Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “tool influence function”

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 127 records · Page 7

Modeling the impact of liana infestation on the demography and carbon cycle of tropical forests

There is mounting empirical evidence that lianas affect the carbon cycle of tropical forests. However, no single vegetation model takes into account this growth form, although such efforts could greatly improve the predictions of carbon dynamics in tropical forests. In this study, we incorporated a novel mechanistic representation of lianas in a dynamic global vegetation model (the Ecosystem Demography Model). We developed a liana-specific plant functional type and mechanisms representing liana-tree interactions (such as light competition, liana-specific allometries, and attachment to host trees) and parameterized them according to a comprehensive literature meta-analysis. We tested the model for an old-growth forest (Paracou, French Guiana) and a secondary forest (Gigante Peninsula, Panama). The resulting model simulations captured many features of the two forests characterized by different levels of liana infestation as revealed by a systematic comparison of the model outputs with empirical data, including local census data from forest inventories, eddy flux tower data, and terrestrial laser scanner-derived forest vertical structure. The inclusion of lianas in the simulations reduced the secondary forest net productivity by up to 0.46 t C ha -1 year -1 , which corresponds to a limited relative reduction of 2.6% in comparison with a reference simulation without lianas. However, this resulted in significantly reduced accumulated above-ground biomass after 70 years of regrowth by up to 20 t C /ha (19% of the reference simulation). Ultimately, the simulated negative impact of lianas on the total biomass was almost completely cancelled out when the forest reached an old-growth successional stage. Our findings suggest that lianas negatively influence the forest potential carbon sink strength, especially for young, disturbed, liana-rich sites. In light of the critical role that lianas play in the profound changes currently experienced by tropical forests, this new model provides a robust numerical tool to forecast the impact of lianas on tropical forest carbon sinks.

59 BASIC BIOLOGICAL SCIENCES↗

Component-Level Analysis of Heating and Cooling Loads in the U.S. Residential Building Stock

The residential building sector accounts for a substantial portion of total energy consumption in the United States and offers a significant opportunity for energy reduction and decarbonization through improvements in energy efficiency. Heating and air conditioning are the primary contributors to residential energy usage and electricity system peak demand. However, due to the diversity of the housing stock and the complexity of factors affecting heating and cooling demand, identifying the relative contributions to heating and cooling loads poses challenges. To address this, we applied the ResStock analysis tool to simulate 550,000 building energy models, providing statistical representation of residential buildings in the contiguous United States. We introduced outputs that quantified the heating and cooling influence of different components of a home, such as air leakage, envelope components (ceilings, walls, windows, foundations), internal heat gains from people, lighting, plug loads, and duct losses and gains. Leveraging the granularity of ResStock, we present a dataset to enable deeper understanding of the contributors to heating and cooling loads as a function of housing characteristics such as location, envelope efficiency, and building type. This work aims to support prioritization of research and development and informed decision-making for residential building decarbonization.

building simulation↗

Impact of hollow-atom formation on coherent x-ray scattering at high intensity

X-ray free-electron lasers (FELs) are promising tools for structural determination of macromolecules via coherent x-ray scattering. During ultrashort and ultraintense x-ray pulses with an atomic-scale wavelength, samples are subject to radiation damage and possibly become highly ionized, which may influence the quality of x-ray scattering patterns. We develop a toolkit to treat detailed ionization, relaxation, and scattering dynamics for an atom within a consistent theoretical framework. The coherent x-ray scattering problem including radiation damage is investigated as a function of x-ray FEL parameters such as pulse length, fluence, and photon energy. We find that the x-ray scattering intensity saturates at a fluence of {approx}10{sup 7} photon/A{sup 2} per pulse but can be maximized by using a pulse duration much shorter than the time scales involved in the relaxation of the inner-shell vacancy states created. Under these conditions, both inner-shell electrons in a carbon atom are removed, and the resulting hollow atom gives rise to a scattering pattern with little loss of quality for a spatial resolution >1 A. Our numerical results predict that in order to scatter from a carbon atom 0.1 photon per x-ray pulse, within a spatial resolution of 1.7 A, a fluence of 1x10{sup 7} photons/A{sup 2} per pulse is required at a pulse length of 1 fs and a photon energy of 12 keV. By using a pulse length of a few hundred attoseconds, one can suppress even secondary ionization processes in extended systems. The present results suggest that high-brightness attosecond x-ray FELs would be ideal for single-shot imaging of individual macromolecules.

74 ATOMIC AND MOLECULAR PHYSICS↗

Final report for Collaborative Research on Ecophysiological Controls on Amazonian Precipitation Seasonality and Variability

The principal objective of our project is to address how vegetation influences climate variability and precipitation over Amazonian rainforests, with an emphasis on plant physiological controls on deep convection triggering along a geographical water stress gradient. Using a new tool of sun-induced chlorophyll fluorescence (SIF), our study integrates plant physiological observations and climate modeling and aims to decrease the gap between field ecology and climate modeling. We have installed sensors at Manaus site (K34) and at Rebio Jaru (Ji-Paraná, Rondonia). From field observations, we found a diversity of hydraulic functioning, indicating that within the same forest we will find different responses to the predicted climatic changes. The hydraulic strategies varied through a continuum of hydraulic security and water transport efficiency, which can be related to a long process of taxa evolution. The fact that taller trees are more hydraulically vulnerable indicates that if these forests go through higher drought conditions and temperature increase, they could experience a massive tree mortality that will promote changes on forest structure, changing the ecosystem functions, such as water fluxes and carbon storage. To investigate the role of water stress on the onset of the rainy season, we have developed high-resolution cloud resolving model (CRM) simulations using a new strategy coupling convective and large-scales. The first results show that the presence of early morning fog reduces the wet season transpiration and carbon uptake over the everwet regions of Amazonia. We have incorporated SIF into CLM to constrain energy, carbon and water flux of tropical forests. By using CloudSat-observed cloud vertical structure, we show an empirical relationship between atmospheric energetics and convective clouds in Amazonia, in comparison with the savanna/shrubland and tropical West Pacific, emphasizing predominant deep convection during the early wet season when Bowen ratio and convective available potential energy (CAPE) increase. Although on average the deep convections are prominent only during the short dry-to-wet transition while shallow/congestus clouds dominate the rest of the year in Amazonia, there is a clear cloud regime shift when CAPE grows beyond a threshold value. Our project helped initiate and maintain collaboration between US and Brazilian scientists. We have been collaborating through yearly field campaigns, weekly international research meetings, and ongoing other collaborative projects. Through these activities, we helped our students and postdocs fill their knowledge gaps and learn new research techniques.

54 ENVIRONMENTAL SCIENCES↗

Collaborative Research on Ecophysiological Controls on Amazonian Precipitation Seasonality and Variability (Final Technical Report)

The principal objective of our project is to address how vegetation influences climate variability and precipitation over Amazonian rainforests, with an emphasis on plant physiological controls on deep convection triggering along a geographical water stress gradient. Using a new tool of sun-induced chlorophyll fluorescence (SIF), our study integrates plant physiological observations and climate modeling and aims to decrease the gap between field ecology and climate modeling. We have installed sensors at Manaus site (K34) and at Rebio Jaru (Ji-Paraná, Rondonia). From field observations, we found a diversity of hydraulic functioning, indicating that within the same forest we will find different responses to the predicted climatic changes. The hydraulic strategies varied through a continuum of hydraulic security and water transport efficiency, which can be related to a long process of taxa evolution. The fact that taller trees are more hydraulically vulnerable indicates that if these forests go through higher drought conditions and temperature increase, they could experience a massive tree mortality that will promote changes on forest structure, changing the ecosystem functions, such as water fluxes and carbon storage. To investigate the role of water stress on the onset of the rainy season, we have developed high-resolution cloud resolving model (CRM) simulations using a new strategy coupling convective and large-scales. The first results show that the presence of early morning fog reduces the wet season transpiration and carbon uptake over the everwet regions of Amazonia. We have incorporated SIF into CLM to constrain energy, carbon and water flux of tropical forests. By using CloudSat-observed cloud vertical structure, we show an empirical relationship between atmospheric energetics and convective clouds in Amazonia, in comparison with the savanna/shrubland and tropical West Pacific, emphasizing predominant deep convection during the early wet season when Bowen ratio and convective available potential energy (CAPE) increase. Although on average the deep convections are prominent only during the short dry-to-wet transition while shallow/congestus clouds dominate the rest of the year in Amazonia, there is a clear cloud regime shift when CAPE grows beyond a threshold value. Our project helped initiate and maintain collaboration between US and Brazilian scientists. We have been collaborating through yearly field campaigns, weekly international research meetings, and ongoing other collaborative projects. Through these activities, we helped our students and postdocs fill their knowledge gaps and learn new research techniques.

54 ENVIRONMENTAL SCIENCES↗

Chemical clocks: using otolith geochemistry to enhance estimation of age and growth of white hake (Urophycis tenuis)

The white hake (Urophycis tenuis) is a groundfish distributed throughout the Gulf of Maine. Catch advice is based on stock assessments done with age-based population dynamics models; however, otolith aging is challenging because of unclear growth increments. To address this concern, we compared the consistency of aging with counts of visual annuli to that of aging with cycles of elemental concentrations measured by using laser ablation inductively coupled plasma mass spectrometry. We tested the hypothesis that oscillations in both environmental conditions and internal physiology through time influence uptake of elements during otolith mineralization. Concentrations of manganese, in comparison with those of the other investigated trace elements (magnesium, strontium, and barium), had the most promising correlation with visual growth increments (~100% age agreement, ±1 year), offering an additional tool to enhance increment identification. In our examination of 550 otoliths collected during 2007–2021, we found that white hake lived a maximum of 10.3 years and exhibited sexual dimorphism in maximum length and age. By using generated von Bertalanffy growth functions, L(t)=110(1–e–0.113(t+0.45)) for males and L(t)=140(1–e–0.113(t–0.30)) for females (where L(t) is length at time t), size and age at maturity were calculated for males (37.4 cm in total length (TL), 3.3 years) and females (47.4 cm TL, 4.2 years). These results demonstrate that otolith geochemistry can be used to improve the accuracy and precision of the estimation of fish age and maturity, even for challenging species.

59 BASIC BIOLOGICAL SCIENCES↗

HEPOM: Using Graph Neural Networks for the Accelerated Predictions of Hydrolysis Free Energies in Different pH Conditions

Hydrolysis is a fundamental family of chemical reactions where water facilitates the cleavage of bonds. The process is ubiquitous in biological and chemical systems, owing to water’s remarkable versatility as a solvent. However, accurately predicting the feasibility of hydrolysis through computational techniques is a difficult task, as subtle changes in reactant structure like heteroatom substitutions or neighboring functional groups can influence the reaction outcome. Furthermore, hydrolysis is sensitive to the pH of the aqueous medium, and the same reaction can have different reaction properties at different pH conditions. In this work, we have combined reaction templates and high-throughput ab initio calculations to construct a diverse data set of hydrolysis free energies. The developed framework automatically identifies reaction centers, generates hydrolysis products, and utilizes a trained graph neural network (GNN) model to predict ΔG values for all potential hydrolysis reactions in a given molecule. The long-term goal of the work is to develop a data-driven, computational tool for high-throughput screening of pH-specific hydrolytic stability and the rapid prediction of reaction products, which can then be applied in a wide array of applications including chemical recycling of polymers and ion-conducting membranes for clean energy generation and storage.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Methods to Register Models and Input/Output Parameters for Integrated Modeling

Significant resources can be required when constructing integrated modeling systems. In a typical application, components (e.g., models and databases) created by different developers are assimilated, requiring the framework’s functionality to bridge the gap between the user’s knowledge of the components being linked. The framework, therefore, needs the capability to assimilate a wide range of model-specific input/output requirements as well as their associated assumptions and constraints. The process of assimilating such disparate components into an integrated modeling framework varies in complexity and difficulty. Several factors influence the relative ease of assimilating components, including, but not limited to, familiarity with the components being assimilated, familiarity with the framework and its tools that support the assimilation process, level of documentation associated with the components and the framework, and design structure of the components and framework. This initial effort reviews different approaches for assimilating models and their model-specific input/output requirements: 1) modifying component models to directly communicate with the framework (i.e., through an Application Programming Interface), 2) developing model-specific external wrappers such that no component model modifications are required, 3) using parsing tools to visually map pre-existing input/output files, and 4) describing and linking models as dynamic link libraries. Most of these approaches are illustrated using the widely distributed modeling system called Framework for Risk Analysis in Multimedia Environmental Systems (FRAMES). The review concludes that each has its strengths and weakness, the factors that determine which approaches work best in a given application.

Data Transfer↗

Exploring the Functional Performance of a Commercial High-Temperature Photopolymer Resin for Vat Photopolymerized Injection Molding Tools

Additive manufacturing (AM) presents a promising approach for rapid tooling, offering significant reductions in production costs and lead times compared to that of conventional moldmaking processes. This work examines the fabrication of molds for injection molding through vat photopolymerization (VPP) AM, using a commercially available photopolymer resin intended for high temperature applications (High Temp V2 resin). The photorheological behavior of the resin is examined to evaluate its curing kinetics and mechanical development during polymerization, ensuring suitability for mold fabrication. The thermal and mechanical properties of the printed photopolymer resin are characterized, alongside assessments of dimensional accuracy and surface finish, which are critical factors influencing the performance and quality of injection molded parts. Additionally, specimens produced using VPP-fabricated molds are evaluated and benchmarked against those molded with traditional metal molds.

Jones, Haley W. [Savannah River National Laborator↗

Global overview and major challenges of host prediction methods for uncultivated phages

Bacterial communities play critical roles across all of Earth’s biomes, affecting human health and global ecosystem functioning. They do so under strong constraints exerted by viruses, i.e., bacteriophages or “phages”. Phages can reshape bacterial communities’ structure, influence long-term evolution of bacterial populations, and alter host cell metabolism during infection. Metagenomics approaches, i.e., shotgun sequencing of environmental DNA or RNA, recently enabled large-scale exploration of phage genomic diversity, yielding several millions of phage genomes now to be further analyzed and characterized. One major challenge however is the lack of direct host information for these phages. Several methods and tools have been proposed to bioinformatically predict the potential host(s) of uncultivated phages based only on genome sequence information. Here we review these different approaches and highlight their distinct strengths and limitations. We also outline complementary experimental assays which are being proposed to validate and refine these bioinformatic predictions.

59 BASIC BIOLOGICAL SCIENCES↗

cMonkey2: Automated, systematic, integrated detection of co-regulated gene modules for any organism

The cMonkey integrated biclustering algorithm identifies conditionally co-regulated modules of genes (biclusters). cMonkey integrates various orthogonal pieces of information which support evidence of gene co-regulation, and optimizes biclusters to be supported simultaneously by one or more of these prior constraints. The algorithm served as the cornerstone for constructing the first global, predictive Environmental Gene Regulatory Influence Network (EGRIN) model for a free-living cell, and has now been applied to many more organisms. However, due to its computational inefficiencies, long run-time and complexity of various input data types, cMonkey was not readily usable by the wider community. To address these primary concerns, we have significantly updated the cMonkey algorithm and refactored its implementation, improving its usability and extendibility. These improvements provide a fully functioning and user-friendly platform for building co-regulated gene modules and the tools necessary for their exploration and interpretation. We show, via three separate analyses of data for E. coli, M. tuberculosis and H. sapiens, that the updated algorithm and inclusion of novel scoring functions for new data types (e.g. ChIP-seq and transcription factor over-expression [TFOE]) improve discovery of biologically informative co-regulated modules. The complete cMonkey2 software package, including source code, is available at https://github.com/baliga-lab/cmonkey2.

59 BASIC BIOLOGICAL SCIENCES↗

Radiological Safety Analysis Computer (RSAC 7.2) Presentation

This presentation will be for the graduate seminar class for the University of New Mexico. We will discuss the Radiological Safety Analysis Computer (RSAC) version 7.2, a critical tool in radiological consequence analysis within the nuclear safety domain. We will begin with an introduction to radiological consequence analysis and the importance of dose modeling in nuclear safety, referencing key regulatory frameworks such as DOE-STD-3009 and 10 CFR 20. The presentation will provide an overview of RSAC, including its history, primary functions, and the inputs and outputs involved in its calculations. A detailed examination of RSAC's atmospheric dispersion model, based on the Gaussian plume methodology, will be presented, highlighting how meteorological factors influence the results. We will discuss the regulatory alignment of RSAC with DOE and NRC guidelines, emphasizing the assumptions required for compliance and the balance between conservative and realistic inputs. The presentation will also address the limitations and assumptions inherent in RSAC, comparing it with other consequence analysis codes like MACCS, HotSpot, and RASCAL. An example RSAC run will be demonstrated, showcasing a hypothetical accident scenario and interpreting the resulting dose calculations. Finally, we will introduce new ideas to enhance the current RSAC framework. This future work aims to provide multi-industry adaptability and integration with new dispersion models, addressing the evolving needs of DOE and NRC. The presentation will conclude with a discussion on RSAC's ongoing role in nuclear safety and the opportunities for future innovation.

computer program↗

PATtyFams: Protein families for the microbial genomes in the PATRIC database

The ability to build accurate protein families is a fundamental operation in bioinformatics that influences comparative analyses, genome annotation, and metabolic modeling. For several years we have been maintaining protein families for all microbial genomes in the PATRIC database (Pathosystems Resource Integration Center, patricbrc.org) in order to drive many of the comparative analysis tools that are available through the PATRIC website. However, due to the burgeoning number of genomes, traditional approaches for generating protein families are becoming prohibitive. In this report, we describe a new approach for generating protein families, which we call PATtyFams. This method uses the k-mer-based function assignments available through RAST (Rapid Annotation using Subsystem Technology) to rapidly guide family formation, and then differentiates the function-based groups into families using a Markov Cluster algorithm (MCL). In conclusion, this new approach for generating protein families is rapid, scalable and has properties that are consistent with alignment-based methods.

59 BASIC BIOLOGICAL SCIENCES↗

Effect of Annealing Temperature on Structure and Magnetic Properties of Zn{sub 0.94}Mg{sub 0.01}Mn{sub 0.05}O Nanoparticles

Structural and magnetic characteristics of Zn{sub 0.94}Mg{sub 0.01}Mn{sub 0.05}O (ZnMgMnO) nanoparticles synthesized by Sol-Gel preparation technique in a wide temperature range were reported. The influence of annealing temperature on the crystal structure of ZnMgMnO nanoparticles was figured out by X-ray diffraction technique (XRD). To optimize the annealing temperature and reveal the particle formation and size, Scanning Electron Microscope (SEM) imaging technique was performed. The required stoichiometry of the synthesized samples was obtained by an energy dispersive X-ray analysis technique (EDX). Quantum Design Vibrating Sample Magnetometer (QDVSM) tool was used for the magnetic characterizations of Zn{sub 0.94}Mg{sub 0.01}Mn{sub 0.05}O composition. Magnetization measurements as a function of magnetic field (M-H) were performed in the magnetic field up to 10 kOe. Temperature dependence of magnetization measurements (M-T) was taken between 10 and 320 K temperature ranges.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Carbon use efficiency diagnostics in Nannochloropsis salina

We report that carbon dioxide (CO 2 ) can be a significant resource input affecting the cost of algae biomass production on an industrial scale. Improvements in biofuel productivity therefore require characterization of CO 2 use efficiency (CUE). RuBisCO saturation with CO 2 is an important factor influencing biomass productivity. During CO2 fixation by RuBisCO, fractionation of carbon isotopes occurs, with preferential fixation of 12 CO 2 , resulting in assimilation of the lighter isotope in algae. This photosynthetic discrimination (Δ DIC-algae ), approximated by the difference between the δ 13 C of external medium and that of algae, is a function of the proportion of CO 2 fixed relative to supply. Δ DIC-algae has been applied to the study of photosynthesis in algae over the past few decades and we have adopted the tool to characterize improvements in conditions optimized for biofuel productivity, such as controlled photobioreactors. We report the use of Δ DIC-algae as a tool for characterizing CO 2 dynamics and RuBisCO saturation in Nannochloropsis salina CCMP 1776, a benchmark strain in algal biofuels research. We use Δ DIC-algae to describe the conditions under which RuBisCO saturation can be achieved to maximize biomass productivity. Finally, our results suggest that determination of Δ DIC-algae can provide important feedback to support engineering and cultural improvements that can impact carbon use efficiency and biomass productivity.

09 BIOMASS FUELS↗

NUMERICAL SIMULATIONS OF CHROMOSPHERIC HARD X-RAY SOURCE SIZES IN SOLAR FLARES

X-ray observations are a powerful diagnostic tool for transport, acceleration, and heating of electrons in solar flares. Height and size measurements of X-ray footpoint sources can be used to determine the chromospheric density and constrain the parameters of magnetic field convergence and electron pitch-angle evolution. We investigate the influence of the chromospheric density, magnetic mirroring, and collisional pitch-angle scattering on the size of X-ray sources. The time-independent Fokker-Planck equation for electron transport is solved numerically and analytically to find the electron distribution as a function of height above the photosphere. From this distribution, the expected X-ray flux as a function of height, its peak height, and full width at half-maximum are calculated and compared with RHESSI observations. A purely instrumental explanation for the observed source size was ruled out by using simulated RHESSI images. We find that magnetic mirroring and collisional pitch-angle scattering tend to change the electron flux such that electrons are stopped higher in the atmosphere compared with the simple case with collisional energy loss only. However, the resulting X-ray flux is dominated by the density structure in the chromosphere and only marginal increases in source width are found. Very high loop densities (>10{sup 11} cm{sup -3}) could explain the observed sizes at higher energies, but are unrealistic and would result in no footpoint emission below about 40 keV, contrary to observations. We conclude that within a monolithic density model the vertical sizes are given mostly by the density scale height and are predicted smaller than the RHESSI results show.

79 ASTRONOMY AND ASTROPHYSICS↗

Hybrid interatomic potential for Sn

To design materials for extreme applications, it is important to understand and predict phase transitions and their influence on material properties under high pressures and temperatures. Atomistic modeling can be a useful tool to assess these behaviors. However, this can be difficult due to the lack of fidelity of the interatomic potentials in reproducing this high pressure and temperature extreme behavior. Here, in this work, a hybrid EAM-R—which is the combination of embedded atom method (EAM) and rapid artificial neural network potential—for Tin (Sn) is described which is capable of accurately modeling the complex sequence of phase transitions between different metallic polymorphs as a function of pressure. This hybrid approach ensures that a basic empirical potential like EAM is used as a lower energy bound. By using the final activation function, the neural network contribution to energy must be positive, assuring stability over the whole configuration space. This implementation has the capacity to reproduce density functional theory results at 6 orders of magnitude slower than a pair potential for molecular dynamics simulation, including elastic and plastic characteristics and relative energies of each phase. Using calculations of the Gibbs free energy, it is demonstrated that the potential precisely predicts the experimentally observed phase changes at temperatures and pressures across the whole phase diagram. At 10.2 GPa, the present potential predicts a first-order phase transition between body-centered tetragonal (BCT) β-Sn and another polymorph of BCT-Sn. This structure transforms into body-centered cubic near the experimentally reported value at 33 GPa. Thus, the Sn potential developed in this paper can be used to study complex deformation mechanisms under extreme conditions of high pressure and strain rates unlike existing potentials. Moreover, the framework developed in this paper can be extended for different material systems with complex phase diagrams.

36 MATERIALS SCIENCE↗

Investigation of hydrometeorological influences on reservoir releases using explainable machine learning methods

Long short-term memory (LSTM) networks have demonstrated successful applications in accurately and efficiently predicting reservoir releases from hydrometeorological drivers including reservoir storage, inflow, precipitation, and temperature. However, due to its black-box nature and lack of process-based implementation, we are unsure whether LSTM makes good predictions for the right reasons. In this work, we use an explainable machine learning (ML) method, called SHapley Additive exPlanations (SHAP), to evaluate the variable importance and variable-wise temporal importance in the LSTM model prediction. In application to 30 reservoirs over the Upper Colorado River Basin, United States, we show that LSTM can accurately predict the reservoir releases with NSE ≥ 0.69 for all the considered reservoirs despite of their diverse storage sizes, functionality, elevations, etc. Additionally, SHAP indicates that storage and inflow are more influential than precipitation and temperature. Moreover, the storage and inflow show a relatively long-term influence on the release up to 7 days and this influence decreases as the lag time increases for most reservoirs. These findings from SHAP are consistent with our physical understanding. However, in a few reservoirs, SHAP gives some temporal importances that are difficult to interpret from a hydrological point of view, probably because of its ignorance of the variable interactions. SHAP is a useful tool for black-box ML model explanations, but the hydrological processes inferred from its results should be interpreted cautiously. More investigations of SHAP and its applications in hydrological modeling is needed and will be pursued in our future study.

54 ENVIRONMENTAL SCIENCES↗