Search NASA⌕ Search

SEARCH · Search NASA

Results for “PREDICTION”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 595 records · Page 33

Predicting carcinogenicity of diverse chemicals using probabilistic neural network modeling approaches

Robust global models capable of discriminating positive and non-positive carcinogens; and predicting carcinogenic potency of chemicals in rodents were developed. The dataset of 834 structurally diverse chemicals extracted from Carcinogenic Potency Database (CPDB) was used which contained 466 positive and 368 non-positive carcinogens. Twelve non-quantum mechanical molecular descriptors were derived. Structural diversity of the chemicals and nonlinearity in the data were evaluated using Tanimoto similarity index and Brock–Dechert–Scheinkman statistics. Probabilistic neural network (PNN) and generalized regression neural network (GRNN) models were constructed for classification and function optimization problems using the carcinogenicity end point in rat. Validation of the models was performed using the internal and external procedures employing a wide series of statistical checks. PNN constructed using five descriptors rendered classification accuracy of 92.09% in complete rat data. The PNN model rendered classification accuracies of 91.77%, 80.70% and 92.08% in mouse, hamster and pesticide data, respectively. The GRNN constructed with nine descriptors yielded correlation coefficient of 0.896 between the measured and predicted carcinogenic potency with mean squared error (MSE) of 0.44 in complete rat data. The rat carcinogenicity model (GRNN) applied to the mouse and hamster data yielded correlation coefficient and MSE of 0.758, 0.71 and 0.760, 0.46, respectively. The results suggest for wide applicability of the inter-species models in predicting carcinogenic potency of chemicals. Both the PNN and GRNN (inter-species) models constructed here can be useful tools in predicting the carcinogenicity of new chemicals for regulatory purposes. - Graphical abstract: Figure (a) shows classification accuracies (positive and non-positive carcinogens) in rat, mouse, hamster, and pesticide data yielded by optimal PNN model. Figure (b) shows generalization and predictive abilities of the interspecies GRNN model to predict the carcinogenic potency of diverse chemicals. - Highlights: • Global robust models constructed for carcinogenicity prediction of diverse chemicals. • Tanimoto/BDS test revealed structural diversity of chemicals and nonlinearity in data. • PNN/GRNN successfully predicted carcinogenicity/carcinogenic potency of chemicals. • Developed interspecies PNN/GRNN models for carcinogenicity prediction. • Proposed models can be used as tool to predict carcinogenicity of new chemicals.

60 APPLIED LIFE SCIENCES↗

In silico prediction of toxicity of non-congeneric industrial chemicals using ensemble learning based modeling approaches

Ensemble learning approach based decision treeboost (DTB) and decision tree forest (DTF) models are introduced in order to establish quantitative structure–toxicity relationship (QSTR) for the prediction of toxicity of 1450 diverse chemicals. Eight non-quantum mechanical molecular descriptors were derived. Structural diversity of the chemicals was evaluated using Tanimoto similarity index. Stochastic gradient boosting and bagging algorithms supplemented DTB and DTF models were constructed for classification and function optimization problems using the toxicity end-point in T. pyriformis. Special attention was drawn to prediction ability and robustness of the models, investigated both in external and 10-fold cross validation processes. In complete data, optimal DTB and DTF models rendered accuracies of 98.90%, 98.83% in two-category and 98.14%, 98.14% in four-category toxicity classifications. Both the models further yielded classification accuracies of 100% in external toxicity data of T. pyriformis. The constructed regression models (DTB and DTF) using five descriptors yielded correlation coefficients (R{sup 2}) of 0.945, 0.944 between the measured and predicted toxicities with mean squared errors (MSEs) of 0.059, and 0.064 in complete T. pyriformis data. The T. pyriformis regression models (DTB and DTF) applied to the external toxicity data sets yielded R{sup 2} and MSE values of 0.637, 0.655; 0.534, 0.507 (marine bacteria) and 0.741, 0.691; 0.155, 0.173 (algae). The results suggest for wide applicability of the inter-species models in predicting toxicity of new chemicals for regulatory purposes. These approaches provide useful strategy and robust tools in the screening of ecotoxicological risk or environmental hazard potential of chemicals. - Graphical abstract: Importance of input variables in DTB and DTF classification models for (a) two-category, and (b) four-category toxicity intervals in T. pyriformis data. Generalization and predictive abilities of the constructed (c) DTB and (d) DTF regression models to predict the T. pyriformis toxicity of diverse chemicals. - Highlights: • Ensemble learning (EL) based models constructed for toxicity prediction of chemicals • Predictive models used a few simple non-quantum mechanical molecular descriptors. • EL-based DTB/DTF models successfully discriminated toxic and non-toxic chemicals. • DTB/DTF regression models precisely predicted toxicity of chemicals in multi-species. • Proposed EL based models can be used as tool to predict toxicity of new chemicals.

60 APPLIED LIFE SCIENCES↗

Subseasonal to Seasonal Prediction of Wintertime Northern Hemisphere Extratropical Cyclone Activity by S2S and NMME Models

In this study, the prediction of wintertime extratropical cyclone activity from subseasonal to seasonal timescale in current dynamical models' reforecasts is investigated. On seasonal time scales, the North American Multi-Model Ensemble (NMME) models show skillful predictions over the eastern North Pacific, North America and the western North Atlantic with at least 5 months lead. The prediction skill is highly related to El Nino-Southern Oscillation (ENSO), as using the ENSO-related SST pattern gives rise to prediction skill with very similar spatial pattern and amplitude. On subseasonal time scales, models in the Seasonal to Sub-seasonal Prediction (S2S) dataset have skillful predictions up to 4 weeks lead over regions from the eastern North Pacific to the western Atlantic, as well as northern Europe, the eastern Atlantic and East Asia. Generally, forecast skill improves with a larger ensemble size. The subseasonal prediction skill from the Pacific to the western Atlantic is related to ENSO, and that over eastern Atlantic, Europe and East Asia are associated with stratospheric polar vortex anomalies. Current models do not show much skill from the Madden-Julian Oscillation (MJO), as the MJO impact on extratropical cyclone activity is not well captured by the models. European Centre for Medium-Range Weather Forecasts (ECMWF) model has the highest single model subseasonal prediction skill. The prediction skill in the ECMWF model is higher than its estimated potential predictability, likely because the signal-to-noise ratio is too low in the model hindcasts.

54 ENVIRONMENTAL SCIENCES↗

Large-scale genomic analyses with machine learning uncover predictive patterns associated with fungal phytopathogenic lifestyles and traits

Abstract Invasive plant pathogenic fungi have a global impact, with devastating economic and environmental effects on crops and forests. Biosurveillance, a critical component of threat mitigation, requires risk prediction based on fungal lifestyles and traits. Recent studies have revealed distinct genomic patterns associated with specific groups of plant pathogenic fungi. We sought to establish whether these phytopathogenic genomic patterns hold across diverse taxonomic and ecological groups from the Ascomycota and Basidiomycota, and furthermore, if those patterns can be used in a predictive capacity for biosurveillance. Using a supervised machine learning approach that integrates phylogenetic and genomic data, we analyzed 387 fungal genomes to test a proof-of-concept for the use of genomic signatures in predicting fungal phytopathogenic lifestyles and traits during biosurveillance activities. Our machine learning feature sets were derived from genome annotation data of carbohydrate-active enzymes (CAZymes), peptidases, secondary metabolite clusters (SMCs), transporters, and transcription factors. We found that machine learning could successfully predict fungal lifestyles and traits across taxonomic groups, with the best predictive performance coming from feature sets comprising CAZyme, peptidase, and SMC data. While phylogeny was an important component in most predictions, the inclusion of genomic data improved prediction performance for every lifestyle and trait tested. Plant pathogenicity was one of the best-predicted traits, showing the promise of predictive genomics for biosurveillance applications. Furthermore, our machine learning approach revealed expansions in the number of genes from specific CAZyme and peptidase families in the genomes of plant pathogens compared to non-phytopathogenic genomes (saprotrophs, endo- and ectomycorrhizal fungi). Such genomic feature profiles give insight into the evolution of fungal phytopathogenicity and could be useful to predict the risks of unknown fungi in future biosurveillance activities.

59 BASIC BIOLOGICAL SCIENCES↗

THE PREDICTIVE POWER OF OHL'S PRECURSOR METHOD

Using linear regression techniques and correlation analysis, the predictive power of Ohl's method is shown to satisfy one of the following relationships. Either a successful prediction of cycle amplitude can be obtained if the correlation between the minimum aa geomagnetic index in the declining phase of a solar cycle and the sunspot maximum of the succeeding cycle becomes stronger, or the prediction error exceeds the expected prediction error if the correlation becomes weaker. The correlation coefficient has a declining secular variation, which leads to a weakening trend in predictive power, as well as a 44 year periodicity that may explain why the prediction method did not work well for solar cycle 23. As this finding only emerged twice in the {approx}100 year period studied, this 44 year periodicity may occur with a degree of uncertainty and may therefore need to be checked in the future. Two quantities, namely a prediction parameter and the prediction index, are proposed to analyze predictive power. Using the above properties, the success probability of a prediction result can be analyzed in advance.

79 ASTRONOMY AND ASTROPHYSICS↗

Improving net ecosystem CO2 flux prediction using memory-based interpretable machine learning

Terrestrial ecosystems play a central role in the global carbon cycle and affect climate change. However, our predictive understanding of these systems is still limited due to their complexity and uncertainty about how key drivers and their legacy effects influence carbon fluxes. Here, we propose an interpretable Long Short-Term Memory (iLSTM) network for predicting net ecosystem CO 2 exchange (NEE) and interpreting the influence on the NEE prediction from environmental drivers and their memory effects. We consider five drivers and apply the method to three forest sites in the United States. Besides performing the prediction in each site, we also conduct transfer learning by using the iLSTM model trained in one site to predict at other sites. Results show that the iLSTM model produces good NEE predictions for all three sites and, more importantly, it provides reasonable interpretations on the input driver's importance as well as their temporal importance on the NEE prediction. Additionally, the iLSTM model demonstrates good across-site transferability in terms of both prediction accuracy and interpretability. The transferability can improve the NEE prediction in unobserved forest sites, and the interpretability advances our predictive understanding and guides process-based model development.

Liu, Siyan↗

ScaleML: Machine Learning based Heap Memory Object Scaling Prediction

Memory subsystem contributes 28-40% of total energy consumption. Several studies investigated energy prediction and consumption via profiling memory object access patterns. However, such profiling leads to higher energy consumption due to intense memory object-level profiling to achieve high prediction accuracy. Further, memory object access pattern prediction has been considered through analyzing the variation between memory object access patterns, referred to as scaling rate. The existing techniques for scaling rate prediction, such as Linear Scaling Rate (LSR), suffer from a high error rate in prediction with changes in access patterns, which leads to a high error rate of energy consumption prediction. In this paper, we compare and evaluate several memory object access pattern prediction models including LSR and machine learning (ML) models. Further, we propose SCALEML, a heap memory object scaling rate prediction mechanism that employs an ML model to achieve high access pattern prediction accuracy with variations in memory object access patterns. We evaluate SCALEML using various application benchmarks. The experimental results show that SCALEML achieves about 20% higher accuracy than the LSR model for predicting the scaling rate of object access patterns and energy estimation.

Park, Joongeon↗

Learning Heterogeneous Features Jointly: A Deep End-to-End Framework for Multi-Step Short-Term Wind Power Prediction

Leveraging multiple heterogeneous measurements to predict wind power has long been a challenging task in the electrical community. In this paper, a deep architecture incorporated with multitask learning and multimodal learning for wind power prediction, termed predictive stacked autoencoder (PSAE), is presented. PSAE is a unified framework integrating multiple stacked autoencoders (SAEs), one feature fusion layer, and one prediction terminal layer, which expands the architecture from two spatial dimensions, including the depth and width, compared to conventional prediction models. Initially, the SAEs at the bottom of PSAE extracted features from multiple kinds of measurements respectively. Following, the feature fusion layer encodes the high-order features extracted by different SAEs into a unified feature that is more informative and representative for wind power prediction. Finally, the prediction terminal layer functions as a regression machine which generates the predicted targets based on the fusion features. Trained in an end-to-end (E2E) manner, PSAE is capable of learning heterogeneous features jointly and achieving the prediction task of sequence-to-sequence (S2S). Experiments for multi-step short-term predictions are conducted on real-world data, and the results demonstrate the superiority of PSAE to prior methods.

17 WIND ENERGY↗

The seasonal climate niche predicts phenology and distribution of an ephemeral annual plant, Mollugo verticillata

Many short-lived species complete their life cycles during brief seasonal windows of favourable environmental conditions. Such species may persist in the face of climate warming by migration to track their seasonal climate niche in space and/or by phenological shifts to track favourable conditions in time within the year. To describe the seasonal climate niche of the short-lived annual Mollugo verticillata in California, we used data from herbarium specimens and historic climate records to estimate environmental conditions at the location, month and year of each collection. We used these data in a MaxEnt framework to construct a seasonal species distribution model (SDM) of the species’ climate niche within the total climate space available across all seasons and locations in California. The model provides fine-scale spatial and temporal predictions of habitat suitability, predicting both where and when the species should be observed. We compared the predictions of the model to those from a conventional SDM based on mean annual climate data. Both models showed that M. verticillata is limited to warm environments within California. However, the seasonal SDM also predicted phenology by mapping climate suitability across the state for each month of the year. Mollugo verticillata is limited to warm months, and its seasonal climate niche shifts in space across California in the course of the year. We used the seasonal SDM to map the predicted future species distribution for each month of the year under three warming scenarios. The species is predicted to expand its range and occur earlier in the year in most locations; in the warmest locations, seasonal suitability is predicted to decline in the warmest months, which may result in bimodal phenology with a mid-summer gap. Synthesis. We developed a novel species distribution model using herbarium records and monthly weather data, which predicts not only where a short-lived species should be found but also when during the year it is predicted to occur in those areas. Furthermore, this model can be used to predict how climate change will affect the species distribution in space as well as seasonal phenology across the landscape.

54 ENVIRONMENTAL SCIENCES↗

SU-E-J-191: Motion Prediction Using Extreme Learning Machine in Image Guided Radiotherapy

Purpose: Real-time motion tracking is a critical issue in image guided radiotherapy due to the time latency caused by image processing and system response. It is of great necessity to fast and accurately predict the future position of the respiratory motion and the tumor location. Methods: The prediction of respiratory position was done based on the positioning and tracking module in ARTS-IGRT system which was developed by FDS Team (www.fds.org.cn). An approach involving with the extreme learning machine (ELM) was adopted to predict the future respiratory position as well as the tumor’s location by training the past trajectories. For the training process, a feed-forward neural network with one single hidden layer was used for the learning. First, the number of hidden nodes was figured out for the single layered feed forward network (SLFN). Then the input weights and hidden layer biases of the SLFN were randomly assigned to calculate the hidden neuron output matrix. Finally, the predicted movement were obtained by applying the output weights and compared with the actual movement. Breathing movement acquired from the external infrared markers was used to test the prediction accuracy. And the implanted marker movement for the prostate cancer was used to test the implementation of the tumor motion prediction. Results: The accuracy of the predicted motion and the actual motion was tested. Five volunteers with different breathing patterns were tested. The average prediction time was 0.281s. And the standard deviation of prediction accuracy was 0.002 for the respiratory motion and 0.001 for the tumor motion. Conclusion: The extreme learning machine method can provide an accurate and fast prediction of the respiratory motion and the tumor location and therefore can meet the requirements of real-time tumor-tracking in image guided radiotherapy.

60 APPLIED LIFE SCIENCES↗

SU-E-T-629: Prediction of the ViewRay Radiotherapy Treatment Time for Clinical Logistics

Purpose: An algorithm is developed in our clinic, given a new treatment plan, to predict treatment delivery time for radiation therapy (RT) treatments of patients on ViewRay magnetic resonance-image guided radiation therapy (MR-IGRT) delivery system. This algorithm is necessary for managing patient treatment appointments, and is useful as an indicator to assess the treatment plan complexity. Methods: A patient’s total treatment delivery time, not including time required for localization, may be described as the sum of four components: (1) the treatment initialization time; (2) the total beam-on time; (3) the gantry rotation time; and (4) the multileaf collimator (MLC) motion time. Each of the four components is predicted separately. The total beam-on time can be calculated using both the planned beam-on time and the decay-corrected delivery dose rate. To predict the remaining components, we quantitatively analyze the patient treatment delivery record files. The initialization time is demonstrated to be random since it depends on the final gantry angle and MLC leaf positions of the previous treatment. Based on modeling the relationships between the gantry rotation angles and the corresponding rotation time, and between the furthest MLC leaf moving distance and the corresponding MLC motion time, the total delivery time is predicted using linear regression. Results: The proposed algorithm has demonstrated the feasibility of predicting the ViewRay treatment delivery time for any treatment plan of any patient. The average prediction error is 0.89 minutes or 5.34%, and the maximal prediction error is 2.09 minutes or 13.87%. Conclusion: We have developed a treatment delivery time prediction algorithm based on the analysis of previous patients’ treatment delivery records. The accuracy of our prediction is sufficient for guiding and arranging patient treatment appointments on a daily basis. The predicted delivery time could also be used as an indicator to assess the treatment plan complexity. This work was supported by a research grant from Viewray Inc.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

MO-FG-303-03: Demonstration of Universal Knowledge-Based 3D Dose Prediction

Purpose: To demonstrate a knowledge-based 3D dose prediction methodology that can accurately predict achievable radiotherapy distributions. Methods: Using previously treated plans as input, an artificial neural network (ANN) was trained to predict 3D dose distributions based on 14 patient-specific anatomical parameters including the distance (r) to planning target volume (PTV) boundary, organ-at-risk (OAR) boundary distances, and angular position ( θ,φ). 23 prostate and 49 stereotactic radiosurgery (SRS) cases with ≥1 nearby OARs were studied. All were planned with volumetric-modulated arc therapy (VMAT) to prescription doses of 81Gy for prostate and 12–30Gy for SRS. Site-specific ANNs were trained using all prostate 23 plans and using a 24 randomly-selected subset for the SRS model. The remaining 25 SRS plans were used to validate the model. To quantify predictive accuracy, the dose difference between the clinical plan and prediction were calculated on a voxel-by-voxel basis δD(r,θ,φ)=Dclin(r,θ,φ)-Dpred(r, θ,φ). Grouping voxels by boundary distance, the mean <δ Dr>=(1/N)Σ -θ,φ D(r,θ,φ) and inter-quartile range (IQR) quantified the accuracy of this method for deriving DVH estimations. The standard deviation (σ) of δ D quantified the 3D dose prediction error on a voxel-by-voxel basis. Results: The ANNs were highly accurate in predictive ability for both prostate and SRS plans. For prostate, <δDr> ranged from −0.8% to +0.6% (max IQR=3.8%) over r=0–32mm, while 3D dose prediction accuracy averaged from σ=5–8% across the same range. For SRS, from r=0–34mm the training set <δDr> ranged from −3.7% to +1.5% (max IQR=4.4%) while the validation set <δDr> ranged from −2.2% to +5.8% (max IQR=5.3%). 3D dose prediction accuracy averaged σ=2.5% for the training set and σ=4.0% over the same interval. Conclusion: The study demonstrates this technique’s ability to predict achievable 3D dose distributions for VMAT SRS and prostate. Future investigations will attempt to separate and quantify modeling error and plan quality variations in the statistical noise. Research funding support from Varian Medical Systems.

60 APPLIED LIFE SCIENCES↗

SU-F-T-348: The Impact of Model Library Population On RapidPlan Based Dose-Volume Histograms (DVHs) Prediction for Rectal Cancer Patients Treated with Volumetric-Modulated Radiotherapy (VMAT)

Purpose: RapidPlan uses a library consisting of expert plans from different patients to create a model that can predict achievable dose-volume histograms (DVHs) for new patients. The goal of this study is to investigate the impacts of model library population (plan numbers) on the DVH prediction for rectal cancer patients treated with volumetric-modulated radiotherapy (VMAT) Methods: Ninety clinically accepted rectal cancer patients’ VMAT plans were selected to establish 3 models, named as Model30, Model60 and Model90, with 30,60, and 90 plans in the model training. All plans had sufficient target coverage and bladder and femora sparings. Additional 10 patients were enrolled to test the DVH prediction differences with these 3 models. The predicted DVHs from these 3 models were compared and analyzed. Results: Predicted V40 (Vx, percent of volume that received x Gy for the organs at risk) and Dmean (mean dose, cGy) of the bladder were 39.84±13.38 and 2029.4±141.6 for the Model30,37.52±16.00 and 2012.5±152.2 for the Model60, and 36.33±18.35 and 2066.5±174.3 for the Model90. Predicted V30 and Dmean of the left femur were 23.33±9.96 and 1443.3±114.5 for the Model30, 21.83±5.75 and 1436.6±61.9 for the Model60, and 20.31±4.6 and 1415.0±52.4 for the Model90.There were no significant differences among the 3 models for the bladder and left femur predictions. Predicted V40 and Dmean of the right femur were 19.86±10.00 and 1403.6±115.6 (Model30),18.97±6.19 and 1401.9±68.78 (Model60), and 21.08±7.82 and 1424.0±85.3 (Model90). Although a slight lower DVH prediction of the right femur was found on the Model60, the mean differences for V30 and mean dose were less than 2% and 1%, respectively. Conclusion: There were no significant differences among Model30, Model60 and Model90 for predicting DVHs on rectal patients treated with VMAT. The impact of plan numbers for model library might be limited for cancers with similar target shape.

60 APPLIED LIFE SCIENCES↗

Dominant Time Scales of Potentially Predictable Precipitation Variations across the Continental United States

While low-frequency variations in precipitation amount, occurrence counts (hereafter “occurrence”), and intensity can take place on seasonal to multidecadal time scales, it is often unclear at which time scales these precipitation variations can be ascribed to potentially predictable, climate-induced changes versus simple, stochastic (i.e., random) precipitation event evolutions. This paper seeks to isolate the dominant time scales at which potentially predictable changes in observed precipitation characteristics occur over the continental United States and analyze sources of revealed potentially predictable precipitation variations for particular regions. The results highlight that at interannual time scales (here defined as those shorter than 7 years), the potential for predicting annual precipitation amounts tends to be higher than for annual event occurrence or intensity, with interannual potential predictability highest in both relatively dry and wet locations and lowest in transition regions. By contrast, at time scales greater than 7 years the potential for predicting annual event occurrence tends to be higher than amount or intensity, with >20-yr time scale potential predictability highest in relatively wet locations and lowest in relatively dry locations. To highlight the utility of this type of analysis, two robust signals are selected for further investigation, including 1) approximately 10-yr time scale variations in potentially predictable annual amounts over the northwestern United States and 2) 20–60-yr time scale variations in potentially predictable annual event occurrence over the southwestern United States. While mechanistic drivers for these observed variations are still being investigated, concurrent and precursor climate-state estimates in the atmosphere and ocean—principally over the Pacific sector—are provided, the monitoring of which may help realize the potential for predicting precipitation variations in these regions.

54 ENVIRONMENTAL SCIENCES↗

Comparison of Machine Learning-Based Predictive Models of the Nutrient Loads Delivered from the Mississippi/Atchafalaya River Basin to the Gulf of Mexico

Predicting nutrient loads is essential to understanding and managing one of the environmental issues faced by the northern Gulf of Mexico hypoxic zone, which poses a severe threat to the Gulf’s healthy ecosystem and economy. The development of hypoxia in the Gulf of Mexico is strongly associated with the eutrophication process initiated by excessive nutrient loads. Due to the complexities in the excessive nutrient loads to the Gulf of Mexico, it is challenging to understand and predict the underlying temporal variation of nutrient loads. The study was aimed at identifying an optimal predictive machine learning model to capture and predict nonlinear behavior of the nutrient loads delivered from the Mississippi/Atchafalaya River Basin (MARB) to the Gulf of Mexico. For this purpose, monthly nutrient loads (N and P) in tons were collected from US Geological Survey (USGS) monitoring station 07373420 from 1980 to 2020. Machine learning models—including autoregressive integrated moving average (ARIMA), gaussian process regression (GPR), single-layer multilayer perceptron (MLP), and a long short-term memory (LSTM) with the single hidden layer—were developed to predict the monthly nutrient loads, and model performances were evaluated by standard assessment metrics—Root Mean Square Error (RMSE) and Correlation Coefficient (R). The residuals of predictive models were examined by the Durbin–Watson statistic. The results showed that MLP and LSTM persistently achieved better accuracy in predicting monthly TN and TP loads compared to GPR and ARIMA. In addition, GPR models achieved slightly better test RMSE score than ARIMA models while their correlation coefficients are much lower than ARIMA models. Moreover, MLP performed slightly better than LSTM in predicting monthly TP loads while LSTM slightly outperformed for TN loads. Furthermore, it was found that the optimizer and number of inputs didn’t show effects on the LSTM performance while they exhibited impacts on MLP outcomes. This study explores the capability of machine learning models to accurately predict nonlinearly fluctuating nutrient loads delivered to the Gulf of Mexico. Further efforts focus on improving the accuracy of forecasting using hybrid models which combine several machine learning models with superior predictive performance for nutrient fluxes throughout the MARB.

54 ENVIRONMENTAL SCIENCES↗

Leveraging 35 years of Pinus taeda research in the southeastern US to constrain forest carbon cycle predictions: regional data assimilation using ecosystem experiments

Predicting how forest carbon cycling will change in response to climate change and management depends on the collective knowledge from measurements across environmental gradients, ecosystem manipulations of global change factors, and mathematical models. Formally integrating these sources of knowledge through data assimilation, or model–data fusion, allows the use of past observations to constrain model parameters and estimate prediction uncertainty. Data assimilation (DA) focused on the regional scale has the opportunity to integrate data from both environmental gradients and experimental studies to constrain model parameters. Here, we introduce a hierarchical Bayesian DA approach (Data Assimilation to Predict Productivity for Ecosystems and Regions, DAPPER) that uses observations of carbon stocks, carbon fluxes, water fluxes, and vegetation dynamics from loblolly pine plantation ecosystems across the southeastern US to constrain parameters in a modified version of the Physiological Principles Predicting Growth (3-PG) forest growth model. The observations included major experiments that manipulated atmospheric carbon dioxide (CO 2 ) concentration, water, and nutrients, along with nonexperimental surveys that spanned environmental gradients across an 8.6 × 10 5 km 2 region. We optimized regionally representative posterior distributions for model parameters, which dependably predicted data from plots withheld from the data assimilation. While the mean bias in predictions of nutrient fertilization experiments, irrigation experiments, and CO 2 enrichment experiments was low, future work needs to focus modifications to model structures that decrease the bias in predictions of drought experiments. Predictions of how growth responded to elevated CO 2 strongly depended on whether ecosystem experiments were assimilated and whether the assimilated field plots in the CO 2 study were allowed to have different mortality parameters than the other field plots in the region. We present predictions of stem biomass productivity under elevated CO 2 , decreased precipitation, and increased nutrient availability that include estimates of uncertainty for the southeastern US. Overall, we (1) demonstrated how three decades of research in southeastern US planted pine forests can be used to develop DA techniques that use multiple locations, multiple data streams, and multiple ecosystem experiment types to optimize parameters and (2) developed a tool for the development of future predictions of forest productivity for natural resource managers that leverage a rich dataset of integrated ecosystem observations across a region.

54 ENVIRONMENTAL SCIENCES↗

The Decadal Climate Prediction Project (DCPP) contribution to CMIP6

The Decadal Climate Prediction Project (DCPP) is a coordinated multi-model investigation into decadal climate prediction, predictability, and variability. The DCPP makes use of past experience in simulating and predicting decadal variability and forced climate change gained from the fifth Coupled Model Intercomparison Project (CMIP5) and elsewhere. It builds on recent improvements in models, in the reanalysis of climate data, in methods of initialization and ensemble generation, and in data treatment and analysis to propose an extended comprehensive decadal prediction investigation as a contribution to CMIP6 (Eyring et al., 2016) and to the WCRP Grand Challenge on Near Term Climate Prediction (Kushnir et al., 2016). The DCPP consists of three components. Component A comprises the production and analysis of an extensive archive of retrospective forecasts to be used to assess and understand historical decadal prediction skill, as a basis for improvements in all aspects of end-to-end decadal prediction, and as a basis for forecasting on annual to decadal timescales. Component B undertakes ongoing production, analysis and dissemination of experimental quasi-real-time multi-model forecasts as a basis for potential operational forecast production. Component C involves the organization and coordination of case studies of particular climate shifts and variations, both natural and naturally forced (e.g. the “hiatus”, volcanoes), including the study of the mechanisms that determine these behaviours. Furthermore, groups are invited to participate in as many or as few of the components of the DCPP, each of which are separately prioritized, as are of interest to them. The Decadal Climate Prediction Project addresses a range of scientific issues involving the ability of the climate system to be predicted on annual to decadal timescales, the skill that is currently and potentially available, the mechanisms involved in long timescale variability, and the production of forecasts of benefit to both science and society.

54 ENVIRONMENTAL SCIENCES↗

Predictive performance of multi-model ensemble forecasts of COVID-19 across European nations

Background: Short-term forecasts of infectious disease burden can contribute to situational awareness and aid capacity planning. Based on best practice in other fields and recent insights in infectious disease epidemiology, one can maximise the predictive performance of such forecasts if multiple models are combined into an ensemble. Here, we report on the performance of ensembles in predicting COVID-19 cases and deaths across Europe between 08 March 2021 and 07 March 2022. Methods: We used open-source tools to develop a public European COVID-19 Forecast Hub. We invited groups globally to contribute weekly forecasts for COVID-19 cases and deaths reported by a standardised source for 32 countries over the next 1–4 weeks. Teams submitted forecasts from March 2021 using standardised quantiles of the predictive distribution. Each week we created an ensemble forecast, where each predictive quantile was calculated as the equally-weighted average (initially the mean and then from 26th July the median) of all individual models’ predictive quantiles. We measured the performance of each model using the relative Weighted Interval Score (WIS), comparing models’ forecast accuracy relative to all other models. We retrospectively explored alternative methods for ensemble forecasts, including weighted averages based on models’ past predictive performance. Results: Over 52 weeks, we collected forecasts from 48 unique models. We evaluated 29 models’ forecast scores in comparison to the ensemble model. We found a weekly ensemble had a consistently strong performance across countries over time. Across all horizons and locations, the ensemble performed better on relative WIS than 83% of participating models’ forecasts of incident cases (with a total N=886 predictions from 23 unique models), and 91% of participating models’ forecasts of deaths (N=763 predictions from 20 models). Across a 1–4 week time horizon, ensemble performance declined with longer forecast periods when forecasting cases, but remained stable over 4 weeks for incident death forecasts. In every forecast across 32 countries, the ensemble outperformed most contributing models when forecasting either cases or deaths, frequently outperforming all of its individual component models. Among several choices of ensemble methods we found that the most influential and best choice was to use a median average of models instead of using the mean, regardless of methods of weighting component forecast models. Conclusions: Our results support the use of combining forecasts from individual models into an ensemble in order to improve predictive performance across epidemiological targets and populations during infectious disease epidemics. Our findings further suggest that median ensemble methods yield better predictive performance more than ones based on means. Our findings also highlight that forecast consumers should place more weight on incident death forecasts than incident case forecasts at forecast horizons greater than 2 weeks. Funding: AA, BH, BL, LWa, MMa, PP, SV funded by National Institutes of Health (NIH) Grant 1R01GM109718, NSF BIG DATA Grant IIS-1633028, NSF Grant No.: OAC-1916805, NSF Expeditions in Computing Grant CCF-1918656, CCF-1917819, NSF RAPID CNS-2028004, NSF RAPID OAC-2027541, US Centers for Disease Control and Prevention 75D30119C05935, a grant from Google, University of Virginia Strategic Investment Fund award number SIF160, Defense Threat Reduction Agency (DTRA) under Contract No. HDTRA1-19-D-0007, and respectively Virginia Dept of Health Grant VDH-21-501-0141, VDH-21-501-0143, VDH-21-501-0147, VDH-21-501-0145, VDH-21-501-0146, VDH-21-501-0142, VDH-21-501-0148. AF, AMa, GL funded by SMIGE - Modelli statistici inferenziali per governare l'epidemia, FISR 2020-Covid-19 I Fase, FISR2020IP-00156, Codice Progetto: PRJ-0695. AM, BK, FD, FR, JK, JN, JZ, KN, MG, MR, MS, RB funded by Ministry of Science and Higher Education of Poland with grant 28/WFSN/2021 to the University of Warsaw. BRe, CPe, JLAz funded by Ministerio de Sanidad/ISCIII. BT, PG funded by PERISCOPE European H2020 project, contract number 101016233. CP, DL, EA, MC, SA funded by European Commission - Directorate-General for Communications Networks, Content and Technology through the contract LC-01485746, and Ministerio de Ciencia, Innovacion y Universidades and FEDER, with the project PGC2018-095456-B-I00. DE., MGu funded by Spanish Ministry of Health / REACT-UE (FEDER). DO, GF, IMi, LC funded by Laboratory Directed Research and Development program of Los Alamos National Laboratory (LANL) under project number 20200700ER. DS, ELR, GG, NGR, NW, YW funded by National Institutes of General Medical Sciences (R35GM119582; the content is solely the responsibility of the authors and does not necessarily represent the official views of NIGMS or the National Institutes of Health). FB, FP funded by InPresa, Lombardy Region, Italy. HG, KS funded by European Centre for Disease Prevention and Control. IV funded by Agencia de Qualitat i Avaluacio Sanitaries de Catalunya (AQuAS) through contract 2021-021OE. JDe, SMo, VP funded by Netzwerk Universitatsmedizin (NUM) project egePan (01KX2021). JPB, SH, TH funded by Federal Ministry of Education and Research (BMBF; grant 05M18SIA). KH, MSc, YKh funded by Project SaxoCOV, funded by the German Free State of Saxony. Presentation of data, model results and simulations also funded by the NFDI4Health Task Force COVID-19 ( https://www.nfdi4health.de/task-force-covid-19-2 ) within the framework of a DFG-project (LO-342/17-1). LP, VE funded by Mathematical and Statistical modelling project (MUNI/A/1615/2020), Online platform for real-time monitoring, analysis and management of epidemic situations (MUNI/11/02202001/2020); VE also supported by RECETOX research infrastructure (Ministry of Education, Youth and Sports of the Czech Republic: LM2018121), the CETOCOEN EXCELLENCE (CZ.02.1.01/0.0/0.0/17-043/0009632), RECETOX RI project (CZ.02.1.01/0.0/0.0/16-013/0001761). NIB funded by Health Protection Research Unit (grant code NIHR200908). SAb, SF funded by Wellcome Trust (210758/Z/18/Z).

60 APPLIED LIFE SCIENCES↗