Search NASA⌕ Search

SEARCH · Search NASA

Results for “Predictors”

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 1,207 records · Page 67

Timing and extent of crop damage by wild pigs (Sus scrofa Linnaeus) to corn and peanut fields

The global expansion of wild pigs over the last few decades has resulted in an increase in extent and distribution of damages to crops, placing a growing strain on agricultural producers and land managers. Despite the extent of wild pig damage to agriculture, there is little data regarding timing and spatial variability of damage to corn (Zea mays Linnaeus) and we found no data regarding the effect of these factors on peanuts (Arachis hypogaea Linnaeus). Our objective was to determine the timing and extent of wild pig damage to corn and peanut fields, as well as the extent to which local habitat attributes are useful predictors of damage to these crops. During 2017–2018 we performed ground-based surveys throughout the growing season for 29 corn and 41 peanut fields in South Carolina, USA to determine the most important growth stages for wild pig depredation in both crops. Damage to corn peaked shortly after planting during the seedling stage, fell to nearly zero during V4–V6 stages, and resumed during the silk and mature stages. Peanut damage was almost exclusively limited to the seedling stage. Landscape models for both crops identified the extent of forested and wetland areas surrounding crop fields as the most important attributes positively associated with wild pig damage, while the amount of adjacent agricultural area and paved roads were associated negatively. The number of wild pigs identified by remote cameras also was an important indicator of the extent of damage to peanut fields. Results suggest management efforts to limit crop depredation by wild pigs should be targeted shortly prior to planting. Further, because damage was positively associated with the availability of wetland and forest habitats, our results suggest agricultural damage by wild pigs may be most severe near areas of preferred native habitats.

54 ENVIRONMENTAL SCIENCES↗

BrainXcan identifies brain features associated with behavioral and psychiatric traits using large-scale genetic and imaging data

Advances in brain MRI have enabled many discoveries in neuroscience. Case-control comparisons of brain MRI features have highlighted potential causes of psychiatric and behavioral disorders. However, due to the cost and difficulty of collecting MRI data, most studies have small sample sizes, limiting their reliability. Furthermore, reverse causality complicates interpretation because many observed brain differences are the result rather than the cause of the disease. Here we propose a method (BrainXcan) that leverages the power of large-scale genomewide association studies (GWAS) and reference brain MRI data to discover new mechanisms of disease etiology and validate existing ones. BrainXcan tests the association with genetic predictors of brain MRI-derived features and complex traits to pinpoint relevant brain-wide and region-specific features. Requiring only genetic data, BrainXcan allows us to test a host of hypotheses on mental illness, across many MRI modalities, using public data resources. For example, our method shows that reduced axonal density across the brain is associated with schizophrenia risk, consistent with the disconnectivity hypothesis. We also find that the hippocampus volume is associated with schizophrenia risk, highlighting the potential of our approach. Taken together, our results show the promise of BrainXcan to provide insights into the biology of GWAS traits.

Association study↗

Predictive models of long COVID

Background: The cause and symptoms of long COVID are poorly understood. It is challenging to predict whether a given COVID-19 patient will develop long COVID in the future. Methods: We used electronic health record (EHR) data from the National COVID Cohort Collaborative to predict the incidence of long COVID. We trained two machine learning (ML) models — logistic regression (LR) and random forest (RF). Features used to train predictors included symptoms and drugs ordered during acute infection, measures of COVID-19 treatment, pre-COVID comorbidities, and demographic information. We assigned the ‘long COVID’ label to patients diagnosed with the U09.9 ICD10-CM code. The cohorts included patients with (a) EHRs reported from data partners using U09.9 ICD10-CM code and (b) at least one EHR in each feature category. We analysed three cohorts: all patients (n = 2,190,579; diagnosed with long COVID = 17,036), inpatients (149,319; 3,295), and outpatients (2,041,260; 13,741). Findings: LR and RF models yielded median AUROC of 0.76 and 0.75, respectively. Ablation study revealed that drugs had the highest influence on the prediction task. The SHAP method identified age, gender, cough, fatigue, albuterol, obesity, diabetes, and chronic lung disease as explanatory features. Models trained on data from one N3C partner and tested on data from the other partners had average AUROC of 0.75. Interpretation: ML-based classification using EHR information from the acute infection period is effective in predicting long COVID. SHAP methods identified important features for prediction. Cross-site analysis demonstrated the generalizability of the proposed methodology.

60 APPLIED LIFE SCIENCES↗

Visions of success and achievement in recreation-related USDA Forest Service NEPA processes

The National Environmental Policy Act (NEPA) is incorporated into the planning and decision-making culture of all natural resource agencies in the U.S. Yet, we know little about how the attitudes and internal interactions of interdisciplinary (ID) teams engaged in NEPA processes influence process outcomes. We conducted a web-based survey of 106 ID team leaders involved with environmental analyses (EA) or environmental impact statements (EIS) for projects dealing with recreation and travel management on national forests. We explore how they define success in these processes and identify factors most powerfully associated with perceptions of positive outcomes. The survey revealed a tremendous diversity in definitions of success. Strong correlations between the perceived importance of particular indicators of success and their achievement suggest that pre-conceived notions may often help to shape process outcomes. Regression analyses revealed the following factors as the best predictors of ID team leaders' perception of an 'excellent outcome': achievement of the agency mission, whether compromise had taken place between the interested parties, team satisfaction and harmony, timely process completion, and project implementation. Yet, respondents consistently ranked compromise with interested parties and team member satisfaction among the least important measures of successful NEPA processes. Results suggest that clarifying appropriate measures of success in NEPA processes across the agency could make ID team performance more consistent. The research also suggests that greater attention to ID team interactions, both internally and between teams and interested publics, could result in better outcomes.

54 ENVIRONMENTAL SCIENCES↗

Appraising the sustainability of project alternatives: An increasing role for cumulative effects assessment

Evaluating and comparing development alternatives with regard to sustainability is an important goal for comprehensive project appraisal. In the United States, this component has been largely missing from standard environmental impact assessment practice. Cumulative effects assessment provides a way to appraise the sustainability of project alternatives in terms of their probable contributions to long-term trends affecting the condition of valued environmental components. Sustainability metrics and predictors are being developed as criteria for rating systems and evaluation processes that are applied to community planning, building design, and transportation infrastructure. Increasing interest in adaptive management is also providing cost-effective solutions to optimizing safety and function throughout the long-term operation of a facility or infrastructure. Recent federal legislation is making it easier to integrate sustainability features into development alternatives through early, community-based planning.

54 ENVIRONMENTAL SCIENCES↗

Project risk and appeals in U.S. Forest Service planning

The National Environmental Policy Act (NEPA) requires U.S. Forest Service planning processes to be conducted by interdisciplinary teams of resource specialists to analyze and disclose the likely environmental impacts of proposed natural resource management actions on Forest Service lands. Multiple challenges associated with these processes have been a source of frustration for the agency. One of these challenges involves administrative appeals through which public entities can challenge a Forest Service decision following a NEPA process. These appeals instigate an internal review process and can result in an affirmation of the Forest Service decision, a reversal of that decision, or additional work that re-initiates all or part of the NEPA process. We examine the best predictors of appeals and their outcomes on a representative sample of 489 Forest Service NEPA processes that were decided between 2007 and 2009. While certain factors associated with pre-existing social contexts (such as a history of controversy) or pre-determined elements of a proposed action (such as the extraction of forest products) predispose certain processes to a higher risk of appeals, other practices and process-related strategies within the control of the agency also appear to bear meaningful influence on the occurrence of appeals and their outcomes. Appeals and their outcomes were most strongly related to programmatic, structural (turnover of personnel in particular), and relationship risks (both internal and external) within the processes, suggesting the need for greater focus within the agency on cultivating positive internal and external relationships to manage the risk of appeals. -- Highlights: ► We examined appeals and their outcomes on 489 U.S. Forest Service NEPA processes. ► Project type, context, team turnover, and personal relationships predicted appeals. ► External relationship management and staff turnover best predicted appeal outcomes. ► Positive internal and external relationships appear to reduce appeal risks.

54 ENVIRONMENTAL SCIENCES↗

Evaluation of noise pollution in urban traffic hubs—Noise maps and measurements

A study was made of some of the main traffic hubs in a Latin American metropolis, in order to determine the presence or absence of noise by means of noise measurements and acoustic mapping. To characterize noise in the evaluated road stretches, 232 measurements were taken at different points. The Predictor software package was used for the noise mapping calculations. Noise sensitive areas, e.g., hospitals, were identified in the evaluated road stretches. Noise maps were calculated for two hospitals, showing the current levels of noise that reach their facades. Hypothetical scenarios were simulated by making changes in the composition of traffic and total number of vehicles, and an assessment was made of the potential influence of these modifications in reducing the noise levels reaching the facades of the buildings in question. The simulations indicated that a 50% reduction in total traffic flow, or a 50% reduction in heavy vehicle traffic flow, would reduce the noise levels by about 3 dB(A). - Highlights: • Evaluation of noise pollution in urban traffic hubs • Street systems • Environmental noise impacts • Noise mapping.

54 ENVIRONMENTAL SCIENCES↗

Multivariate pluvial flood damage models

Depth–damage-functions, relating the monetary flood damage to the depth of the inundation, are commonly used in the case of fluvial floods (floods caused by a river overflowing). We construct four multivariate damage models for pluvial floods (caused by extreme rainfall) by differentiating on the one hand between ground floor floods and basement floods and on the other hand between damage to residential buildings and damage to housing contents. We do not only take into account the effect of flood-depth on damage, but also incorporate the effects of non-hazard indicators (building characteristics, behavioural indicators and socio-economic variables). By using a Tobit-estimation technique on identified victims of pluvial floods in Flanders (Belgium), we take into account the effect of cases of reported zero damage. Our results show that the flood depth is an important predictor of damage, but with a diverging impact between ground floor floods and basement floods. Also non-hazard indicators are important. For example being aware of the risk just before the water enters the building reduces content damage considerably, underlining the importance of warning systems and policy in this case of pluvial floods. - Highlights: • Prediction of damage of pluvial floods using also non-hazard information • We include ‘no damage cases’ using a Tobit model. • The damage of flood depth is stronger for ground floor than for basement floods. • Non-hazard indicators are especially important for content damage. • Potential gain of policies that increase awareness of flood risks.

54 ENVIRONMENTAL SCIENCES↗

A machine learning pipeline for identifying infiltration managed aquifer recharge locations from satellite imagery in the San Joaquin Valley, California

This study focuses on an agricultural region in California’s Central Valley, USA, where Managed Aquifer Recharge (MAR) is widely implemented to mitigate groundwater depletion under increasing water demand and climate variability. A deep learning and machine learning framework was developed to identify infiltration-MAR locations using satellite imagery and environmental data. The framework integrates surface water detection from Sentinel-2 imagery, geospatial delineation of water bodies, spatiotemporal tracking of water body dynamics, and supervised classification using meteorological, environmental, and topographic variables. The framework was applied to a 2379 km² study area southwest of Fresno, where 765 water bodies were detected, including 139 identified MAR sites based on publicly available datasets and expert knowledge. The classification model achieved an accuracy of 0.94 and an F1 score of 0.85. Feature importance analysis indicates that cropland, normalized difference vegetation index (NDVI), and evaporation are among the most influential predictors for infiltration-MAR. Notably, the framework suggests that engineered water management in infiltration-MAR systems can disrupt or even reverse the expected positive correlation between surface water extent and precipitation. These findings provide physically interpretable insights into the characteristics of existing infiltration-MAR facilities and demonstrate the potential of the proposed framework as a reproducible, interpretable, and potentially transferable tool for data-driven infiltration-MAR identification and inventory development under growing climatic and hydrological uncertainty.

Classification↗

Effective moisture penetration depth model for residential buildings: Sensitivity analysis and guidance on model inputs

Moisture buffering of building materials has a significant impact on the building's indoor humidity, and building energy simulations need to model this buffering to accurately predict the humidity. Researchers requiring a simple moisture-buffering approach typically rely on the effective-capacitance model, which has been shown to be a poor predictor of actual indoor humidity. This paper describes an alternative two-layer effective moisture penetration depth (EMPD) model and its inputs. While this model has been used previously, there is a need to understand the sensitivity of this model to uncertain inputs. In this paper, we use the moisture-adsorbent materials exposed to the interior air: drywall, wood, and carpet. We use a global sensitivity analysis to determine which inputs are most influential and how the model's prediction capability degrades due to uncertainty in these inputs. We then compare the model's humidity prediction with measured data from five houses, which shows that this model, and a set of simple inputs, can give reasonable prediction of the indoor humidity.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Exploratory analysis and performance prediction of big data transfer in High-performance Networks

Big data transfer in large-scale scientific and business applications is increasingly carried out over connections with guaranteed bandwidth provisioned in High-performance Networks (HPNs) via advance bandwidth reservation. Provisioning agents need to carefully schedule data transfer requests, compute network paths, and allocate appropriate bandwidths. Such reserved bandwidths, if not fully utilized, could be simply wasted due to the exclusive access during the approved time window, and cause extra overhead and complexity for resource management. This calls for accurate performance prediction to reserve bandwidths that match actual needs and avoid over-provisioning. We employ machine learning algorithms to predict big data transfer performance based on extensive performance measurements collected in the past several years from data transfer tests using different protocols and toolkits between various end sites on several real-life physical or emulated testbeds. We first analyze the performance patterns in response to a comprehensive list of parameters in end-host systems, network connections, and data transfer applications, which motivate the use of machine learning and also help us identify the effects of latent factors. We then propose threshold- and clustering-based methods to eliminate negative effects of latent factors in data preprocessing and build a robust performance predictor based on customized domain-oriented loss functions. The performance of the proposed methods is verified by extensive experiments using SVR and RFR as well as theoretical analysis of the general performance bound.

97 MATHEMATICS AND COMPUTING↗

An Affinity-to-Commons Model of Public Support For Environmental Energy Policy

As atmospheric CO 2 continues to rise above 450 PPM, policymakers struggle with uncertainty concerning predictors of citizen support for environmental energy policies (EEPs) and preferences for their design, topics which have received limited attention in empirical literature. Here we present an original model of policy support based on citizens’ affinity-to-commons: pathways by which individuals enjoy natural public goods that in turn shape preferences between alternative policy mechanisms. We evaluate this model using a survey of southern California electricity customers, with results indicating the model's utility in predicting public support of EEP. Stronger community ties are associated with preferences for “pull”-type subsidies, whereas stronger connections to natural commons are linked to support for both “pull” and “push”-type sanctions. Findings have implications for coalition building as advocates may engender support for green energy policy by framing sanctions as protecting natural commons, and framing subsidies either in this same way and/or as producing benefits for communities.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Leveraging hyperspectral phenotyping for accurate, non-destructive prediction of metabolite profiles in poplar under drought stress

Accurately predicting drought tolerance in woody perennial bioenergy crops is critical for sustainable biomass production under fluctuating precipitation. Hyperspectral imaging (HSI) in the visible-near-infrared (VNIR) and shortwave-infrared (SWIR) ranges offers a promising approach for predicting plant biochemical traits, yet its application in metabolite profiling remains underexplored. We integrated VNIR+SWIR HSI with untargeted metabolomics to investigate drought-induced metabolic shifts in Populus leaves from eight Populus genotypes. Metabolite profiling identified 127 compounds, with 73 showing significant drought responses spanning amino acids (AA), carbohydrates (CHO), phenolic glycosides (PG), organic acids (OA), fatty acids and alcohols (FA), terpenes (T), phenolic metabolites (P), and unclassified metabolites. Spectral analysis revealed consistently higher reflectance across VNIR and SWIR wavelengths in drought-stressed plants, corresponding with increased accumulation of AA and reduced CHO and PG levels. Least absolute shrinkage and selection operator (LASSO) regression modeling identified robust spectral predictors of metabolite concentrations, associating VNIR wavelengths (500–700 nm) predominantly with AA and P, whereas SWIR wavelengths (1680–1700 nm) reliably predicted CHO, OA, and T. Several stable spectral-metabolite associations persisted across the two watering regimes (drought vs. well-watered), highlighting their potential as spectral biomarkers for non-destructive stress monitoring. Minimal genotype-specific variation suggests that observed spectral and metabolic responses were driven primarily by environmental factors, likely reflecting limited genetic diversity among the commercial Populus genotypes examined. This work establishes VNIR+SWIR hyperspectral imaging as a powerful, non-destructive phenotyping tool for precision monitoring and targeted improvement of drought resilience in bioenergy crops.

Biochemical trait prediction↗

Characterizing and predicting ultrafine particle counts in Canadian classrooms during the winter months: Model development and evaluation

School classrooms are potentially important micro-environments for childhood exposures owing to the large amount of time children spend in these locations. While a number of airborne contaminants may be present in schools, to date few studies have examined ultrafine particle (0.02-1 {mu}m) (UFP) levels in classrooms. In this study, our objective was to characterize UFP counts (cm{sup -3}) in classrooms during the winter months and to develop a model to predict such exposures based on ambient weather conditions and outdoor UFPs, as well as classroom characteristics such as size, temperature, relative humidity, and carbon dioxide levels. In total, UFP count data were collected on 60 occasions in 37 occupied classrooms at one elementary school and one secondary school in Pembroke, Ontario. On average, outdoor UFP levels exceeded indoor measures by 8989 cm{sup -3} (95% confidence interval (CI): 6382, 11 596), and classroom UFP counts were similar at both schools with a combined average of 5017 cm{sup -3} (95% CI: 4300, 5734). Of the variables examined only wind speed and outdoor UFPs were important determinants of classrooms UFP levels. Specifically, each 10 km/h increase in wind speed corresponded to an 1873 cm{sup -3} (95% CI: 825, 2920) decrease in classroom UFP counts, and each 10 000 cm{sup -3} increase in outdoor UFPs corresponded to a 1550 cm{sup -3} (95% CI: 930, 2171) increase in classroom UFP levels. However, high correlations between these two predictors meant that the independent effects of wind speed and outdoor UFPs could not be separated in multivariable models, and only outdoor UFP counts were included in the final predictive model. To evaluate model performance, classroom UFP counts were collected for 8 days at two new schools and compared to predicted values based on outdoor UFP measures. A moderate correlation was observed between measured and predicted classroom UFP counts (r=0.63) for both schools combined, but this relationship was not valid on days in which a strong indoor UFP source (electric kitchen stove) was active in schools. In general, our findings suggest that reasonable estimates of classroom UFP counts may be obtained from outdoor UFP data but that the accuracy of such estimates are limited in the presence of indoor UFP sources.

60 APPLIED LIFE SCIENCES↗

Uptake of cadmium from Pacific oysters (Crassostrea gigas) in British Columbia oyster growers

Background: Pacific oysters along the North American coast from Washington to Alaska contain concentrations of cadmium (Cd) that are high by comparison with Atlantic oysters, frequently exceeding 2 {mu}g/g wet weight, but it is unclear whether this Cd is absorbed by consumers. Objectives: To determine the effect of oyster consumption on Cd in blood and urine among a group with high oyster consumption. Methods: Sixty-one non-smoking oyster growers and family members with a mean age of 47.3{+-}7.6 years (range 33-64) were interviewed by telephone to assess their oyster consumption and other sources of Cd exposure at present and 5 years prior to the start of oyster farming. Their blood and urine Cd concentrations were measured. Results: The geometric mean Cd concentration in blood was 0.83 {mu}g/L and in urine was 0.76 {mu}g/g creatinine. Thirty-six percent of participants had urinary Cd levels above 1 {mu}g/g creatinine and 5% were above 2 {mu}g/g creatinine. Recent (last 12 months) and long-term oyster consumptions were positive predictors of blood Cd but did not directly predict urinary Cd. The optimal model for predicting the variance in blood Cd included recent intake of oyster-derived Cd, serum iron concentration and recent ketchup consumption (R{sup 2}=0.34, p=0.00004), with the latter two variables showing a protective effect. The factors found to predict urinary Cd were blood Cd concentration and duration of oyster farming. A rise in blood Cd was observed after 12 years of farming oysters, likely caused by higher consumption of oysters during this period. Conclusions: Oyster-derived Cd is bioavailable and affects body stores of the metal.

60 APPLIED LIFE SCIENCES↗

Heavy metals exposures among Mexican farmworkers in eastern North Carolina

Background: Immigrant farmworkers are a population at risk for numerous environmental and occupational exposures. The metals arsenic, lead, mercury, and cadmium are known neurotoxins to which workers can be exposed both in the US and in their country of origin. Because farmworkers are exposed to neurotoxic pesticides, they may be at risk for adverse health effects from the combined exposure. Objectives: To examine the relationship between exposure to metals, as measured in urine, with personal and work-related characteristics of Mexican migrant and seasonal farmworkers in the US. Methods: We analyzed data on metals found in urine of 258 farmworkers recruited from 44 camps in eastern North Carolina in 2007. Geometric means and 95% confidence intervals (CI) were used to compare data with data from the National Health and Nutrition Examination Survey (NHANES). We used multivariate regression models fitted for each metal to estimate the association of creatinine-corrected urinary metals and worker characteristics related to environmental and occupational exposures. Results: Geometric mean urinary metals concentrations ({mu}g/g creatinine) exceeded NHANES reference values for arsenic (13.23 [CI 11.11, 15.35] vs. 8.55 [CI 7.23, 9.86]) and lead (1.26 [CI 1.08, 1.43] vs. 0.63 [CI 0.60, 0.66]). Age, being from the central region of Mexico, and pack years of cigarette smoking were significant predictors of metals exposure; being a current smoker and years worked in US agriculture were not. Conclusions: This first study to examine indicators of worker body burdens of metals shows that workers have body burdens related to exposures other than work in the US. Further research should address their risk for adverse health outcomes due to combined exposures to neurotoxins in pesticides.

60 APPLIED LIFE SCIENCES↗

Distribution of toenail selenium levels in young adult Caucasians and African Americans in the United States: The CARDIA Trace Element Study

Background: Data on selenium (Se) levels in American young adults, especially in African Americans, are lacking. Objective: This study presented toenail Se distributions in American young adults of both genders, including both Caucasians and African Americans; and explored potential predictors of toenail Se levels. Data and methods: Data from the Coronary Artery Risk Development in Young Adults study among 4252 American young adults, aged 20-32 in 1987 was used to examine toenail Se levels by instrumental neutron-activation analysis. The distribution of Se levels was described and multivariable linear regression was used to examine potential modifiers of toenail Se concentration within ethnicity-gender subgroups. Results: The geometric mean of toenail Se in this cohort was 0.844 {mu}g/g (95% CI, 0.840-0.849 {mu}g/g) and the median was 0.837 {mu}g/g (95% CI, 0.833-0.844 {mu}g/g). Median levels from lowest to highest quintile were 0.691, 0.774, 0.838, 0.913 and 1.037 {mu}g/g. Se levels varied geographically, and were generally in accordance with its concentrations in local soil. Males, African Americans, current smokers, heavy drinkers and less educated participants were more likely to have low Se levels. Conclusion: This study suggests that toenail Se levels vary geographically depending on soil Se concentrations. In addition to gender, ethnicity and education level, smoking status and alcohol consumption are two important indicators of Se status since they are modifiable lifestyle factors. Findings from this study might aid public health professionals in identifying people at relatively high or low Se levels, so that chronic disease prevention efforts can be directed toward these subgroups. - Research highlights: {yields} Average of toenail Se levels in this cohort was 0.844 {mu}g/g (95% CI, 0.840-0.849 {mu}g/g). {yields} Toenail Se levels vary geographically depending on soil Se concentrations. {yields} Males, African Americans and less educated participants have low Se levels. {yields} Smoking status and alcohol consumption are two important indicators of Se status.

54 ENVIRONMENTAL SCIENCES↗