Process-based Model Research Articles

Tracking the nitrogen leaching from different sources in bioretention systems with a process-based model

Bioretention systems are a typical stormwater management technology intended to reduce runoff and non-point source pollution. Multiple sources can contribute to nitrogen (N) leaching from bioretention systems, including the current rainfall-runoff (i.e., immediate leaching), recent rainfall-runoffs (i.e., fast leaching) and long-ago legacy N (i.e., slow leaching), however, there is a lack of methods to quantify these contributions. Here, a process-based Bioretention System (BRS) model for N leaching simulations was verified with artificially-labeled 15N data besides N data to reduce predictive uncertainties. The BRS model was then coupled with an Nitrogen Source Apportionment (NSA) module for N leaching tracking to quantify the contributions from different sources. Results from a bioretention system showed that additional 15N data reduced the predictive uncertainty of total N (TN) concentrations by 23 %. Immediate leaching, fast leaching and slow leaching increased, decreased and remained steady during the leaching process and accounted for 79.54 ± 5.08 %, 2.80 ± 1.74 % and 17.66 ± 4.11 % of N leaching, respectively. The dominant immediate leaching can be significantly weakened by a submerged zone, a shorter antecedent dry period (ADP), a lower rainfall intensity or influent concentration. This study provides a framework to quantify N leaching from multiple sources with implications for long-term N leaching management.

A family of process-based models to simulate landscape use by multiple taxa

ContextLand-use change is a key driver of biodiversity loss. Models that accurately predict how biodiversity might be affected by land-use changes are urgently needed, to help avoid further negative impacts and inform landscape-scale restoration projects. To be effective, such models must balance model realism with computational tractability and must represent the different habitat and connectivity requirements of multiple species.ObjectivesWe explored the extent to which process-based modelling might fulfil this role, examining feasibility for different taxa and potential for informing real-world decision-making.MethodsWe developed a family of process-based models (*4pop) that simulate landscape use by birds, bats, reptiles and amphibians, derived from the well-established poll4pop model (designed to simulate bee populations). Given landcover data, the models predict spatially-explicit relative abundance by simulating optimal home-range foraging, reproduction, dispersal of offspring and mortality. The models were co-developed by researchers, conservation NGOs and volunteer surveyors, parameterised using literature data and expert opinion, and validated against observational datasets collected across Great Britain.ResultsThe models were able to simulate habitat specialists, generalists, and species requiring access to multiple habitats for different types of resources (e.g. breeding vs foraging). We identified model refinements required for some taxa and considerations for modelling further species/groups.ConclusionsWe suggest process-based models that integrate multiple forms of knowledge can assist biodiversity-inclusive decision-making by predicting habitat use throughout the year, expanding the range of species that can be modelled, and enabling decision-makers to better account for landscape context and habitat configuration effects on population persistence.

Interpretable Prediction of SARS-CoV-2 Epitope-Specific TCR Recognition Using a Pre-Trained Protein Language Model.

The emergence of the novel coronavirus, designated as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), has posed a significant threat to public health worldwide. There has been progress in reducing hospitalizations and deaths due to SARS-CoV-2. However, challenges stem from the emergence of SARS-CoV-2 variants, which exhibit high transmission rates, increased disease severity, and the ability to evade humoral immunity. Epitope-specific T-cell receptor (TCR) recognition is key in determining the T-cell immunogenicity for SARS-CoV-2 epitopes. Although several data-driven methods for predicting epitope-specific TCR recognition have been proposed, they remain challenging due to the enormous diversity of TCRs and the lack of available training data. Self-supervised transfer learning has recently been proven useful for extracting information from unlabeled protein sequences, increasing the predictive performance of fine-tuned models, and using a relatively small amount of training data. This study presents a deep-learning model generated by fine-tuning pre-trained protein embeddings from a large corpus of protein sequences. The fine-tuned model showed markedly high predictive performance and outperformed the recent Gaussian process-based prediction model. The output attentions captured by the deep-learning model suggested critical amino acid positions in the SARS-CoV-2 epitope-specific TCRβ sequences that are highly associated with the viral escape of T-cell immune response.

Modeling and evaluation of centrifugal pump performance degradation model based on stochastic process

In order to clarify the performance degradation problem of fuel centrifugal pumps, a stochastic process-based performance degradation model modeling method is given with an aero-engine fuel centrifugal pump as the object of study, and the reliability assessment is carried out. First, the degradation modeling method based on stochastic process is given. Secondly, PumpLinx is utilized to obtain the performance degradation data of the centrifugal pump and perform data processing. Gamma and Wiener stochastic process models are given based on MATLAB. Finally, based on the centrifugal pump performance degradation data for centrifugal pump reliability assessment, and compare and analyze the advantages and disadvantages of the two modeling methods, and found that the degradation modeling method based on the Gamma process is more suitable for centrifugal pump reliability assessment.

Enhancing physically-based hydrological modeling with an ensemble of machine-learning reservoir operation modules under heavy human regulation using easily accessible data

Dams and reservoirs have significantly altered river flow dynamics worldwide. Accurately representing reservoir operations in hydrological models is crucial yet challenging. Detailed reservoir operation data is often inaccessible, leading to relying on simplified reservoir operation modules in most hydrological models. To improve the capability of hydrological models to capture flow variability influenced by reservoirs, this study proposes a hybrid hydrological modeling framework, which combines a process-based hydrological model with a machine-learning-based reservoir operation module designed to simulate runoff under reservoir operations. The reservoir operation module employs an ensemble of three machine learning models: random forest, support vector machine, and AutoGluon. These models predict reservoir outflows using precipitation and temperature data as inputs. The Soil and Water Assessment Tool (SWAT) then integrates these outflow predictions to simulate runoff. To evaluate the performance of this hybrid approach, the Xijiang Basin within the Pearl River Basin, China, is used as a case study. The results highlight the superiority of the SWAT model coupled with machine learning-based reservoir operation models compared to alternative modeling approaches. This hybrid model effectively captures peak flows and dry period runoff. The Nash-Sutcliffe Efficiency (NSE) in daily runoff simulations shows substantial improvement, ranging from 0.141 to 0.780, with corresponding enhancements in the coefficient of determination (R2) by 0.098–0.397 when compared to the original reservoir operation modules in SWAT. In comparison to parameterization techniques lacking a dedicated reservoir module, NSE enhancements range from 0.068 to 0.537, and R2 improvements range from 0.027 to 0.139. The proposed hybrid modeling approach effectively characterizes the impact of reservoir operations on river flow dynamics, leading to enhanced accuracy in runoff simulation. These findings offer valuable insights for hydrological forecasting and water resources management in regions influenced by reservoir operations.

Research progress on cambium activity and radial growth dynamics monitoring of coniferous species

The radial growth of trees plays a crucial role in determining forest carbon sequestration capacity. Understanding the growth dynamics of trees and their response to environmental factors is essential for predicting forest's carbon sink potential under future climate change. Coniferous forest trees are particularly sensitive to climate change, with growth dynamics responding rapidly to environmental shifts. We collected and analyzed data from 99 papers published between 1975 and 2023, and examined the effects of exogenous factors (such as temperature, water, and photoperiod) and endogenous factors (including tree age and species) on cambial activity and radial growth in conifers. We further explored the mechanisms underlying these effects. The results showed that climate warming had the potential to advance the onset while delayed the end of xylem differentiation stages in conifers in temperate and boreal regions. Water availability played a crucial role in regulating the timing of cambial phenology and wood formation by influencing water potential and cell turgor. Additionally, the photoperiod not only participated in regulating the start and end times of growth, but also influenced the timing of maximum growth rate occurrence. Future climate warming was expected to extend the growing season, leading to increase in growth of conifers in boreal regions and expanding forests to higher altitudes or latitudes. However, changes in precipitation patterns and increased evapotranspiration resulting from temperature increases might advance the end of growing season and reduce growth rate in arid areas. To gain a more comprehensive understanding of the relationship between radial growth and climatic factors, it is necessary to develop process-based models to elucidate the physiological mechanisms underlying wood formation and the response of trees to climatic factors.

Battery degradation stage detection and life prediction without accessing historical operating data

Degradation stage detection and life prediction are important for battery health management and safe reuse. This study first proposes a method of detecting whether a battery has entered a rapid degradation stage without accessing historical operating data. In addition, to alleviate the burden of extensive training data, an effective method of selecting training data with high physical similarity to the test data is developed improving the model's overall performance. Six physics-informed features are extracted from an equivalent circuit model using the relaxation voltage. Unsupervised learning is employed to analyze the internal physics similarity of batteries. The relationship between features and degradation stages or battery life is described by the Gaussian process-based machine learning model. The proposed method is validated using 65 batteries of two types. The results demonstrate that the detection accuracy of the degradation stage exceeds 90 %, and the performance of the life prediction model achieves an improvement of up to 53.56 % in terms of the root mean square error compared to that of the benchmark. Both detection and prediction can be independent of historical data, showing promise in assessing whether a battery can be used in the second life and predicting battery life in real time.

Allometric relationships and trade-offs in 11 common Mediterranean-climate grasses.

Biomass allocation in plants is the foundation for understanding dynamics in ecosystem carbon balance, species competition, and plant-environment interactions. However, existing work on plant allometry has mainly focused on trees, with fewer studies having developed allometric equations for grasses. Grasses with different life histories can vary in their carbon investment by prioritizing the growth of specific organs to survive, outcompete co-occurring plants, and ensure population persistence. Further, because grasses are important fuels for wildfire, the lack of grass allocation data adds uncertainty to process-based models that relate plant physiology to wildfire dynamics. To fill this gap, we conducted a greenhouse experiment with 11 common California grasses varying in photosynthetic pathway and growth form. We measured plant sizes and harvested above- and belowground biomass throughout the life cycle of annual species, while for the establishment stage of perennial grasses to quantify allometric relationships for leaf, stem, and root biomass, as well as plant height and canopy area. We used basal diameter as a reference measure of plant size. Overall, basal diameter is the best predictor for leaf and stem biomass, height, and canopy area. Including height as another predictor can improve model accuracy in predicting leaf and stem biomass and canopy area. Fine root biomass is a function of leaf biomass alone. Species vary in their allometric relationships, with most variation occurring for plant height, canopy area, and stem biomass. We further explored potential trade-offs in biomass allocation across species between leaf and fine root, leaf and stem, and allocation to reproduction. Consistent with our expectation, we found that fast-growing plants allocated a greater fraction to reproduction. Additionally, plant height and specific leaf area negatively influenced the leaf-to-stem ratio. However, contrary to our hypothesis, there were no differences in root-to-leaf ratio between perennial and annual or C4 and C3 plants. Our study provides species-specific and functional-type-specific allometry equations for both above- and belowground organs of 11 common California grass species, enabling nondestructive biomass assessment in California grasslands. These allometric relationships and trade-offs in carbon allocation across species can improve ecosystem model predictions of grassland species interactions and environmental responses through differences in morphology.

Procedural justice and probation officer legitimacy: Testing the process-based model in community supervision

PurposeThis study provides an empirical test of Tyler's (2006) process-based model by investigating the relationship between procedural justice, legitimacy, and felt obligation to obey among people sanctioned to county-level probation. MethodsUsing a cross-sectional design, self-reported data were collected from a sample of individuals on probation in a western state (n = 185). Confirmatory factor analyses and full structural equation modeling were used to test a measurement and structural component on the relationship between procedural justice and clients' legitimacy and obligation to obey. ResultsResults demonstrate construct and discriminant validity across measures of client's procedural justice and legitimacy attitudes. In addition, SEM results revealed a positive and statistically significant association between procedural justice with probationers' legitimacy beliefs. In addition, perceived legitimacy was associated with clients' felt obligation to obey their probation officer, though indirect effects were non-significant but potentially suggested partial mediation. ConclusionsThis study provides empirical evidence supporting the positive relationship between perceptions of procedural justice with clients' legitimacy and perceived PO legitimacy with their felt obligation. Study results highlight key theoretical and policy considerations regarding the measurement and application of the process-based model in community corrections.

California’s 2023 snow deluge: Contextualizing an extreme snow year against future climate change

The increasing prevalence of low snow conditions in a warming climate has attracted substantial attention in recent years, but a focus exclusively on low snow leaves high snow years relatively underexplored. However, these large snow years are hydrologically and economically important in regions where snow is critical for water resources. Here, we introduce the term "snow deluge" and use anomalously high snowpack in California's Sierra Nevada during the 2023 water year as a case study. Snow monitoring sites across the state had a median 41 y return interval for April 1 snow water equivalent (SWE). Similarly, a process-based snow model showed a 54 y return interval for statewide April 1 SWE (90% CI: 38 to 109 y). While snow droughts can result from either warm or dry conditions, snow deluges require both cool and wet conditions. Relative to the last century, cool-season temperature and precipitation during California's 2023 snow deluge were both moderately anomalous, while temperature was highly anomalous relative to recent climatology. Downscaled climate models in the Shared Socioeconomic Pathway-370 scenario indicate that California snow deluges-which we define as the 20 y April 1 SWE event-are projected to decline with climate change (58% decline by late century), although less so than median snow years (73% decline by late century). This pattern occurs across the western United States. Changes to snow deluge, and discrepancies between snow deluge and median snow year changes, could impact water resources and ecosystems. Understanding these changes is therefore critical to appropriate climate adaptation.

Aggregation of activity data on crop management can induce large uncertainties in estimates of regional nitrogen budgets

A complete understanding of the nexus between productivity and sustainability of agricultural production systems calls for a comprehensive assessment of the nitrogen budget (NB). In our study, data from the well-monitored Danish Agricultural Watershed Monitoring Program (LOOP-program; 2013–2019) is used for a quantitative inter-comparison of three different approaches to drive the process-based model LandscapeDNDC on the regional scale. The aim is to assess how assumptions and simplifications about farm management activities at a regional scale induce previously unquantified uncertainties in the simulation of yields and the NB of cropping systems. Our findings reveal that the approach based on detailed field-level management data (A) performs the best in simulation of yield (r2 = 0.93). In contrast, the other two different data aggregation approaches (B: Sequential mono-cropping of six major crops with simulation results averaged according to proportional area, and C: simulation of 20 most frequent crop rotations) have lower correlations to the observed yields (r2 = 0.92 and 0.77, respectively) but are still statistically significant at p < 0.05 level. Notable differences arise between detailed and more aggregated crop system simulations concerning the NB, particularly concerning N losses to the environment. Compared to the detailed approach (A) (gaseous N fluxes: 24.3 kg-N ha−1 year−1; nitrate leaching: 14.7 kg-N ha−1 year−1), the aggregation approach B leads to a 31.4% over-estimation in total gaseous N fluxes (+7.6 kg-N ha−1 year−1), while nitrate leaching shows a similar average with a distinct pattern. Conversely, employing aggregation approach C leads to a 17.6% over-estimation in total gaseous fluxes (+4.3 kg-N ha−1 year−1) and a 204.9% over-estimation in nitrate leaching (+30.2 kg-N ha−1 year−1). These findings suggest that management representation should be chosen carefully because it can induce large uncertainties, especially when simulating large-scale NBs or assessing the environmental impact of cropping management. This may compromise the accuracy of national and international nutrient budgets, and preclude comparisons among different sources when the approaches for management representation differ.

Improving Decision Support Tools for Quantifying GHG Emissions from Organic Production Systems

AbstractAs food companies have adopted sustainability metrics to quantify the environmental impacts of supply chains, we need data-driven decision support tools that represent organic management practices. Decision support tools such as COMET-Farm and the Cool Farm Tool have been developed to estimate management practice impacts on soil carbon and greenhouse gas emissions from agricultural systems, but these tools have primarily been developed and used to evaluate conventional management systems. We provide an overview of the research, outreach, and educational activities used to improve these tools to better integrate organic management practices with a focus on cover crops. We summarize our previously published findings from a meta-analysis of the average potential soil carbon benefits of cover crops in temperate climates that identified planting window, biomass production, and soil texture as important predictors of cover crop soil carbon outcomes. We demonstrate how these findings were applied to improvements in process-based models and the parameterization of empirical models. In addition, we solicited feedback from organic community members on the utility of these tools and identified barriers to adoption. Finally, we evaluated both tools as resources for teaching undergraduate students about organic management systems and their impacts on greenhouse gas emissions. While both tools contain a range of customizable, organic amendment options, grazing management options still need further improvement. These improved decision support systems can help identify opportunities for enhancing the sustainability of organic systems.

Improving prediction accuracy of CSM-CERES-Wheat model for water and nitrogen response using a modified Penman-Monteith equation in a semi-arid region

Context or problemThe implementation of crop models for water and nitrogen management are important in preserving water resources and reducing groundwater pollution. The Decision Support System for Agrotechnology Transfer (DSSAT) is a collection of field-scale process-based crop models that simulate crop growth and yield in response to water and nitrogen management as well as many other factors. Objective or research questionThe objective of this study was to improve the simulation of irrigation and nitrogen fertilizer effects on growth and yield of winter wheat using locally modified Penman-Monteith (PM) equation. Specifically, we focused on improving the estimation of ETo using a locally modified Penman-Monteith (PM) equation in the CSM-CERES-Wheat model. MethodsField measured data from a two-year experiment (2009–2010 and 2010–2011) in a semi-arid region were used to calibrate and evaluate the CSM-CERES-Wheat model in a region with strong advective condition. In this experiment, the irrigation treatments were 1.2, 1.0, 0.8, 0.5 times by full irrigation requirements and rain-fed, and nitrogen treatments were 0, 46, 92, and 136 kg N ha−1. ResultsUsing the CSM-CERES-Wheat model with the unmodified PM equation resulted in some growth variables such as: aboveground dry matter, grain yield, and grain nitrogen uptake with high normalized root mean square error (NRMSE) of 0.51, 0.35, and 0.32, respectively. However, using the modified PM equation for ETo module under semi-arid condition improved the soil water content estimation and the model outputs, such as the above ground dry matter, grain yield and grain nitrogen uptake with NRMSE of 0.2, 0.15 and 0.16, respectively, that are accurate. This modification provided more accurate estimations, particularly in our experimental site in semi-aid region with strong advective conditions. ConclusionsThe findings of this study indicated that the locally modified PM equation in the CSM-CERES-Wheat model enhanced the accuracy of estimating irrigation and nitrogen fertilizer on the growth and yield of winter wheat. By improving the estimation of ETo value, the model's performance in simulating crop responses to water and nitrogen management was significantly enhanced. Implications or significanceThe utilization of the accurately estimated ETo value of the modified PM equation has significant importance in the practical application of the CSM-CERES-Wheat model. Furthermore, this modification greatly enhances the accuracy of model outputs, particularly at our experimental site located in a semi-arid region with robust advective conditions.

Disturbance rejection design for Gaussian process-based model predictive control using extended state observer

Gaussian process (GP) regression has gained significant popularity in machine learning because it has the intrinsic capability to capture uncertainty in function prediction and requires a limited number of hyperparameters to be optimized. In this study, a Gaussian process model predictive control (GPMPC) algorithm is proposed to model the unknown dynamics of the process using Gaussian process regression. The GPMPC incorporates the expected variance of the GP model to account for the model's uncertainty and to achieve prudent control. Meanwhile, the extended state observer (ESO) is introduced for the GPMPC, which can estimate the unmodeled dynamics and unknown disturbance. With the designed feedforward gain, the proposed extended state observer-based GPMPC (GPMPC-ESO) method can achieve offset-free performance. Theoretical analysis is conducted to evaluate the stability and disturbance rejection performance of the control system. Finally, the algorithms are validated by simulation in continuous stirred tank reactor (CSTR) process control.

Bayesian hidden mark interaction model for detecting spatially variable genes in imaging-based spatially resolved transcriptomics data.

Recent technology breakthroughs in spatially resolved transcriptomics (SRT) have enabled the comprehensive molecular characterization of cells whilst preserving their spatial and gene expression contexts. One of the fundamental questions in analyzing SRT data is the identification of spatially variable genes whose expressions display spatially correlated patterns. Existing approaches are built upon either the Gaussian process-based model, which relies on ad hoc kernels, or the energy-based Ising model, which requires gene expression to be measured on a lattice grid. To overcome these potential limitations, we developed a generalized energy-based framework to model gene expression measured from imaging-based SRT platforms, accommodating the irregular spatial distribution of measured cells. Our Bayesian model applies a zero-inflated negative binomial mixture model to dichotomize the raw count data, reducing noise. Additionally, we incorporate a geostatistical mark interaction model with a generalized energy function, where the interaction parameter is used to identify the spatial pattern. Auxiliary variable MCMC algorithms were employed to sample from the posterior distribution with an intractable normalizing constant. We demonstrated the strength of our method on both simulated and real data. Our simulation study showed that our method captured various spatial patterns with high accuracy; moreover, analysis of a seqFISH dataset and a STARmap dataset established that our proposed method is able to identify genes with novel and strong spatial patterns.

Climate warming effects in stratified reservoirs: Thorough assessment for opportunities and limits of machine learning techniques versus process-based models in thermal structure projections

It is nowadays a hot topic to apply machine learning (ML) algorithms to illustrate water temperature dynamics in lentic waters. Due to the limited amount of in-situ temperature measurements from traditional sampling programmes, most of the related studies, however, restricted their analysis within the surface and rarely checked the results for whole depth profiles. Moreover, capability of such methods in projecting thermal dynamics under future climate conditions is even less illustrated. To fill the gap, we collected an unparalleled huge database including 9 million water temperature measurements, from 2013 to 2022, in Rappbode Reservoir, Germany as well as the corresponding climatic and hydrological observations, to train three commonly used ML models (Random Forest, XGBoost, and Long Short-Term Memory) and comprehensively evaluate their performance in reproducing thremal structure within the whole water column. We also systemically compared such simulation results with a well-established process-driven model (CE-QUAL-W2). Going beyond a pure reproduction of observatiosn, we further evaluated the ability of CE-QUAL-W2 and ML models in projecting thermal dynamics under RCP8.5 climate scenario up to 2100. Our results suggested three ML methods yielded high accuracy in capturing water temperature dynamics from the surface (epilimnion) to bottom (hypolimnion), and also satisfactorily reproduced temporal development of the stratification pattern, with the results corresponding well with those from CE-QUAL-W2. By contrast, the projections by ML models were rather insensitive to future climate warming under RCP8.5 comparing with those by CE-QUAL-W2, pointing to an important risk whenever ML-based models are extrapolated beyond their training data range. Supported by millions of in situ measurements, our research clearly illustrated the opportunities (limits) of ML models in projecting thermal dynamics in deep waters, and the conclusion also provides key guidance for scientists to select the appropriate tools in projecting water temperature, and other environmental factors, under changing climate.

Does self and role efficacy navigate effectiveness among MSME managers? A process-based perspective

PurposeThe paper aims to propose and validate a process-based model to enhance managerial effectiveness among micro, small and medium enterprises (MSMEs). It has been observed that business uncertainties and inadequate financial resources that MSME entrepreneurs and managers face require them to constantly engage in strong self-awareness and self-regulating behavior to enhance the efficacy in their roles and, henceforth, their role performance effectiveness.Design/methodology/approachThe approach for data collection was based on the clustering of MSMEs belonging to the clusters machine tool, pump manufacturing, foundry, textile and auto-component clusters in India. The respondents to the study were MSME entrepreneurs and managers who oversee and manage multiple functions like operations, quality, marketing, sales, supply chain management, procurement, personnel and administration and general administration.FindingsThe self-efficacy of entrepreneurial managers of MSMEs is observed to play an integral role in enhancing the efficacy of their roles, thus highlighting the use of a process-based perspective while dealing with constant resource constraints and excessive dynamism in their business contexts. The ability to handle multiple tasks effectively and resilience to manage challenges enhances their role-making process, which is significant in achieving and sustaining goal-oriented behavior among MSME entrepreneurs and managers.Practical implicationsThis paper would serve as an effective model for entrepreneurs and managers to enhance their efficacy in the individual and interdependent role context, which would help achieve their individual and organizational goals. The model emphasizes a process-based perspective that thrusts the need to relate to the organizational context, enhancing individual confidence for goal-related behavior and fulfilling their role-related expectations.Originality/valueThis paper presents a model of enhancing managerial effectiveness that discusses self-efficacy as antecedent behavior. Here, personal and environmental factors aid cognition to one’s capability to construct reality, self-regulate, encode information and engage in effective managerial action.

Exploring Ecological, Morphological, and Environmental Controls on Coastal Foredune Evolution at Annual Scales Using a Process-Based Model

Coastal communities commonly rely upon foredunes as the first line of defense against sea-level rise and storms, thus requiring management guidance to optimize their protective services. Here, we use the AeoLiS model to simulate wind-driven accretion and wave-driven erosion patterns on foredunes with different morphologies and ecological properties under modern-day conditions. Additional sets of model runs mimic potential future climate changes to inform how both morphological and ecological properties may have differing contributions to net dune changes under evolving environmental forcing. This exploratory study, applied to represent the morphological, environmental, and ecological conditions of the northern Outer Banks, North Carolina, USA, finds that dunes experiencing minimal wave collision have similar net volumetric growth rates regardless of beach morphology, though the location and density of vegetation influence sediment deposition patterns across the dune profile. The model indicates that high-density, uniform planting strategies trap sediment close to the dune toe, whereas low-density plantings may allow for accretion across a broader extent of the dune face. The initial beach and dune shape generally plays a larger role in annual-scale dune evolution than vegetation cover. For steeper beach slopes and/or low dune toe elevations, the model generally predicts wave-driven dune erosion at the annual scale.

Early forecasting of corn yield and price variations using satellite vegetation products

Several approaches based on process-based, statistical or economic models have been developed and are currently used by market analysts and governmental agencies to forecast agricultural yields and commodity prices at the national and global scales. However, these models require input data that are often difficult to obtain before crop harvest and their predictions are not always reliable. Satellite data products provide easily accessible sources of information on vegetation characteristics, potentially relevant for predicting crop yield and commodity prices at a large scale, before harvest. This study presents and evaluates several methods for forecasting variations in US corn yield and global corn prices using regional vegetation satellite products. The spatial dimension of maps of vegetation indices derived from satellite products in the US Corn Belt is first reduced either by their spatial means or by principal components derived using the Empirical Orthogonal Function (EOF) method. Mean values and EOF principal components are then used as explanatory variables in generalized LASSO linear models to forecast corn price and yield changes at different time horizons, before harvest. A cross-validation method was implemented to assess the model performances using Area under the ROC curve (AUC), the Brier Skill Score (BSS) and the Matthew Correlation Coefficient (MCC). The results revealed that EOF principal components derived from satellite-based gross primary production (GPP) in July could be used to predict variations in US corn yields, with an AUC of 0.97 (95% CI 0.92-1), a BSS of 0.75 and an MCC of 0.83. The results also showed that the July spatial averages of satellite GPP data obtained in the US Corn Belt could be used to forecast the forthcoming increase or decrease of global corn prices as evidenced by an AUC reaching up to 0.91, a BSS up to 0.5, and an MCC up to 0.67. Finally, we compared our forecasts of yield and price variations to those derived from the WASDE forecasting method currently implemented in the US. This comparison showed that our approach was at least as accurate and often more accurate than WASDE. This study demonstrates how satellite data can be used to forecast variations in crop yields and corn prices several months before harvest.

Beyond assimilation of leaf area index: Leveraging additional spectral information using machine learning for site-specific soybean yield prediction

Assimilating external observations of crop state in cropping system models is essential for making spatially explicit predictions of crop variables relevant in precision agriculture. Satellite-based leaf area index (LAI) estimates have been the most frequent variable used as a proxy of actual crop growth. However, additional information beyond LAI, like canopy N content, water content, and structure, can be retrieved from satellite observations. Including such variables by data assimilation directly is difficult because many crop models do not have corresponding state variables or the relationship between the observations and the process that regulates crop growth is unclear. Therefore, other approaches are required to include such information. In this study, we investigate the improvement in the predicted yield and feature impact on model outputs by using a hybrid approach that combines observations from Sentinel-1 and 2 time-series with the outputs from a process-based model embedded in a data assimilation framework and uses the Gradient-boosted trees regressor (GBTR) as predictive model. We used two regions with soybean fields: the US (13 K points) and Uruguay (400 K points). We found an advantage when using the GBTR as the predictive model (reduced RRMSE by ∼16%) compared to data assimilation. Adding the vegetation indices had a marginal improvement (reduced RRMSE by ∼1%), while the impact of adding reflectance and backscatter values was negative. The satellite-based features had a very small importance score, while features' impact on prediction was predominantly unclear, explaining the marginal predictive power added by satellite-based features. We found that features from the reproductive stages had the highest importance, while the importance of an index related to drought stress (NMDI) across the growing season provided insights for further improvement of data assimilation methods. However, more studies are required to better disentangle pathways towards further improvement in constraining crop models by ingesting satellite observations.