closure. A timeline for the 4-point real-time measurement process and synchronized ABSS samplings is outlined in Supplemental Table ST2. After each two-week cycle, the two bags from each ABSS unit were retrieved and transported to the lab, where the N 2 O concentration in each bag was measured with Picarro G2508 (Picarro, Inc., Santa Clara, CA, USA). Average hourly N 2 O flux over the two-week measurement period was calculated using the equation: $$F=({C}_{cc}-{C}_{oc})\stackrel{-}{\left(\frac{P}{T}\right)}\frac{V{M}_{N}}{RA}\times\frac{60\text{min/h}}{D}$$ 2 Where \({C}_{cc}\) and \({C}_{oc}\) are the concentrations of N 2 O in the closed-chamber and open-chamber composite sampling bags, respectively, \(\stackrel{-}{\left(\frac{P}{T}\right)}\) is the mean ratio of atmospheric pressure to temperature (atm K − 1 ) over the two-week sample collection period, V is the headspace volume of the GHG chamber (L), M N is the molecular weight of N 2 O-N (28 µ g/ µ mol), R is the ideal gas constant (0.082057 atm L mole − 1 K − 1 ), A is soil surface area inside the GHG chamber (m 2 ), and D is the closed chamber duration between the OC and CC samplings (10 minutes). 2.3.4 Estimating “expected” bulk concentrations from real-time N 2 O flux data In addition to directly comparing the flux estimation from the real-time vs. the ABSS methods, we wanted to investigate the specific potential sources of discrepancy between the two methods. The potential sources were 1) differences in sampling times, 2) two- vs. four-point flux estimation, and 3) the loss of specific meteorological data at the time of sample collection. Each of these sources and our correction approach is described below. We examined each of these sources of error by generating a synthetic “expected” bulk sample dataset derived from the real-time data. The ABSS collected ambient samples before chamber closure and concentrated samples 10 minutes after closure, while the real-time samples were taken at 15-minute intervals starting at various times after chamber closure. We addressed this misalignment of sampling times by estimating real-time N 2 O concentrations at 0 and 10 minutes using the regression models from which we had obtained dC/dt . This approach also allowed us to estimate the inherent error introduce solely from the ABSS’s two-point sampling approach. Finally, we used air density data to account for variation in the number of molecules added to the bulk bags with each sample, assuming a constant sample volume. The resulting “expected” bulk concentration dataset derived from the real-time data was then used to investigate the contribution of each potential source of error inherent in the bulk sampling approach. 2.3.4 Statistical Analysis for Field Validation Hourly flux from each chamber measurement system was compared using ordinary least squares regression ( ols function in statsmodels package). The precision of the measurements, i.e. how reliably a real-time flux estimated corresponded to a particular ABSS flux value, was interpreted as the r 2 value of the model. We estimated ABSS system bias as the slope of the regression line, specifically its deviation from a slope of 1. We compared open-chamber and closed-chamber bag concentration measurements from the ABSS system with the synthetic “expected” bulk concentrations using one-way ANOVA to estimate the error contribution directly from the bulking process. We assumed independence between groups, and so did not include a random chamber effect in the model. 2.4 Meteorological variation assessment Fluctuations in atmospheric temperature and pressure cause air density variation across the 2-week sample collection periods and introduce unresolvable measurement error due to unknown gas sample density at collection time. To estimate the potential error contribution of air density variation, historic weather datasets were used, where two-week periods from 10 years of hourly historic weather data (IBM, 2024 ) was randomly sampled at 49 gridded locations across the US Corn Belt (Supplemental Figure SF6) to construct 800 hourly weather datasets per location. The historic weather datasets were then overlaid with flux measurements from the field validation experiment to generate 4000 simulated ABSS datasets per location. Comparisons of ABSS flux measurements with and without correction for the known air density at each sampling event were then used to evaluate air density contributions to bulk sampling measurement error. 2.5 Gas stability testing The stability of the N 2 O gas concentrations in the gas collection bags over 2-week periods was evaluated by filling 3 bags with N 2 O concentrations in range with previously collected closed-chamber samples (~ 0.5ppm). The filled and closed bags were then stored in the lab or greenhouse and the N 2 O gas concentrations were measured at 0, 1, 7 and 15 days after filling to observe any changes in the N 2 O concentrations over time. 3. Results 3.1 Lab dosing experiment N 2 O concentration readings from samples collected by the ABSS showed high precision to those measured directly from the chambers (r 2 = 0.998), and a slope not different from the 1:1 line (Fig. 2 ). However, further scrutiny of deviation in ABSS concentration readings showed a statistically significant departure from direct chamber measurements, with an overestimation of N 2 O at low concentrations to underestimation at high concentrations (Fig. 2 inset). These parameters suggest the ABSS on average overestimates typical ambient N 2 O concentrations (334 ppb) by 1.2 ppb in the lab while underestimating concentrations over 356 ppb, though the bias is small (~ 0.9%). This inference was supported by a t-test of open-chamber readings, which shows a statistically significant 1.2 ppb higher average reading from ABSS samples than from the real-time gas analyzer (t 44 = 4.56, p < 0.0001, Supplemental Table ST3). 3.2 Field validation We successfully captured paired 2-week flux data for 9 chambers over two different sampling periods. Most lost data came from malfunctions in the chamber lid control and custom multiplex valve system. Average N 2 O flux over the 2-week sampling period ranged from 41.9–175.4 ug N 2 O-N m − 2 hr − 1 . The synthetic “expected” bulk sample dataset derived from the real-time measurement provided an idealized maximum accuracy of the ABSS, due to the lost meteorological information and the limitations of the two-sample flux estimation procedure. This “expected” bulk dataset underestimated the real-time flux by 6% (F 1,8 = 752, p < 0.0001, r 2 = 0.99, red shaded interval in Fig. 3 ; Supplemental Figure SF5a). Direct comparison of the ABSS and real-time analyzer calculated hourly flux indicated an overall underestimation by the ABSS system across all flux levels of 24.7% (t = -5.9, p < 0.0001, 95% CI: 15.1–34.3%, r 2 = 0.79; black trend line in Fig. 3 ). This additional overall system bias (i.e. the additional grey shaded wedge in Fig. 3 ) indicates further deviation of flux measurements in practice from the theoretical “best” (i.e. red shaded wedge). An initial regression of the difference between systems against direct measurements indicated the intercept was not significantly different from zero (p = 0.99). 3.3 Sources of Field Validation Bias ABSS flux estimates were biased by an apparent overestimation of open-chamber N 2 O concentrations. We found that the bulked ABSS open-chamber samples had N 2 O concentrations 15% higher than the “expected” concentrations, (mean difference: 5.6 ppb; Table 1 , Fig. 4 ). Closed-chamber measurements were underestimated by an average of 1.93 ppb, though this did not represent a significant difference between systems (Table 1 ). Figure 4 illustrates the close agreement between observed and expected closed concentrations compared with a lack of correlation in open chamber readings. Table 1 ANOVA results assessing the difference in N 2 O concentration measurements between ABSS and “expected” bulk concentrations at t = 0 (Open-Chamber) and t = 10 minutes (Closed-Chamber; pooled standard error = 1.68). Source Mean Difference in ABSS vs. “Expected” Units p-value % Error Open-Chamber N 2 O 5.60 ppb N 2 O 0.004 14.6 Closed-Chamber N 2 O -1.93 ppb N 2 O 0.266 -5.0 3.4 Estimating air density variation contribution to error Air density across our analysis area and time window showed 2-week variations ranging from 0.041 to 0.26 kg m − 3 , which corresponds to 0.2% − 1.1% error in flux rate (Supplemental Figure SF6, Panel C). Absolute simulated flux error increased with increasing flux, with average error ranging from 0.0004 to 0.0448 ug N 2 O-N m − 1 h − 1 for low and high flux rates, respectively (Supplemental Figure SF6, Panel D). 3.5 Gas stability testing The gas stability test showed no significant change in N 2 O concentration with a 95% confidence interval of concentration change ranging from − 5.006 to 7.074 ppb after a 15-day storage period (n = 3; Supplemental Table ST4). 4. Discussion 4.1 Utility of ABSS The results presented here demonstrate that the ABSS can provide consistent, intercomparable results compared to real time analyzers for integrated two-week trace N 2 O gas flux estimations. Our reported results suggest that as currently deployed, the ABSS may underestimate total emissions, but would provide an accurate comparison within a given study design. The ABSS approach is best suited for studies aiming to compare bi-weekly average or cumulative emissions in replicated field experiments, such as those looking to evaluate management impacts. The modular, solar-powered operation and low cost for deployment of the ABSS fills an important technological gap for taking high-frequency measurements in large-scale experiments and landscapes. Compared to manual static chamber sampling, hourly sampling over the entire collection period by the ABSS can reduce potential bias introduced by diurnal flux variation and low sampling frequency (Ferrari Machado et al., 2019 ). 4.2 Limitations of ABSS Deployment of the ABSS for N 2 O flux measurements presents some limitations, specifically regarding absolute flux estimation and detection of short-term emission spikes. Though the ABSS was able to capture 80% of the variability in average flux for our field validation experiment, the 25% flux underestimation is a concern. Our results are comparable to those reported by Ambus et al. ( 2010 ), who reported up to a 28% negative bias in their 3-bag SIGMA system compared with conventional static and real-time automated chamber methods. Investigation into the sources of bias identified the open-chamber N 2 O concentration as the largest source of error, which may be remedied through simple changes to the operational sequence i.e. longer flushing durations when clearing sampling tubes. Our approach of estimating “expected” bag concentrations using the real-time data also revealed a 6% error due to the sample bulking process, a relatively small but unavoidable contribution. The ABSS is best suited to research questions concerned with integrated or cumulative emissions, and within-study comparisons. As currently deployed, the system provides average flux estimates for two-week periods, and so would be unable to provide estimates at finer temporal resolution scales that may be desired for some modeling purposes. For example, most “hot moments” last hours or a few days (Wagner-Riddle et al., 2020 ), behaviors that would not be directly detectable within ABSS data, though would still be captured within the cumulative samples. Recent advances in biogeochemical modeling methods to predict N 2 O emissions depend on sub-daily flux measurements, though also utilize manual weekly chamber measurements for validation (Saha et al., 2021 ). The version of ABSS presented here relied on a two-bag system for calculation of average flux. This approach was meant to deliver a low-cost, low-effort system that minimized labor and resources. In general, four gas samples per closure time is considered the “gold standard” for flux rate determination from static chamber approaches, which allows for more precise error calculation (Venterea et al., 2020 ). While saturation effects or nonlinearity of chamber concentration during the closure period often makes nonlinear flux calculation approaches more accurate, they are also more prone to random error and thus less precise, or repeatable, than basic linear calculation methods (Venterea et al., 2020 ). In their assessment of chamber linearity assumptions from a variety of datasets, Chadwick et al. ( 2014 ) concluded that in most circumstances, an assumption of linearity and two or three headspace samples are sufficient, given a relatively short chamber closure period. While the short chamber closure time helps to minimize potential saturation effects, this approach does require that the gas analyzer employed has appropriate sensitivity to detect the expected gas concentration differences between the sample bags. The ABSS chamber closure time can be modified depending on expected flux rates and gas analyzer detectability limits. 4.3. Scalability and statistical power The relatively low cost of ABSS provides benefits for increasing replication and overall statistical power. A 16-unit deployment of ABSS costs about $ 25,000, excluding the laboratory-based gas analyzer and labor for bag collection and analysis every two weeks. In contrast, the same number of chambers and the real-time analyzer system can cost over $ 250,000 when factoring in the dedicated gas analyzer (~ $ 100,000), multiplexor system, housing for the analyzer in the field, and chambers, excluding the labor and expertise for deployment. Importantly, real-time systems constrain the experimental footprint to generally within tens of meters from the analyzer and within range of a reliable power source. The increased statistical power from higher chamber replication possible when using the ABSS is likely to provide much greater benefits compared to the uncertainty introduced by sample bulking and two-point flux estimation. Chadwick et al. ( 2014 ) found a 10-fold reduction in error as chamber replication increased from two to five under large flux conditions in nitrogen fertilization rate experiments across the UK. The authors concluded that spatial variability between chambers generally introduces much greater uncertainty than assumptions of linearity in flux calculations, suggesting a beneficial tradeoff between chamber number versus number of headspace samples (Chadwick et al., 2014 ). Replication is particularly important for generating the statistical power needed to detect treatment differences, necessary for robustly assessing the impact of management interventions on N 2 O emissions. As ABSS chambers may deployed at several times the density of a real-time analyzer system for the same budget, the benefit to statistical power likely far outweighs the inherent temporal and sample pooling constraints. 5. Conclusion The Automated Bulk Sampling System (ABSS) is a low-cost, remotely deployable trace gas measurement system that has advantages over current technologies, including high-resolution sampling and low-cost deployment. Our results demonstrated the ability for the ABSS system to capture high-precision estimation of soil N 2 O flux in agricultural systems. While our validation experiments highlight some inherent biases compared with real-time analyzer systems, the high precision of the system demonstrates high potential for capturing treatment effects in experiments or relative differences in flux along environmental gradients. Additional system modifications, including increased flushing of the sampling tubes, may improve accuracy, and correction factors may also be developed. Deployable technologies like the ABSS have the potential to rapidly augment collection of N 2 O flux data, facilitating new insights into the management of soil N 2 O production in agricultural systems. Declarations Consent to Publish declaration Not applicable. Ethics and Consent to Participate declarations: Not applicable. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Systems designs are also available upon request. Competing Interests The authors declare that they have no competing interests Funding This work was supported by Corteva Agriscience. Authors’ contributions Randy Clark : Conceptualization, Methodology, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. Nick Friedenberg : Methodology, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. Dan Chamberlain : Conceptualization, Methodology, Investigation. Data curation. Chris Parry: Methodology, Investigation. Timothy Hart: Methodology, Investigation, Data curation. Jessica Garcia: Data Curation, Formal analysis. Julie Abendroth: Project administration, Writing – review & editing. Courtland Kelly : Project administration, Visualization, Writing – original draft, Writing – reviewing and editing Acknowledgements The authors would like to thank Jennifer Soong for providing scientific leadership during the early phases of the research; Robert Hall for providing technical expertise during the methodology discussions and validation studies; Don McDonald, Alex Steurrys, Shane Rich, Brandon Maurer and other members of the Corteva engineering team for their guidance on system designs and troubleshooting; and John Arbuckle for his leadership, support and vision towards research innovation. References Ambus P, Skiba U, Drewer J, Jones SK, Carter MS, Albert KR, Sutton MA. Development of an accumulation-based system for cost‐effective chamber measurements of inert trace gas fluxes. Eur J Soil Sci. 2010;61(5):785–92. https://doi.org/10.1111/j.1365-2389.2010.01272.x . Arias-Navarro C, Díaz-Pinés E, Kiese R, Rosenstock TS, Rufino MC, Stern D, Neufeldt H, Verchot LV, Butterbach-Bahl K. Gas pooling: A sampling technique to overcome spatial heterogeneity of soil carbon dioxide and nitrous oxide fluxes. Soil Biol Biochem. 2013;67:20–3. https://doi.org/10.1016/j.soilbio.2013.08.011 . Butterbach-Bahl K, Baggs EM, Dannenmann M, Kiese R, Zechmeister-Boltenstern S. Nitrous oxide emissions from soils: how well do we understand the processes and their controls? Philosophical Trans Royal Soc B-Biological Sci. 2013;368(1621). https://doi.org/10.1098/rstb.2013.0122 . Chadwick DR, Cardenas L, Misselbrook TH, Smith KA, Rees RM, Watson CJ, McGeough KL, Williams JR, Cloy JM, Thorman RE, Dhanoa MS. Optimizing chamber methods for measuring nitrous oxide emissions from plot-based agricultural experiments. Eur J Soil Sci. 2014;65(2):295–307. https://doi.org/10.1111/ejss.12117 . Charles A, Rochette P, Whalen JK, Angers DA, Chantigny MH, Bertrand N. Global nitrous oxide emission factors from agricultural soils after addition of organic amendments: A meta-analysis. Agric Ecosyst Environ. 2017;236:88–98. https://doi.org/10.1016/j.agee.2016.11.021 . Charteris AF, Chadwick DR, Thorman RE, Vallejo A, de Klein CAM, Rochette P, Cárdenas LM. Global Research Alliance N2O chamber methodology guidelines: Recommendations for deployment and accounting for sources of variability. J Environ Qual. 2020;49(5):1092–109. https://doi.org/10.1002/jeq2.20126 . Davis MP, Groh TA, Parkin TB, Williams RJ, Isenhart TM, Hofmockel KS. Portable Automation of Static Chamber Sample Collection for Quantifying Soil Gas Flux. J Environ Qual. 2018;47(2):270–5. https://doi.org/10.2134/jeq2017.10.0387 . Della Chiesa T, Northrup D, Miguez FE, Archontoulis SV, Baum ME, Venterea RT, Emmett BD, Malone RW, Iqbal J, Necpalova M, Castellano MJ. Reducing greenhouse gas emissions from North American soybean production. Nat Sustain. 2024;7(12):1608–15. https://doi.org/10.1038/s41893-024-01458-9 . Ferrari Machado PV, Wagner-Riddle C, MacTavish R, Voroney PR, Bruulsema TW. Diurnal Variation and Sampling Frequency Effects on Nitrous Oxide Emissions Following Nitrogen Fertilization and Spring-Thaw Events. Soil Sci Soc Am J. 2019;83(3):743–50. https://doi.org/10.2136/sssaj2018.10.0365 . Fuchs K, Merbold L, Buchmann N, Bretscher D, Brilli L, Fitton N, Topp CFE, Klumpp K, Lieffering M, Martin R, Newton PCD, Rees RM, Rolinski S, Smith P, Snow V. Multimodel Evaluation of Nitrous Oxide Emissions From an Intensively Managed Grassland. J Geophys Research-Biogeosciences. 2020;125(1). https://doi.org/https://doi.org/10.1029/2019JG005261 . IBM. (2024). IBM Environmental Intelligence Suite . In IBM Corporation. https://www.ibm.com/products/environmental-intelligence-suite Lawrence NC, Hall SJ. Capturing temporal heterogeneity in soil nitrous oxide fluxes with a robust and low-cost automated chamber apparatus. Atmos Meas Tech. 2020;13(7):4065–78. https://doi.org/10.5194/amt-13-4065-2020 . Maier M, Weber TKD, Fiedler J, Fuß R, Glatzel S, Huth V, Jordan S, Jurasinski G, Kutzbach L, Schäfer K, Weymann D, Hagemann U. Introduction of a guideline for measurements of greenhouse gas fluxes from soils using non-steady‐state chambers. J Plant Nutr Soil Sci. 2022;185(4):447–61. https://doi.org/10.1002/jpln.202200199 . Mbow HOP, Reisinger A, Canadell J, O’Brien P. (2017). Special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems (SR2). Olander LP, Wollenberg E, Tubiello FN, Herold M. Synthesis and Review: Advancing agricultural greenhouse gas quantification. Environ Res Lett. 2014;9(7). https://doi.org/https://doi.org/10.1088/1748-9326/9/7/075003 . Parkin TB, Venterea RT, Hargreaves SK. Calculating the Detection Limits of Chamber-based Soil Greenhouse Gas Flux Measurements. J Environ Qual. 2012;41(3):705–15. https://doi.org/10.2134/jeq2011.0394 . Reay DS, Davidson EA, Smith KA, Smith P, Melillo JM, Dentener F, Crutzen PJ. Global agriculture and nitrous oxide emissions. Nat Clim Change. 2012;2(6):410–6. https://doi.org/10.1038/Nclimate1458 . Saha D, Basso B, Robertson GP. Machine learning improves predictions of agricultural nitrous oxide (N2O) emissions from intensively managed cropping systems. Environ Res Lett. 2021;16(2). https://doi.org/10.1088/1748-9326/abd2f3 . Seabold S, Perktold J. (2010). stastmodels: Econometric and statistical modeling with Python . In Proceeding of the 9th Python in Science Conference https://github.com/statsmodels/statsmodels/ Storer K, Coggan A, Ineson P, Hodge A. Arbuscular mycorrhizal fungi reduce nitrous oxide emissions from N2O hotspots. New Phytol. 2018;220(4):1285–95. https://doi.org/10.1111/nph.14931 . Tian HQ, Xu RT, Canadell JG, Thompson RL, Winiwarter W, Suntharalingam P, Davidson EA, Ciais P, Jackson RB, Janssens-Maenhout G, Prather MJ, Regnier P, Pan NQ, Pan SF, Peters GP, Shi H, Tubiello FN, Zaehle S, Zhou F, Yao YZ. A comprehensive quantification of global nitrous oxide sources and sinks. Nature. 2020;586(7828):248–. https://doi.org/10.1038/s41586-020-2780-0 . Venterea RT, Petersen SO, de Klein CAM, Pedersen AR, Noble ADL, Rees RM, Gamble JD, Parkin TB. Global Research Alliance N(2) O chamber methodology guidelines: Flux calculations. J Environ Qual. 2020;49(5):1141–55. https://doi.org/10.1002/jeq2.20118 . Wagner SW, Reicosky DC, Alessi RS. Regression models for calculating gas fluxes measured with a closed chamber. Agron J. 1997;89(2):279–84. https://doi.org/DOI 10.2134/agronj1997.00021962008900020021x . Wagner-Riddle C, Baggs EM, Clough TJ, Fuchs K, Petersen SO. Mitigation of nitrous oxide emissions in the context of nitrogen loss reduction from agroecosystems: managing hot spots and hot moments. Curr Opin Environ Sustain. 2020;47:46–53. https://doi.org/10.1016/j.cosust.2020.08.002 . Additional Declarations No competing interests reported. Supplementary Files SupplPlantMethods.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 18 May, 2026 Reviewers agreed at journal 18 May, 2026 Reviewers agreed at journal 18 May, 2026 Reviews received at journal 16 Mar, 2026 Reviewers agreed at journal 09 Mar, 2026 Reviewers invited by journal 08 Mar, 2026 Editor assigned by journal 07 Mar, 2026 Submission checks completed at journal 07 Mar, 2026 First submitted to journal 02 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9013846","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Method Article","associatedPublications":[],"authors":[{"id":604459008,"identity":"bdc69f7e-9b57-4798-8214-3de5a8f6278c","order_by":0,"name":"Randy Clark","email":"","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":false,"prefix":"","firstName":"Randy","middleName":"","lastName":"Clark","suffix":""},{"id":604459009,"identity":"de0af0f3-98f6-4104-86d0-1f839870554e","order_by":1,"name":"Nick Friedenberg","email":"","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":false,"prefix":"","firstName":"Nick","middleName":"","lastName":"Friedenberg","suffix":""},{"id":604459010,"identity":"f5780613-884d-478e-804e-f375532007ef","order_by":2,"name":"Dan Chamberlain","email":"","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":false,"prefix":"","firstName":"Dan","middleName":"","lastName":"Chamberlain","suffix":""},{"id":604459011,"identity":"efc47b76-5800-4e3e-a30a-37d4f27550ff","order_by":3,"name":"Chris Parry","email":"","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":false,"prefix":"","firstName":"Chris","middleName":"","lastName":"Parry","suffix":""},{"id":604459012,"identity":"2358066a-23dc-419e-a2d5-6fe4fea04cbb","order_by":4,"name":"Timothy Hart","email":"","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":false,"prefix":"","firstName":"Timothy","middleName":"","lastName":"Hart","suffix":""},{"id":604459013,"identity":"54dc9e41-9a73-4a86-8bcd-b92b6db99324","order_by":5,"name":"Jessica Garcia","email":"","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":false,"prefix":"","firstName":"Jessica","middleName":"","lastName":"Garcia","suffix":""},{"id":604459014,"identity":"9f4db469-e004-40aa-8179-95f8fc662332","order_by":6,"name":"Julie Abendroth","email":"","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":false,"prefix":"","firstName":"Julie","middleName":"","lastName":"Abendroth","suffix":""},{"id":604459015,"identity":"7bdab615-d4c8-4750-ab0d-0dbe92de6d82","order_by":7,"name":"Courtland Kelly","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBUlEQVRIiWNgGAWjYNCCAiA+wMbA8KCAQY44HQcMoFoSDBiMgUzGBpK0JDYQ0sLf3mP2+YOBXR7fAbbEBwkGh9P723ufP/jYZgeU6k7ApkXizBnjGQcMkoslD7AdNgBqyZ1x5rhh48y2ZKDU2Q3YtBhI5BgDHcacuOEAe5sESMsGiTTGZp4zzECpXHxa6kFa2n+AHGYA0vLnTD0hLYeBWtiOAb1/OAGshaHiME4tEmeOFTOcMTieOPMwWzLQYemGM84cY5zZU3GcB5df+NubNzNUVFQn9h1vM/zwocJanr+9jeHDD4NqOf72XqxaEIAZTDbD+Tz4lSNAHbEKR8EoGAWjYAQBAI+WZKueAMkHAAAAAElFTkSuQmCC","orcid":"","institution":"Corteva Agriscience","correspondingAuthor":true,"prefix":"","firstName":"Courtland","middleName":"","lastName":"Kelly","suffix":""}],"badges":[],"createdAt":"2026-03-02 21:38:04","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9013846/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9013846/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104468186,"identity":"6df7396f-a8f5-4785-83c8-3f004e72fbdb","added_by":"auto","created_at":"2026-03-12 06:46:16","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":75783,"visible":true,"origin":"","legend":"\u003cp\u003eImages of the ABSS soil N\u003csub\u003e2\u003c/sub\u003eO collection system deployed in a field setting. Each automated chamber is attached to one enclosure system and is powered with a solar panel. \u0026nbsp;\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9013846/v1/505b871d0931c21a14fff3b8.png"},{"id":104468184,"identity":"4f1d51e0-15ea-4d69-9dff-85c23a20f9f0","added_by":"auto","created_at":"2026-03-12 06:46:16","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":99969,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between N\u003csub\u003e2\u003c/sub\u003eO gas concentration readings from ABSS and real-time data in laboratory dosing experiments. Open circles are open-chamber readings; closed circles are measurements taken after N\u003csub\u003e2\u003c/sub\u003eO was dosed into closed chambers. Solid 1:1 line in main figure indicates perfect alignment, dotted line is the robust linear regression (r\u003csup\u003e2\u003c/sup\u003e = 0.998). Inset: the percent error of the ABSS system as function of direct measurements. \u0026nbsp;The solid line indicates 0% error and dotted line is robust linear regression (intercept: 3.5 ± 1.5 ppb, z = 0.6, p \u0026lt; 0.0001; slope: -0.009 ± 0.002, z = -9.4, p \u0026lt; 0.0001).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9013846/v1/290548368053180f652bc5e6.png"},{"id":104780601,"identity":"bd2fa07b-a2c9-4b89-a151-ef80cef619af","added_by":"auto","created_at":"2026-03-17 07:53:20","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":101507,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of ABSS and real-time analyzer results for in-field validation experiments occurring in two different 2-week measurement periods. The different measurement periods are indicated by the point shape. The solid grey line represents a 1:1 relationship and the black line is the linear regression result with equation as indicated (r\u003csup\u003e2\u003c/sup\u003e\u0026nbsp;= 0.79). The red shaded area is the portion of underestimation attributable to two-point sampling and sample bulking, derived from our synthetic “expected” bulk dataset, and the grey shaded area the additional underestimation due to overall system bias.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9013846/v1/17057f07149386661ec8cef4.png"},{"id":104468187,"identity":"ca063365-bb08-4c7e-a425-226e22028ea8","added_by":"auto","created_at":"2026-03-12 06:46:16","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":782962,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of open-chamber N\u003csub\u003e2\u003c/sub\u003eO concentration measurements (open points) and 10-minute closed-chamber concentrations (black points) between measured ABSS samples and “expected” bulk concentrations in our field experiment. Point shape (circle or square) signifies the 2-week sampling period. The grey solid line is the 1:1 line of perfect fit.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-9013846/v1/9c498fb4b5a887ad795d33a8.png"},{"id":104808469,"identity":"741bd9a4-097c-4a3e-a75a-ca61955c70b5","added_by":"auto","created_at":"2026-03-17 12:37:50","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2220669,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9013846/v1/6627df1e-98d3-4d2e-86f7-e9c1da0ce4b7.pdf"},{"id":104468188,"identity":"d93e5159-4312-4ea4-a78e-0d91f6c52054","added_by":"auto","created_at":"2026-03-12 06:46:16","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1700255,"visible":true,"origin":"","legend":"","description":"","filename":"SupplPlantMethods.docx","url":"https://assets-eu.researchsquare.com/files/rs-9013846/v1/5bdeebf94199c3a60d33e095.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"The Automated Bulk Sampling System (ABSS), a low-cost solution for integrated nitrous oxide emission quantification in field studies","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eNitrous oxide (N\u003csub\u003e2\u003c/sub\u003eO) is a potent greenhouse gas (GHG) and is the largest direct component of on-farm GHG emissions in row-crop agriculture (Mbow et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Reay et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Tian et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Emissions of N\u003csub\u003e2\u003c/sub\u003eO from agricultural soils are a byproduct of nitrification and denitrification by native soil microbes and are driven by environmental and management factors including temperature, water, oxygen, pH, substrate availability, and energy resources (Butterbach-Bahl et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Considering the large role of agriculture in anthropogenic N\u003csub\u003e2\u003c/sub\u003eO emissions, there is a critical need to quantify management impacts on soil N\u003csub\u003e2\u003c/sub\u003eO emissions (Olander et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite decades of research, mechanistic models of N\u003csub\u003e2\u003c/sub\u003eO flux often struggle to replicate empirical measurements, emphasizing gaps in our understanding that require additional data collection efforts to address (Fuchs et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Reay et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). A major challenge is the high spatiotemporal variation in N\u003csub\u003e2\u003c/sub\u003eO emissions influenced by micro-climate and micro-site factors, often called \u0026ldquo;hot spots\u0026rdquo; and \u0026ldquo;hot moments\u0026rdquo; (Wagner-Riddle et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Relatively infrequent and low-density gas flux measurements often miss key locations or events in the flux landscape, leaving the links between agricultural management with N\u003csub\u003e2\u003c/sub\u003eO emissions uncertain (Charteris et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Some agricultural practices have relatively well-known effects on emissions (i.e. fertilizer reduction, the use of nitrification inhibitors or other enhanced efficiency fertilizers), but other potentially mitigative factors such as tillage, cover crop use, biological and organic amendments, and cropping system and crop variety are less understood (Charles et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Della Chiesa et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Olander et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Storer et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe large cost, complexity, and effort required for N\u003csub\u003e2\u003c/sub\u003eO data collection limits the frequency and replication of measurements, while also creating both resource and technical barriers to understanding agriculture management impacts on GHG emissions. Chamber headspace sampling methods are the primary approach to measuring N\u003csub\u003e2\u003c/sub\u003eO and other GHG emissions from plot-scale agricultural experiments. Manual non-steady-state (i.e. closed) chamber collection methods are the most common and inexpensive and are generally the most time-consuming and laborious. In these approaches, chambers are installed and sampled multiple times while fully closed for a period, usually less than an hour, then the samples are transported back to a laboratory for analysis (Charteris et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Portable, \u0026ldquo;fast analyzers\u0026rdquo; may also be employed to measure gas fluxes directly in the field by transporting a flow-through analyzer to each collar site manually, which saves some time and effort (Maier et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These manually measured chamber site locations are typically visited at weekly to monthly intervals and allow for flexibility in spatial layouts (Charteris et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and the ability to measure multiple plots or locations with a single portable chamber.\u003c/p\u003e \u003cp\u003eThe use of automated chamber methods, by contrast, provides greater temporal resolution through higher-frequency measurements throughout the experimental period. These methods either use automatic samplers, which store individual samples in separate vials, or utilize real-time gas analyzer systems to record trace gas concentrations at the time of sampling with no lab analysis required (Davis et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Lawrence \u0026amp; Hall, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Both automated chamber approaches have drawbacks, however. Automatic samplers may offer greater flexibility for deployment and scaling, but they can generate a cumbersome number of samples, leading to relatively high labor and analysis costs. Real-time analyzers coupled to automated chambers are costly and require significant expertise to deploy, power, and maintain, while also limiting the experimental footprint and the number of chambers that can be monitored.\u003c/p\u003e \u003cp\u003eBulk sampling of gas, wherein multiple samples are accumulated into a single container over time, provides an intermediate solution, leading to a reduced lab analysis workload while capturing cumulative emissions with complete, high-frequency temporal coverage (Ambus et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Like automatic samplers, bulk sampling minimizes the in-field labor requirements and separates the sampling apparatus from the more expensive analytical instrument, allowing scalability and flexibility of deployment. An example, the SIGMA bulk sampler (Ambus et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), was used in a 2009 field experiment (Juszczak and Augustin, 2013) but has not seen wider use, perhaps due to cost or lack of commercial models. Sample pooling has also been applied spatially, where gas samples from multiple chambers are pooled to account for spatial heterogeneity and reduce analytical effort (Arias-Navarro et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). With relatively simple engineering, bulk sampling systems could provide an affordable and flexible route to broader N\u003csub\u003e2\u003c/sub\u003eO mitigation investigations. However, potential biases and sources of uncertainty must be properly considered.\u003c/p\u003e \u003cp\u003eIn consideration of the need for low-cost, scalable N\u003csub\u003e2\u003c/sub\u003eO measurement systems and the potential of the \u0026ldquo;sample bulking\u0026rdquo; approach, we sought to design and develop a low-cost bulk sampling system to measure N\u003csub\u003e2\u003c/sub\u003eO (and other trace GHGs) in remote field settings compatible with agricultural small plot and on-farm experimental designs. Our second objective was to validate this system and assess sources of measurement uncertainty, including systematic and mechanical contributions, as well as those introduced by natural variation in atmospheric conditions over the sample collection period. To address the first objective, we designed, fabricated, and validated the Automated Bulk Sampling System (ABSS). The system collects and composites high-frequency samples of chamber headspace gas into air-tight gas collection bags that can be retrieved, replaced, and analyzed every two weeks. To validate the system and the general sample bulking approach, we conducted lab dosing experiments and a field test in corn (\u003cem\u003eZea mays\u003c/em\u003e L.) in Iowa, USA, where we directly compared soil N\u003csub\u003e2\u003c/sub\u003eO flux estimates generated from the ABSS system with fluxes measured with a real-time in-field analyzer. We investigated different sources of error within our field-generated dataset, including estimating data loss due to the \u0026ldquo;bulking\u0026rdquo; approach. We also used historic meteorological data to estimate the likely error contribution from weather variation.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Design and fabrication\u003c/h2\u003e \u003cp\u003eThe ABSS system was designed and fabricated using low-cost materials and is powered by a solar panel and battery for remote deployment. It consists of cylindrical chamber with an automatic, air-tight lid connected to an insulated enclosure housing the electrical components, valves, and the two gas sample collections bags (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The system collects specific gas volume samples from the headspace of an automated chamber at pre-defined periods (i.e. hourly for two weeks) from two different chamber closure states: open-chamber (OC) and closed-chamber (CC). All samples are combined into single gas sample bags for each closure state. Details on the chamber components, construction and design are described in the Supplemental Methods.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Lab validation\u003c/h2\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.2.1 Lab experimental setup\u003c/h2\u003e \u003cp\u003eOur initial lab validation experiments used gas injection to simulate soil fluxes and compared composite N\u003csub\u003e2\u003c/sub\u003eO measurements from the ABSS with concentration measurements taken in real-time from automatic chambers. In this approach, we focused on comparing direct N\u003csub\u003e2\u003c/sub\u003eO concentrations of open-chamber air and closed-chamber air that was injected with N\u003csub\u003e2\u003c/sub\u003eO calibration gas, as these two values underpin the estimation of flux. A set of four automated chambers were programmed to cycle through open-chamber and closed-chamber states 336 times over 2.8 days. A set amount of N\u003csub\u003e2\u003c/sub\u003eO calibration gas (Gasco Affiliates LLC, Oldsmar, FL) was injected into the chamber headspace during the closed-chamber state to simulate a target flux rate. Individual experiments were run with consistent, non-varying target flux rates, as well as with \u0026ldquo;spikes\u0026rdquo; representative of short term, higher flux moments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2 Real-time analyzer measurements in lab validation\u003c/h2\u003e \u003cp\u003eAll automated chambers were connected in parallel to a Picarro G2580, with a valve controller system that allowed each individual chamber to be measured separately. The analyzer system cycled through the four chambers so that the real-time measurements were taken for 80 seconds with 10 seconds of ambient air reading between chambers. The chamber closure timing was staggered so that the real-time analyzer cycled through all four chambers in an open state and then again when all four chambers were in a closed state at the same phase of their cycle. The chamber headspace concentration reading was taken as the average the last 30 seconds of readings from the 80 second measurement time. The average concentration over all the cycles in the experiment was then calculated to compare to the ABSS bulk measurement, described below.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.2.3 ABSS measurement in lab validation\u003c/h2\u003e \u003cp\u003eEach automated chamber was also connected to an ABSS gas sampling system so that headspace gas samples were collected during the open-chamber and closed-chamber phases of the cycles. The ABSS gas samples were synchronized to be collected in the last 10 seconds of the real-time analyzer measurement time. All samples taken during the open-chamber phase of the cycles were collected in the same open-chamber sample bag, while all samples taken from the closed-chamber phase were collected into the same closed-chamber bag, resulting in two sample bags at the end of each experiment. After completion, the bag N\u003csub\u003e2\u003c/sub\u003eO concentrations were measured on the same Picarro G2580 used for the real-time measurements. See Figure S4 for experimental set-up and example data plots.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.2.4 Statistical Analysis for Lab Validation\u003c/h2\u003e \u003cp\u003eThe agreement between ABSS bulk sample concentrations and those measured by the real-time analyzer were assessed by direct regression to assess overall precision (r\u003csup\u003e2\u003c/sup\u003e) and bias (deviation from the 1:1 line). We further explored the bias by regressing the difference (or error) in concentration readings from the two methods against the real-time readings. Analysis of the specific concentrations from the open-chamber and closed-chambers phases, instead of the overall concentration change (flux), allowed us to better examine sources of error between the two measurement approaches. The distribution of gas concentrations investigated was non-normally distributed. Analysis employed iteratively weighted robust regression to reduce the influence of high measured N\u003csub\u003e2\u003c/sub\u003eO fluxes on the fit while preserving interpretation in natural rather transformed units. All analyses were performed using the \u003cem\u003estatsmodels\u003c/em\u003e v 0.14.4 Python package and Python 3.11 (Seabold \u0026amp; Perktold, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Weights were obtained from the \u003cem\u003erlm\u003c/em\u003e (robust linear model) function and used as inputs to the \u003cem\u003ewls\u003c/em\u003e (weighted least squares) function to obtain traditional model summary statistics.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Field Validation Experiment\u003c/h2\u003e \u003cp\u003eThe robustness of the ABSS and 2-point bulk sampling method under field conditions was empirically tested during the spring and summer of 2023. In this evaluation, the average soil N\u003csub\u003e2\u003c/sub\u003eO fluxes captured by the ABSS were compared to average soil N\u003csub\u003e2\u003c/sub\u003eO flux computed from averaging real-time flux measurements taken over 2-week sample collection periods.\u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 Field Experiment Design\u003c/h2\u003e \u003cp\u003eSoil gas measurements were collected from a corn field in Johnston, IA at the Corteva Agriscience Research Farm from September to October 2023. Average temperature during the sampling period was 18.9\u0026deg;C and rainfall totaled 6.2 cm in September. The soil is classified as a Wiota silty clay loam.\u003c/p\u003e \u003cp\u003eAirtight PVC chambers with automated lids were mounted on top of stainless-steel collars that were inserted 10 cm into the soil and secured with gaffer\u0026rsquo;s tape. Once installed, the inside dimensions of the chamber headspace above the soil measured 30 cm tall by 25.45 cm diameter.\u003c/p\u003e \u003cp\u003eField validation testing consisted of two separate two-week sample collection periods, where 8 ABSS were connected to separate GHG chambers that were part of an ongoing 12 chamber real-time N\u003csub\u003e2\u003c/sub\u003eO measurement experiment. In this design, ABSS gas samples were collected from the same chambers that were being monitored by a real-time N\u003csub\u003e2\u003c/sub\u003eO analyzer system.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 Real-Time Analyzer Measurements in Field Validation\u003c/h2\u003e \u003cp\u003eReal-time field measurements of N\u003csub\u003e2\u003c/sub\u003eO concentration in chamber headspaces were made via a custom multiplex valve system that allowed each chamber to be isolated in a closed loop with an LGR-ICOS\u0026trade; GLA151- N\u003csub\u003e2\u003c/sub\u003eO M1 portable N\u003csub\u003e2\u003c/sub\u003eO /CH\u003csub\u003e4\u003c/sub\u003e analyzer (LGR/ABB Inc., Zurich, Switzerland). At the beginning of each measurement cycle, all 12 automated chamber lids closed simultaneously. The real-time analyzer system then cycled through each chamber, measuring headspace gas for 76 seconds each time, and cycling through all chambers 4 times before all chambers opened again. As a result, each chamber has four headspace measurements, but all at different times within the chamber closure period, which lasted about 60 minutes. This closed period was followed by a 60-minute open chamber period to allow the concentrated gases in the chamber headspaces to vent and equilibrate with the surrounding environment.\u003c/p\u003e \u003cp\u003eThe rate of chamber concentration change was calculated by regression using best-fit model selection between quadratic or linear models. Quadratic fits are a simple and performant way to capture possible concentration saturation over the course of a chamber measurement (Parkin et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Wagner et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). The quadratic model was rejected when its quadratic term was positive, when the model was not statistically significant at the \u003cem\u003eα\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.05 level, or when the quadratic model had a Bayesian Information Criterion (BIC) score less than 2 units lower than that of a linear fit. The rate of concentration change (ppm/min), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\frac{dC}{dt}\\)\u003c/span\u003e\u003c/span\u003e, was taken as the linear term of the selected model or 0 if neither model was statistically significant, equivalent to the first derivative of the regression model evaluated at t\u0026thinsp;=\u0026thinsp;0.\u003c/p\u003e \u003cp\u003eFlux of N\u003csub\u003e2\u003c/sub\u003eO was estimated using air pressure, \u003cem\u003eP\u003c/em\u003e (atmospheres), and temperature, \u003cem\u003eT\u003c/em\u003e (degrees Kelvin) from on-site weather station with 15-minute resolution, using the measurement most coincident with chamber closure time. The formula for flux, \u003cem\u003eF\u003c/em\u003e, estimated from real-time analyzer data was\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$F=\\frac{dC}{dt}\\frac{P}{T}\\frac{V{M}_{N}}{RA}\\times60\\text{min/h}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e,\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\frac{dC}{dt}\\)\u003c/span\u003e\u003c/span\u003e is the rate of concentration change (ppm/min) extracted from the best-fit regression line, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\frac{P}{V}\\)\u003c/span\u003e\u003c/span\u003e is the air density coefficient extracted from the weather station, \u003cem\u003eV\u003c/em\u003e is the headspace volume of the GHG chamber (L), \u003cem\u003eM\u003c/em\u003e\u003csub\u003e\u003cem\u003eN\u003c/em\u003e\u003c/sub\u003e is the molecular weight of N\u003csub\u003e2\u003c/sub\u003eO-N (28 \u003cem\u003e\u0026micro;\u003c/em\u003eg/\u003cem\u003e\u0026micro;\u003c/em\u003emol), \u003cem\u003eR\u003c/em\u003e is the ideal gas constant (0.082057 atm L mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e K\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), and \u003cem\u003eA\u003c/em\u003e is soil surface area inside the GHG chamber (m\u003csup\u003e2\u003c/sup\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e2.3.3 ABSS Measurements in Field Validation\u003c/h2\u003e \u003cp\u003eSimilar to the laboratory validation experiment, ABSS system headspace samples were collected from the same chambers as the real-time analyzer. However, while real-time analyzer measurements were taken at varying times in the chamber closure cycle, all ABSS samples were coordinated to be taken simultaneously to correspond with the synchronized chamber closure. Thus, all ABSS closed-chamber samples were collected 10 minutes after chamber closure to avoid saturation effects. The open-chamber samples were collected at the end of the 60-minute open-chamber phase, just before chamber closure. A timeline for the 4-point real-time measurement process and synchronized ABSS samplings is outlined in Supplemental Table ST2.\u003c/p\u003e \u003cp\u003eAfter each two-week cycle, the two bags from each ABSS unit were retrieved and transported to the lab, where the N\u003csub\u003e2\u003c/sub\u003eO concentration in each bag was measured with Picarro G2508 (Picarro, Inc., Santa Clara, CA, USA). Average hourly N\u003csub\u003e2\u003c/sub\u003eO flux over the two-week measurement period was calculated using the equation:\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$F=({C}_{cc}-{C}_{oc})\\stackrel{-}{\\left(\\frac{P}{T}\\right)}\\frac{V{M}_{N}}{RA}\\times\\frac{60\\text{min/h}}{D}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({C}_{cc}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({C}_{oc}\\)\u003c/span\u003e\u003c/span\u003e are the concentrations of N\u003csub\u003e2\u003c/sub\u003eO in the closed-chamber and open-chamber composite sampling bags, respectively, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\stackrel{-}{\\left(\\frac{P}{T}\\right)}\\)\u003c/span\u003e\u003c/span\u003e is the mean ratio of atmospheric pressure to temperature (atm K\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) over the two-week sample collection period, \u003cem\u003eV\u003c/em\u003e is the headspace volume of the GHG chamber (L), \u003cem\u003eM\u003c/em\u003e\u003csub\u003e\u003cem\u003eN\u003c/em\u003e\u003c/sub\u003e is the molecular weight of N\u003csub\u003e2\u003c/sub\u003eO-N (28 \u003cem\u003e\u0026micro;\u003c/em\u003eg/\u003cem\u003e\u0026micro;\u003c/em\u003emol), \u003cem\u003eR\u003c/em\u003e is the ideal gas constant (0.082057 atm L mole\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e K\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), \u003cem\u003eA\u003c/em\u003e is soil surface area inside the GHG chamber (m\u003csup\u003e2\u003c/sup\u003e), and \u003cem\u003eD\u003c/em\u003e is the closed chamber duration between the OC and CC samplings (10 minutes).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003e2.3.4 Estimating \u0026ldquo;expected\u0026rdquo; bulk concentrations from real-time N\u003csub\u003e2\u003c/sub\u003eO flux data\u003c/h2\u003e \u003cp\u003eIn addition to directly comparing the flux estimation from the real-time vs. the ABSS methods, we wanted to investigate the specific potential sources of discrepancy between the two methods. The potential sources were 1) differences in sampling times, 2) two- vs. four-point flux estimation, and 3) the loss of specific meteorological data at the time of sample collection. Each of these sources and our correction approach is described below. We examined each of these sources of error by generating a synthetic \u0026ldquo;expected\u0026rdquo; bulk sample dataset derived from the real-time data.\u003c/p\u003e \u003cp\u003eThe ABSS collected ambient samples before chamber closure and concentrated samples 10 minutes after closure, while the real-time samples were taken at 15-minute intervals starting at various times after chamber closure. We addressed this misalignment of sampling times by estimating real-time N\u003csub\u003e2\u003c/sub\u003eO concentrations at 0 and 10 minutes using the regression models from which we had obtained \u003cem\u003edC/dt\u003c/em\u003e. This approach also allowed us to estimate the inherent error introduce solely from the ABSS\u0026rsquo;s two-point sampling approach. Finally, we used air density data to account for variation in the number of molecules added to the bulk bags with each sample, assuming a constant sample volume. The resulting \u0026ldquo;expected\u0026rdquo; bulk concentration dataset derived from the real-time data was then used to investigate the contribution of each potential source of error inherent in the bulk sampling approach.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e2.3.4 Statistical Analysis for Field Validation\u003c/h2\u003e \u003cp\u003eHourly flux from each chamber measurement system was compared using ordinary least squares regression (\u003cem\u003eols\u003c/em\u003e function in \u003cem\u003estatsmodels\u003c/em\u003e package). The precision of the measurements, i.e. how reliably a real-time flux estimated corresponded to a particular ABSS flux value, was interpreted as the r\u003csup\u003e2\u003c/sup\u003e value of the model. We estimated ABSS system bias as the slope of the regression line, specifically its deviation from a slope of 1.\u003c/p\u003e \u003cp\u003eWe compared open-chamber and closed-chamber bag concentration measurements from the ABSS system with the synthetic \u0026ldquo;expected\u0026rdquo; bulk concentrations using one-way ANOVA to estimate the error contribution directly from the bulking process. We assumed independence between groups, and so did not include a random chamber effect in the model.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Meteorological variation assessment\u003c/h2\u003e \u003cp\u003eFluctuations in atmospheric temperature and pressure cause air density variation across the 2-week sample collection periods and introduce unresolvable measurement error due to unknown gas sample density at collection time. To estimate the potential error contribution of air density variation, historic weather datasets were used, where two-week periods from 10 years of hourly historic weather data (IBM, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) was randomly sampled at 49 gridded locations across the US Corn Belt (Supplemental Figure SF6) to construct 800 hourly weather datasets per location. The historic weather datasets were then overlaid with flux measurements from the field validation experiment to generate 4000 simulated ABSS datasets per location. Comparisons of ABSS flux measurements with and without correction for the known air density at each sampling event were then used to evaluate air density contributions to bulk sampling measurement error.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Gas stability testing\u003c/h2\u003e \u003cp\u003eThe stability of the N\u003csub\u003e2\u003c/sub\u003eO gas concentrations in the gas collection bags over 2-week periods was evaluated by filling 3 bags with N\u003csub\u003e2\u003c/sub\u003eO concentrations in range with previously collected closed-chamber samples (~\u0026thinsp;0.5ppm). The filled and closed bags were then stored in the lab or greenhouse and the N\u003csub\u003e2\u003c/sub\u003eO gas concentrations were measured at 0, 1, 7 and 15 days after filling to observe any changes in the N\u003csub\u003e2\u003c/sub\u003eO concentrations over time.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Lab dosing experiment\u003c/h2\u003e \u003cp\u003eN\u003csub\u003e2\u003c/sub\u003eO concentration readings from samples collected by the ABSS showed high precision to those measured directly from the chambers (r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.998), and a slope not different from the 1:1 line (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eHowever, further scrutiny of deviation in ABSS concentration readings showed a statistically significant departure from direct chamber measurements, with an overestimation of N\u003csub\u003e2\u003c/sub\u003eO at low concentrations to underestimation at high concentrations (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e inset). These parameters suggest the ABSS on average overestimates typical ambient N\u003csub\u003e2\u003c/sub\u003eO concentrations (334 ppb) by 1.2 ppb in the lab while underestimating concentrations over 356 ppb, though the bias is small (~\u0026thinsp;0.9%). This inference was supported by a t-test of open-chamber readings, which shows a statistically significant 1.2 ppb higher average reading from ABSS samples than from the real-time gas analyzer (t\u003csub\u003e44\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;4.56, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, Supplemental Table ST3).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Field validation\u003c/h2\u003e \u003cp\u003eWe successfully captured paired 2-week flux data for 9 chambers over two different sampling periods. Most lost data came from malfunctions in the chamber lid control and custom multiplex valve system. Average N\u003csub\u003e2\u003c/sub\u003eO flux over the 2-week sampling period ranged from 41.9\u0026ndash;175.4 ug N\u003csub\u003e2\u003c/sub\u003eO-N m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e hr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe synthetic \u0026ldquo;expected\u0026rdquo; bulk sample dataset derived from the real-time measurement provided an idealized maximum accuracy of the ABSS, due to the lost meteorological information and the limitations of the two-sample flux estimation procedure. This \u0026ldquo;expected\u0026rdquo; bulk dataset underestimated the real-time flux by 6% (F\u003csub\u003e1,8\u003c/sub\u003e = 752, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.99, red shaded interval in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e; Supplemental Figure SF5a).\u003c/p\u003e \u003cp\u003eDirect comparison of the ABSS and real-time analyzer calculated hourly flux indicated an overall underestimation by the ABSS system across all flux levels of 24.7% (t = -5.9, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, 95% CI: 15.1\u0026ndash;34.3%, r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.79; black trend line in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This additional overall system bias (i.e. the additional grey shaded wedge in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) indicates further deviation of flux measurements in practice from the theoretical \u0026ldquo;best\u0026rdquo; (i.e. red shaded wedge). An initial regression of the difference between systems against direct measurements indicated the intercept was not significantly different from zero (p\u0026thinsp;=\u0026thinsp;0.99).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Sources of Field Validation Bias\u003c/h2\u003e \u003cp\u003eABSS flux estimates were biased by an apparent overestimation of open-chamber N\u003csub\u003e2\u003c/sub\u003eO concentrations. We found that the bulked ABSS open-chamber samples had N\u003csub\u003e2\u003c/sub\u003eO concentrations 15% higher than the \u0026ldquo;expected\u0026rdquo; concentrations, (mean difference: 5.6 ppb; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Closed-chamber measurements were underestimated by an average of 1.93 ppb, though this did not represent a significant difference between systems (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e illustrates the close agreement between observed and expected closed concentrations compared with a lack of correlation in open chamber readings.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eANOVA results assessing the difference in N\u003csub\u003e2\u003c/sub\u003eO concentration measurements between ABSS and \u0026ldquo;expected\u0026rdquo; bulk concentrations at t\u0026thinsp;=\u0026thinsp;0 (Open-Chamber) and t\u0026thinsp;=\u0026thinsp;10 minutes (Closed-Chamber; pooled standard error\u0026thinsp;=\u0026thinsp;1.68).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSource\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean Difference in ABSS vs. \u0026ldquo;Expected\u0026rdquo;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUnits\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e% Error\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOpen-Chamber N\u003csub\u003e2\u003c/sub\u003eO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eppb N\u003csub\u003e2\u003c/sub\u003eO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e14.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClosed-Chamber N\u003csub\u003e2\u003c/sub\u003eO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-1.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eppb N\u003csub\u003e2\u003c/sub\u003eO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.266\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-5.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Estimating air density variation contribution to error\u003c/h2\u003e \u003cp\u003eAir density across our analysis area and time window showed 2-week variations ranging from 0.041 to 0.26 kg m\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e, which corresponds to 0.2% \u0026minus;\u0026thinsp;1.1% error in flux rate (Supplemental Figure SF6, Panel C). Absolute simulated flux error increased with increasing flux, with average error ranging from 0.0004 to 0.0448 ug N\u003csub\u003e2\u003c/sub\u003eO-N m\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e h\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for low and high flux rates, respectively (Supplemental Figure SF6, Panel D).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Gas stability testing\u003c/h2\u003e \u003cp\u003eThe gas stability test showed no significant change in N\u003csub\u003e2\u003c/sub\u003eO concentration with a 95% confidence interval of concentration change ranging from \u0026minus;\u0026thinsp;5.006 to 7.074 ppb after a 15-day storage period (n\u0026thinsp;=\u0026thinsp;3; Supplemental Table ST4).\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Utility of ABSS\u003c/h2\u003e \u003cp\u003eThe results presented here demonstrate that the ABSS can provide consistent, intercomparable results compared to real time analyzers for integrated two-week trace N\u003csub\u003e2\u003c/sub\u003eO gas flux estimations. Our reported results suggest that as currently deployed, the ABSS may underestimate total emissions, but would provide an accurate comparison within a given study design. The ABSS approach is best suited for studies aiming to compare bi-weekly average or cumulative emissions in replicated field experiments, such as those looking to evaluate management impacts.\u003c/p\u003e \u003cp\u003eThe modular, solar-powered operation and low cost for deployment of the ABSS fills an important technological gap for taking high-frequency measurements in large-scale experiments and landscapes. Compared to manual static chamber sampling, hourly sampling over the entire collection period by the ABSS can reduce potential bias introduced by diurnal flux variation and low sampling frequency (Ferrari Machado et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Limitations of ABSS\u003c/h2\u003e \u003cp\u003eDeployment of the ABSS for N\u003csub\u003e2\u003c/sub\u003eO flux measurements presents some limitations, specifically regarding absolute flux estimation and detection of short-term emission spikes. Though the ABSS was able to capture 80% of the variability in average flux for our field validation experiment, the 25% flux underestimation is a concern. Our results are comparable to those reported by Ambus et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), who reported up to a 28% negative bias in their 3-bag SIGMA system compared with conventional static and real-time automated chamber methods. Investigation into the sources of bias identified the open-chamber N\u003csub\u003e2\u003c/sub\u003eO concentration as the largest source of error, which may be remedied through simple changes to the operational sequence i.e. longer flushing durations when clearing sampling tubes. Our approach of estimating \u0026ldquo;expected\u0026rdquo; bag concentrations using the real-time data also revealed a 6% error due to the sample bulking process, a relatively small but unavoidable contribution.\u003c/p\u003e \u003cp\u003eThe ABSS is best suited to research questions concerned with integrated or cumulative emissions, and within-study comparisons. As currently deployed, the system provides average flux estimates for two-week periods, and so would be unable to provide estimates at finer temporal resolution scales that may be desired for some modeling purposes. For example, most \u0026ldquo;hot moments\u0026rdquo; last hours or a few days (Wagner-Riddle et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), behaviors that would not be directly detectable within ABSS data, though would still be captured within the cumulative samples. Recent advances in biogeochemical modeling methods to predict N\u003csub\u003e2\u003c/sub\u003eO emissions depend on sub-daily flux measurements, though also utilize manual weekly chamber measurements for validation (Saha et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe version of ABSS presented here relied on a two-bag system for calculation of average flux. This approach was meant to deliver a low-cost, low-effort system that minimized labor and resources. In general, four gas samples per closure time is considered the \u0026ldquo;gold standard\u0026rdquo; for flux rate determination from static chamber approaches, which allows for more precise error calculation (Venterea et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). While saturation effects or nonlinearity of chamber concentration during the closure period often makes nonlinear flux calculation approaches more accurate, they are also more prone to random error and thus less precise, or repeatable, than basic linear calculation methods (Venterea et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In their assessment of chamber linearity assumptions from a variety of datasets, Chadwick et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) concluded that in most circumstances, an assumption of linearity and two or three headspace samples are sufficient, given a relatively short chamber closure period.\u003c/p\u003e \u003cp\u003eWhile the short chamber closure time helps to minimize potential saturation effects, this approach does require that the gas analyzer employed has appropriate sensitivity to detect the expected gas concentration differences between the sample bags. The ABSS chamber closure time can be modified depending on expected flux rates and gas analyzer detectability limits.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e4.3. Scalability and statistical power\u003c/h2\u003e \u003cp\u003eThe relatively low cost of ABSS provides benefits for increasing replication and overall statistical power. A 16-unit deployment of ABSS costs about \u003cspan\u003e$\u003c/span\u003e25,000, excluding the laboratory-based gas analyzer and labor for bag collection and analysis every two weeks. In contrast, the same number of chambers and the real-time analyzer system can cost over \u003cspan\u003e$\u003c/span\u003e250,000 when factoring in the dedicated gas analyzer (~\u003cspan\u003e$\u003c/span\u003e100,000), multiplexor system, housing for the analyzer in the field, and chambers, excluding the labor and expertise for deployment. Importantly, real-time systems constrain the experimental footprint to generally within tens of meters from the analyzer and within range of a reliable power source.\u003c/p\u003e \u003cp\u003eThe increased statistical power from higher chamber replication possible when using the ABSS is likely to provide much greater benefits compared to the uncertainty introduced by sample bulking and two-point flux estimation. Chadwick et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) found a 10-fold reduction in error as chamber replication increased from two to five under large flux conditions in nitrogen fertilization rate experiments across the UK. The authors concluded that spatial variability between chambers generally introduces much greater uncertainty than assumptions of linearity in flux calculations, suggesting a beneficial tradeoff between chamber number versus number of headspace samples (Chadwick et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Replication is particularly important for generating the statistical power needed to detect treatment differences, necessary for robustly assessing the impact of management interventions on N\u003csub\u003e2\u003c/sub\u003eO emissions. As ABSS chambers may deployed at several times the density of a real-time analyzer system for the same budget, the benefit to statistical power likely far outweighs the inherent temporal and sample pooling constraints.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThe Automated Bulk Sampling System (ABSS) is a low-cost, remotely deployable trace gas measurement system that has advantages over current technologies, including high-resolution sampling and low-cost deployment. Our results demonstrated the ability for the ABSS system to capture high-precision estimation of soil N\u003csub\u003e2\u003c/sub\u003eO flux in agricultural systems. While our validation experiments highlight some inherent biases compared with real-time analyzer systems, the high precision of the system demonstrates high potential for capturing treatment effects in experiments or relative differences in flux along environmental gradients. Additional system modifications, including increased flushing of the sampling tubes, may improve accuracy, and correction factors may also be developed. Deployable technologies like the ABSS have the potential to rapidly augment collection of N\u003csub\u003e2\u003c/sub\u003eO flux data, facilitating new insights into the management of soil N\u003csub\u003e2\u003c/sub\u003eO production in agricultural systems.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eConsent to Publish declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics and Consent to Participate declarations:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Systems designs are also available upon request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by Corteva Agriscience.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRandy Clark\u003c/strong\u003e: Conceptualization, Methodology, Formal analysis, Visualization, Writing – original draft, Writing – review \u0026amp; editing. \u0026nbsp;\u003cstrong\u003eNick Friedenberg\u003c/strong\u003e: Methodology, Formal analysis, Visualization, Writing – original draft, Writing – review \u0026amp; editing. \u003cstrong\u003eDan Chamberlain\u003c/strong\u003e: Conceptualization, Methodology, Investigation. Data curation. \u003cstrong\u003eChris Parry:\u003c/strong\u003e Methodology, Investigation. \u003cstrong\u003eTimothy Hart:\u003c/strong\u003e Methodology, Investigation, Data curation. \u0026nbsp;\u003cstrong\u003eJessica Garcia:\u003c/strong\u003e Data Curation, Formal analysis. \u0026nbsp;\u003cstrong\u003eJulie Abendroth:\u003c/strong\u003e Project administration, Writing – review \u0026amp; editing. \u003cstrong\u003eCourtland Kelly\u003c/strong\u003e: Project administration, Visualization, Writing – original draft, Writing – reviewing and editing\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors would like to thank Jennifer Soong for providing scientific leadership during the early phases of the research; Robert Hall for providing technical expertise during the methodology discussions and validation studies; Don McDonald, Alex Steurrys, Shane Rich, Brandon Maurer and other members of the Corteva engineering team for their guidance on system designs and troubleshooting; and John Arbuckle for his leadership, support and vision towards research innovation.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAmbus P, Skiba U, Drewer J, Jones SK, Carter MS, Albert KR, Sutton MA. Development of an accumulation-based system for cost‐effective chamber measurements of inert trace gas fluxes. Eur J Soil Sci. 2010;61(5):785\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2389.2010.01272.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2389.2010.01272.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArias-Navarro C, D\u0026iacute;az-Pin\u0026eacute;s E, Kiese R, Rosenstock TS, Rufino MC, Stern D, Neufeldt H, Verchot LV, Butterbach-Bahl K. Gas pooling: A sampling technique to overcome spatial heterogeneity of soil carbon dioxide and nitrous oxide fluxes. Soil Biol Biochem. 2013;67:20\u0026ndash;3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.soilbio.2013.08.011\u003c/span\u003e\u003cspan address=\"10.1016/j.soilbio.2013.08.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eButterbach-Bahl K, Baggs EM, Dannenmann M, Kiese R, Zechmeister-Boltenstern S. Nitrous oxide emissions from soils: how well do we understand the processes and their controls? Philosophical Trans Royal Soc B-Biological Sci. 2013;368(1621). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rstb.2013.0122\u003c/span\u003e\u003cspan address=\"10.1098/rstb.2013.0122\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChadwick DR, Cardenas L, Misselbrook TH, Smith KA, Rees RM, Watson CJ, McGeough KL, Williams JR, Cloy JM, Thorman RE, Dhanoa MS. Optimizing chamber methods for measuring nitrous oxide emissions from plot-based agricultural experiments. Eur J Soil Sci. 2014;65(2):295\u0026ndash;307. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ejss.12117\u003c/span\u003e\u003cspan address=\"10.1111/ejss.12117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCharles A, Rochette P, Whalen JK, Angers DA, Chantigny MH, Bertrand N. Global nitrous oxide emission factors from agricultural soils after addition of organic amendments: A meta-analysis. Agric Ecosyst Environ. 2017;236:88\u0026ndash;98. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.agee.2016.11.021\u003c/span\u003e\u003cspan address=\"10.1016/j.agee.2016.11.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCharteris AF, Chadwick DR, Thorman RE, Vallejo A, de Klein CAM, Rochette P, C\u0026aacute;rdenas LM. Global Research Alliance N2O chamber methodology guidelines: Recommendations for deployment and accounting for sources of variability. J Environ Qual. 2020;49(5):1092\u0026ndash;109. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jeq2.20126\u003c/span\u003e\u003cspan address=\"10.1002/jeq2.20126\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavis MP, Groh TA, Parkin TB, Williams RJ, Isenhart TM, Hofmockel KS. Portable Automation of Static Chamber Sample Collection for Quantifying Soil Gas Flux. J Environ Qual. 2018;47(2):270\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2134/jeq2017.10.0387\u003c/span\u003e\u003cspan address=\"10.2134/jeq2017.10.0387\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDella Chiesa T, Northrup D, Miguez FE, Archontoulis SV, Baum ME, Venterea RT, Emmett BD, Malone RW, Iqbal J, Necpalova M, Castellano MJ. Reducing greenhouse gas emissions from North American soybean production. Nat Sustain. 2024;7(12):1608\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41893-024-01458-9\u003c/span\u003e\u003cspan address=\"10.1038/s41893-024-01458-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFerrari Machado PV, Wagner-Riddle C, MacTavish R, Voroney PR, Bruulsema TW. Diurnal Variation and Sampling Frequency Effects on Nitrous Oxide Emissions Following Nitrogen Fertilization and Spring-Thaw Events. Soil Sci Soc Am J. 2019;83(3):743\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2136/sssaj2018.10.0365\u003c/span\u003e\u003cspan address=\"10.2136/sssaj2018.10.0365\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFuchs K, Merbold L, Buchmann N, Bretscher D, Brilli L, Fitton N, Topp CFE, Klumpp K, Lieffering M, Martin R, Newton PCD, Rees RM, Rolinski S, Smith P, Snow V. Multimodel Evaluation of Nitrous Oxide Emissions From an Intensively Managed Grassland. J Geophys Research-Biogeosciences. 2020;125(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1029/2019JG005261\u003c/span\u003e\u003cspan address=\"10.1029/2019JG005261\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIBM. (2024). \u003cem\u003eIBM Environmental Intelligence Suite\u003c/em\u003e. In IBM Corporation. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ibm.com/products/environmental-intelligence-suite\u003c/span\u003e\u003cspan address=\"https://www.ibm.com/products/environmental-intelligence-suite\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLawrence NC, Hall SJ. Capturing temporal heterogeneity in soil nitrous oxide fluxes with a robust and low-cost automated chamber apparatus. Atmos Meas Tech. 2020;13(7):4065\u0026ndash;78. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/amt-13-4065-2020\u003c/span\u003e\u003cspan address=\"10.5194/amt-13-4065-2020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaier M, Weber TKD, Fiedler J, Fu\u0026szlig; R, Glatzel S, Huth V, Jordan S, Jurasinski G, Kutzbach L, Sch\u0026auml;fer K, Weymann D, Hagemann U. Introduction of a guideline for measurements of greenhouse gas fluxes from soils using non-steady‐state chambers. J Plant Nutr Soil Sci. 2022;185(4):447\u0026ndash;61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jpln.202200199\u003c/span\u003e\u003cspan address=\"10.1002/jpln.202200199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMbow HOP, Reisinger A, Canadell J, O\u0026rsquo;Brien P. (2017). \u003cem\u003eSpecial report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems (SR2).\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOlander LP, Wollenberg E, Tubiello FN, Herold M. Synthesis and Review: Advancing agricultural greenhouse gas quantification. Environ Res Lett. 2014;9(7). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1088/1748-9326/9/7/075003\u003c/span\u003e\u003cspan address=\"10.1088/1748-9326/9/7/075003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParkin TB, Venterea RT, Hargreaves SK. Calculating the Detection Limits of Chamber-based Soil Greenhouse Gas Flux Measurements. J Environ Qual. 2012;41(3):705\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2134/jeq2011.0394\u003c/span\u003e\u003cspan address=\"10.2134/jeq2011.0394\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReay DS, Davidson EA, Smith KA, Smith P, Melillo JM, Dentener F, Crutzen PJ. Global agriculture and nitrous oxide emissions. Nat Clim Change. 2012;2(6):410\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/Nclimate1458\u003c/span\u003e\u003cspan address=\"10.1038/Nclimate1458\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaha D, Basso B, Robertson GP. Machine learning improves predictions of agricultural nitrous oxide (N2O) emissions from intensively managed cropping systems. Environ Res Lett. 2021;16(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1088/1748-9326/abd2f3\u003c/span\u003e\u003cspan address=\"10.1088/1748-9326/abd2f3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeabold S, Perktold J. (2010). \u003cem\u003estastmodels: Econometric and statistical modeling with Python\u003c/em\u003e. In \u003cem\u003eProceeding of the 9th Python in Science Conference\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/statsmodels/statsmodels/\u003c/span\u003e\u003cspan address=\"https://github.com/statsmodels/statsmodels/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStorer K, Coggan A, Ineson P, Hodge A. Arbuscular mycorrhizal fungi reduce nitrous oxide emissions from N2O hotspots. New Phytol. 2018;220(4):1285\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/nph.14931\u003c/span\u003e\u003cspan address=\"10.1111/nph.14931\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTian HQ, Xu RT, Canadell JG, Thompson RL, Winiwarter W, Suntharalingam P, Davidson EA, Ciais P, Jackson RB, Janssens-Maenhout G, Prather MJ, Regnier P, Pan NQ, Pan SF, Peters GP, Shi H, Tubiello FN, Zaehle S, Zhou F, Yao YZ. A comprehensive quantification of global nitrous oxide sources and sinks. Nature. 2020;586(7828):248\u0026ndash;. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-020-2780-0\u003c/span\u003e\u003cspan address=\"10.1038/s41586-020-2780-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVenterea RT, Petersen SO, de Klein CAM, Pedersen AR, Noble ADL, Rees RM, Gamble JD, Parkin TB. Global Research Alliance N(2) O chamber methodology guidelines: Flux calculations. J Environ Qual. 2020;49(5):1141\u0026ndash;55. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jeq2.20118\u003c/span\u003e\u003cspan address=\"10.1002/jeq2.20118\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagner SW, Reicosky DC, Alessi RS. Regression models for calculating gas fluxes measured with a closed chamber. Agron J. 1997;89(2):279\u0026ndash;84. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/DOI 10.2134/agronj1997.00021962008900020021x\u003c/span\u003e\u003cspan address=\"DOI 10.2134/agronj1997.00021962008900020021x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagner-Riddle C, Baggs EM, Clough TJ, Fuchs K, Petersen SO. Mitigation of nitrous oxide emissions in the context of nitrogen loss reduction from agroecosystems: managing hot spots and hot moments. Curr Opin Environ Sustain. 2020;47:46\u0026ndash;53. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cosust.2020.08.002\u003c/span\u003e\u003cspan address=\"10.1016/j.cosust.2020.08.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"plant-methods","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"plme","sideBox":"Learn more about [Plant Methods](http://plantmethods.biomedcentral.com/)","snPcode":"13007","submissionUrl":"https://submission.nature.com/new-submission/13007/3","title":"Plant Methods","twitterHandle":"@PlantMethods","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Greenhouse Gas, Nitrous Oxide, Trace Gas Flux, Agriculture, Chamber, Soil","lastPublishedDoi":"10.21203/rs.3.rs-9013846/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9013846/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAgricultural soils are a major source of nitrous oxide (N\u003csub\u003e2\u003c/sub\u003eO), a potent greenhouse gas, but challenges in measuring its highly dynamic flux in field settings hamper modeling and mitigation efforts. We developed a remotely-deployable, high-frequency sampling system that lowers the cost of N\u003csub\u003e2\u003c/sub\u003eO flux measurement and minimizes laboratory analysis. The Automated Bulk Sampling System (ABSS) is self-powered and accumulates hourly open- and closed-chamber headspace gas samples into two separate gas collection bags. The system produces two bulked gas samples at the end of the measurement period, allowing average hourly N\u003csub\u003e2\u003c/sub\u003eO flux over 2-week collection periods to be estimated. Lab-based validation experiments showed high agreement between real-time analyzer and accumulated ABSS concentration readings (r\u003csup\u003e2\u003c/sup\u003e: 0.998, bias: -0.009\u0026thinsp;\u0026plusmn;\u0026thinsp;0.002). The system also showed high precision, or repeatability (r\u003csup\u003e2\u003c/sup\u003e: 0.791) in field validation experiments, but an underestimation bias of 25% for N\u003csub\u003e2\u003c/sub\u003eO fluxes was observed when compared to 2-week average real-time analyzer results. In exploring sources of error, we found overestimation of ambient, open-chamber samples by the ABSS to be the largest source of error (15%), augmented by underestimation of closed-chamber sample concentrations (5%). Loss of information from meteorological variation and two-point flux calculation contributed slightly to underestimation bias (6%). We used historic weather data from the U.S. Corn Belt to simulate the potential error contribution from air density variation, and found an average error of 0.049%, with the largest range in error occurring at lower fluxes. Our results demonstrate that ABSS is a valuable low-cost and low-labor solution for integrated estimates of soil N\u003csub\u003e2\u003c/sub\u003eO flux in large-footprint, replicated plot experimental contexts and can help resolve critical questions in managing soil N\u003csub\u003e2\u003c/sub\u003eO emissions in agricultural systems.\u003c/p\u003e","manuscriptTitle":"The Automated Bulk Sampling System (ABSS), a low-cost solution for integrated nitrous oxide emission quantification in field studies","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-12 06:46:05","doi":"10.21203/rs.3.rs-9013846/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-18T09:23:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"263974148869588704921338724640294012728","date":"2026-05-18T08:27:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"178404377890373442734036790774909112167","date":"2026-05-18T06:56:24+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-16T14:16:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"222948913885704587955193316268419965737","date":"2026-03-09T10:39:53+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-09T01:19:59+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-07T06:32:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-07T06:32:35+00:00","index":"","fulltext":""},{"type":"submitted","content":"Plant Methods","date":"2026-03-02T21:22:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"plant-methods","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"plme","sideBox":"Learn more about [Plant Methods](http://plantmethods.biomedcentral.com/)","snPcode":"13007","submissionUrl":"https://submission.nature.com/new-submission/13007/3","title":"Plant Methods","twitterHandle":"@PlantMethods","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"22df80e6-30b0-464f-9e2f-d5167551b2e6","owner":[],"postedDate":"March 12th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-18T09:23:07+00:00","index":58,"fulltext":""},{"type":"reviewerAgreed","content":"263974148869588704921338724640294012728","date":"2026-05-18T08:27:04+00:00","index":57,"fulltext":""},{"type":"reviewerAgreed","content":"178404377890373442734036790774909112167","date":"2026-05-18T06:56:24+00:00","index":56,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-12T06:46:05+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-12 06:46:05","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9013846","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9013846","identity":"rs-9013846","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}