Search

Agricultural nitrous oxide (N2O) emission accounts for a non-trivial fraction of global greenhouse gas (GHG) budget. To date, estimating N2O fluxes from cropland remains a challenging task because the related microbial processes (e.g., nitrification and denitrification) are controlled by complex interactions among climate, soil, plant and human activities. Existing approaches such as process-based (PB) models have well-known limitations due to insufficient representations of the processes or uncertainties of model parameters, and due to leverage recent advances in machine learning (ML) a new method is needed to unlock the "black box" to overcome its limitations such as low interpretability, out-of-sample failure and massive data demand. In this study, we developed a first-of-its-kind knowledge-guided machine learning model for agroecosystems (KGML-ag) by incorporating biogeophysical and chemical domain knowledge from an advanced PB model, ecosys, and tested it by comparing simulating daily N2O fluxes with real observed data from mesocosm experiments. The gated recurrent unit (GRU) was used as the basis to build the model structure. To optimize the model performance, we have investigated a range of ideas, including (1) using initial values of intermediate variables (IMVs) instead of time series as model input to reduce data demand; (2) building hierarchical structures to explicitly estimate IMVs for further N2O prediction; (3) using multi-task learning to balance the simultaneous training on multiple variables; and (4) pre-training with millions of synthetic data generated from ecosys and fine-tuning with mesocosm observations. Six other pure ML models were developed using the same mesocosm data to serve as the benchmark for the KGML-ag model. Results show that KGML-ag did an excellent job in reproducing the mesocosm N2O fluxes (overall r2=0.81 , and RMSE=3.6 mgNm-2d-1 from cross validation). Importantly, KGML-ag always outperforms the PB model and ML models in predicting N2O fluxes, especially for complex temporal dynamics and emission peaks. Besides, KGML-ag goes beyond the pure ML models by providing more interpretable predictions as well as pinpointing desired new knowledge and data to further empower the current KGML-ag. We believe the KGML-ag development in this study will stimulate a new body of research on interpretable ML for biogeochemistry and other related geoscience processes. [ABSTRACT FROM AUTHOR]

Agricultural nitrous oxide (N2O) emission accounts for a non-trivial fraction of global greenhouse gases (GHGs) budget. To date, estimating N2O fluxes from cropland remains a challenging task because the related microbial processes (e.g., nitrification and denitrification) are controlled by complex interactions among climate, soil, plant and human activities. Existing approaches such as process-based (PB) models have well-known limitations due to insufficient representations of the processes or constraints of model parameters, and to leverage recent advances in machine learning (ML) new method is needed to unlock the 'black box' to overcome its limitations due to low interpretability, out-of- sample failure and massive data demand. In this study, we developed a first of its kind knowledge-guided machine learning model for agroecosystems (KGML-ag), by incorporating biogeophysical/chemical domain knowledge from an advanced PB model, ecosys, and tested it by simulating daily N2O fluxes with real observed data from mesocosm experiments. The Gated Recurrent Unit (GRU) was used as the basis to build the model structure. To optimize the model performance, we have investigated a range of ideas, including: 1) Using initials of intermediate variables (IMVs) instead of time series as model input to reduce data demand; 2) Building hierarchical structures to explicitly estimate IMVs for further N2O prediction; 3) Using multitask learning to balance the simultaneous training on multiple variables; and 4) Pretraining with millions of synthetic data generated from ecosys and fine tuning with mesocosm observations. Six other pure ML models were developed using the same mesocosm data to serve as the benchmark for the KGML-ag model. Results show that KGML-ag did an excellent job in reproducing the mesocosm N2O fluxes (overall r2 = 0.81, and RMSE = 3.6 mg N m-2 day-1 from cross-validation). Importantly KGML-ag always outperforms the PB model and ML models in predicting N2O fluxes, especially for complex temporal dynamics and emission peaks. Besides, KGML-ag goes beyond the pure ML models by providing more interpretable predictions as well as pinpointing desired new knowledge and data to further empower the current KGML- ag. We believe the KGML-ag development in this study will stimulate a new body of research on interpretable ML for biogeochemistry and other related geoscience processes. [ABSTRACT FROM AUTHOR]

The atmospheric carbon dioxide (CO 2) mixing ratio and its carbon isotope (δ13 C-CO 2) composition contain important CO 2 sink and source information spanning from ecosystem to global scales. The observation and simulation for both CO 2 and δ13 C-CO 2 can be used to constrain regional emissions and better understand the anthropogenic and natural mechanisms that control δ13 C-CO 2 variations. Such work remains rare for urban environments, especially megacities. Here, we used near-continuous CO 2 and δ13 C-CO 2 measurements, from September 2013 to August 2015, and inverse modeling to constrain the CO 2 budget and investigate the main factors that dominated δ13 C-CO 2 variations for the Yangtze River delta (YRD) region, one of the largest anthropogenic CO 2 hotspots and densely populated regions in China. We used the WRF-STILT model framework with category-specified EDGAR v4.3.2 CO 2 inventories to simulate hourly CO 2 mixing ratios and δ13 C-CO 2 , evaluated these simulations with observations, and constrained the total anthropogenic CO 2 emission. We show that (1) top-down and bottom-up estimates of anthropogenic CO 2 emissions agreed well (bias < 6 %) on an annual basis, (2) the WRF-STILT model can generally reproduce the observed diel and seasonal atmospheric δ13 C-CO 2 variations, and (3) anthropogenic CO 2 emissions played a much larger role than ecosystems in controlling the δ13 C-CO 2 seasonality. When excluding ecosystem respiration and photosynthetic discrimination in the YRD area, δ13 C-CO 2 seasonality increased from 1.53 ‰ to 1.66 ‰. (4) Atmospheric transport processes in summer amplified the cement CO 2 enhancement proportions in the YRD area, which dominated monthly δs (the mixture of δ13 C-CO 2 from all regional end-members) variations. These findings show that the combination of long-term atmospheric carbon isotope observations and inverse modeling can provide a powerful constraint on the carbon cycle of these complex megacities. [ABSTRACT FROM AUTHOR]

We apply airborne measurements across three seasons (summer, winter and spring 2017–2018) in a multi-inversion framework to quantify methane emissions from the US Corn Belt and Upper Midwest, a key agricultural and wetland source region. Combing our seasonal results with prior fall values we find that wetlands are the largest regional methane source (32 %, 20 [16–23] Gg/d), while livestock (enteric/manure; 25 %, 15 [14–17] Gg/d) are the largest anthropogenic source. Natural gas/petroleum, waste/landfills, and coal mines collectively make up the remainder. Optimized fluxes improve model agreement with independent datasets within and beyond the study timeframe. Inversions reveal coherent and seasonally dependent spatial errors in the WetCHARTs ensemble mean wetland emissions, with an underestimate for the Prairie Pothole region but an overestimate for Great Lakes coastal wetlands. Wetland extent and emission temperature dependence have the largest influence on prediction accuracy; better representation of coupled soil temperature–hydrology effects is therefore needed. Our optimized regional livestock emissions agree well with the Gridded EPA estimates during spring (to within 7 %) but are ∼ 25 % higher during summer and winter. Spatial analysis further shows good top-down and bottom-up agreement for beef facilities (with mainly enteric emissions) but larger (∼ 30 %) seasonal discrepancies for dairies and hog farms (with >40 % manure emissions). Findings thus support bottom-up enteric emission estimates but suggest errors for manure; we propose that the latter reflects inadequate treatment of management factors including field application. Overall, our results confirm the importance of intensive animal agriculture for regional methane emissions, implying substantial mitigation opportunities through improved management. [ABSTRACT FROM AUTHOR]

The atmospheric CO2 mixing ratio and its δ13C-CO2 composition contain important CO2 sink and source information spanning from ecosystem to global scales. The observation and simulation for both CO2 and its carbon isotope ratio (δ13C-CO2) can be used to constrain regional emissions and better understand the anthropogenic and natural mechanisms that control δ13C-CO2 variations. Such work remains rare for urban environments, especially megacities. Here, we used near-continuous CO2 and δ13C-CO2 measurements, from September 2013 to August 2015, and inverse modeling to constrain the CO2 budget and investigate the main factors that dominated δ13C-CO2 variations for the Yangtze River Delta (YRD) region, one of the largest anthropogenic CO2 hotspots and densely populated regions in China. We used the WRF-STILT model framework with category-specified EDGAR v432 CO2 inventories to simulate hourly CO2 mixing ratios and δ13C-CO2, evaluated these simulations with observations, and constrained the anthropogenic CO2 emission categories. Our study shows that: (1) Top-down and bottom-up estimates of anthropogenic CO2 emissions agreed well (bias < 6 %) on an annual basis; (2) The WRF-STILT model performed well in reproducing the observed diel and seasonal atmospheric δ13C-CO2 variations; (3) Anthropogenic CO2 emissions played a much larger role than ecosystems in controlling the δ13C-CO2 seasonality. When excluding ecosystem respiration and photosynthetic discrimination in the YRD area, δ13C-CO2 seasonality increased from 1.53 ‰ to 1.66 ‰; (4) Atmospheric transport processes in summer amplified the cement CO2 enhancement proportions in the YRD area, which dominated monthly δs variations. These findings support that the combination of long-term atmospheric carbon isotope observations and inverse modeling can provide a powerful constraint on the carbon cycle of these complex megacities. [ABSTRACT FROM AUTHOR]

We present top-down constraints on global monthly N2O emissions for 2011 from a multi-inversion approach and an ensemble of surface observations. The inversions employ the GEOS-Chem adjoint and an array of aggregation strategies to test how well current observations can constrain the spatial distribution of global N2O emissions. The strategies include (1) a standard 4D-Var inversion at native model resolution (4° 5°), (2) an inversion for six continental and three ocean regions, and (3) a fast 4D-Var inversion based on a novel dimension reduction technique employing randomized singular value decomposition (SVD). The optimized global flux ranges from 15.9 TgNyr-1 (SVDbased inversion) to 17.5-17.7 TgNyr-1 (continental-scale, standard 4D-Var inversions), with the former better capturing the extratropical N2O background measured during the HIAPER Pole-to-Pole Observations (HIPPO) airborne campaigns. We find that the tropics provide a greater contribution to the global N2O flux than is predicted by the prior bottomup inventories, likely due to underestimated agricultural and oceanic emissions. We infer an overestimate of natural soil emissions in the extratropics and find that predicted emissions are seasonally biased in northern midlatitudes. Here, optimized fluxes exhibit a springtime peak consistent with the timing of spring fertilizer and manure application, soil thawing, and elevated soil moisture. Finally, the inversions reveal a major emission underestimate in the US Corn Belt in the bottom-up inventory used here.We extensively test the impact of initial conditions on the analysis and recommend formally optimizing the initial N2O distribution to avoid biasing the inferred fluxes. We find that the SVD-based approach provides a powerful framework for deriving emission information from N2O observations: by defining the optimal resolution of the solution based on the information content of the inversion, it provides spatial information that is lost when aggregating to political or geographic regions, while also providing more temporal information than a standard 4D-Var inversion. [ABSTRACT FROM AUTHOR]

We present top-down constraints on global, monthly N2O emissions for 2011 from a multi-inversion approach and an ensemble of surface observations. The inversions employ the GEOS-Chem adjoint and an array of aggregation strategies to test how well current observations can constrain the spatial distribution of global N2O emissions. The strategies include: (1) a standard 4D-Var inversion at native model resolution (4° × 5°), (2) an inversion for six continental and three ocean regions, and (3) a fast 4D-Var inversion based on a novel dimension reduction technique employing randomized singular value decomposition (SVD). The optimized global flux ranges from 15.9 Tg N yr-1 (SVD-based inversion) to 17.5-17.7 Tg N yr-1 (continental-scale, standard 4D-Var inversions), with the former better capturing the N2O background measured during the HIAPER Pole-to-Pole Observations (HIPPO) airborne campaigns. We find that the tropics provide a greater contribution to the global N2O flux than is predicted by the prior bottom-up inventories, likely due to underestimated agricultural and oceanic emissions. We infer an overestimate of natural soil emissions in the extratropics, and find that predicted emissions are seasonally biased in northern midlatitudes. Here, optimized fluxes exhibit a springtime versus summertime peak more consistent with the timing of fertilizer application, soil thawing, and elevated soil moisture. Finally, the inversions reveal a major emission underestimate in the US Corn Belt (which may extend to other intensive agricultural regions), likely from underrepresentation of indirect N2O emissions from leaching and runoff. We extensively test the impact of initial conditions on the analysis and recommend formally optimizing the initial N2O distribution to avoid aliasing the inferred fluxes. We find that the SVD-based approach provides a powerful framework for deriving emission information from N2O observations: by defining the state vector based on the information content of the inversion, it provides useful spatial information that is lost when aggregating to ad-hoc regions, while also better resolving temporal features than a standard 4D-Var inversion. [ABSTRACT FROM AUTHOR]

Nitrous oxide (N2O) has a high global warming potential and depletes stratospheric ozone. The U. S. Corn Belt plays an important role in the global anthropogenic N2O budget. To date, studies on local surface N2O emission and the atmospheric N2O budget have commonly used Lagrangian models. In the present study, we used an Eulerian model - Weather Research and Forecasting Chemistry (WRF-Chem) model to investigate the relationships between N2O emission in the Corn Belt and observed atmospheric N2O mixing ratios. Modeled hourly N2O mixing ratios were combined with continuous atmospheric N2O measurements at the KCMP tall tower in Minnesota to constrain agricultural N2O emissions. The modeled spatial patterns of atmospheric N2O were validated against discrete observations at multiple tall towers in the NOAA flask network. After optimization of the surface flux, the model reproduced reasonably well the hourly N2O mixing ratios monitored at the KCMP tower. Agricultural N2O emissions in the EDGAR42 database needed to be scaled up by 19.0 to 28.1 fold to represent the true emission in the Corn Belt from June 1-20, 2010 - a peak emission period. Optimized total N2O emissions were 3.00-4.38, 1.52-2.08, 0.61-0.81 and 0.56-0.75 nmol m-2 s-1 from June 1-20, August 1-20, October 1-20 and December 1-20, 2010, respectively. The simulated spatial patterns of atmospheric N2O mixing ratios were in good agreement with the NOAA discrete observations during the strong emission peak in June. Such spatial patterns illustrate that the IPCC (Inter-governmental Panel on Climate Change) underestimate of emissions is not dependent on tower measurement location. [ABSTRACT FROM AUTHOR]

Increasing atmospheric humidity and convective precipitation over land provide evidence of intensification of the hydrologic cycle - an expected response to surface warming. The extent to which terrestrial ecosystems modulate these hydrologic factors is important to understand feedbacks in the climate system. We measured the oxygen and hydrogen isotope composition of water vapor at a very tall tower (185 m) in the upper Midwest, United States, to diagnose the sources, transport, and fractionation of water vapor in the planetary boundary layer (PBL) over a 3-year period (2010 to 2012). These measurements represent the first set of annual water vapor isotope observations for this region. Several simple isotope models and cross-wavelet analyses were used to assess the importance of the Rayleigh distillation process, evaporation, and PBL entrainment processes on the isotope composition of water vapor. The vapor isotope composition at this tall tower site showed a large seasonal amplitude (mean monthly δ18Ov ranged from -40.2 to -15.9‰ and δ2Hv ranged from -278.7 to -113.0 ‰) and followed the familiar Rayleigh distillation relation with water vapor mixing ratio when considering the entire hourly data set. However, this relation was strongly modulated by evaporation and PBL entrainment processes at timescales ranging from hours to several days. The wavelet coherence spectra indicate that the oxygen isotope ratio and the deuterium excess (dv) of water vapor are sensitive to synoptic and PBL processes. According to the phase of the coherence analyses, we show that evaporation often leads changes in dv, confirming that it is a potential tracer of regional evaporation. Isotope mixing models indicate that on average about 31% of the growing season PBL water vapor is derived from regional evaporation. However, isoforcing calculations and mixing model analyses for high PBL water vapor mixing ratio events (>25 mmol mol-1) indicate that regional evaporation can account for 40 to 60% of the PBL water vapor. These estimates are in relatively good agreement with that derived from numerical weather model simulations. This relatively large fraction of evaporation-derived water vapor implies that evaporation has an important impact on the precipitation recycling ratio within the region. Based on multiple constraints, we estimate that the summer season recycling fraction is about 30 %, indicating a potentially important link with convective precipitation. [ABSTRACT FROM AUTHOR]