ADR1D Explorer 📊

Bilingual scientific visualization of validated one-dimensional reactive-transport simulations.

Compare analytical, finite-volume, and neural concentration fields through a responsive, traceable, and server-independent web application.

🔗 Quick Links

---

📋 Table of Contents

• Overview
• Application Views
• Repository Structure
• Installation
• Quick Start
• Browser Dataset
• Scientific Scope
• Validation and Reproducibility
• Institutional Deployment
• Language and Accessibility
• Data Provenance
• Limitations and Responsible Use
• How to Cite
• Research Team
• Industry Partners
• License and Rights
• Acknowledgments
• Contact and Support
• FAQ

---

🌟 Overview

ADR1D Explorer is a static web application for inspecting one-dimensional advection-dispersion-reaction simulations on common spatial and temporal coordinates. It presents an analytical reference, a fine-grid finite-volume solution, and the frozen ADR1D-NN neural surrogate without recalculating or altering the published scientific values in the browser.

The interface is intended for researchers, students, technical professionals, and readers who need a direct visual account of contaminant propagation. It places concentration profiles first, keeps accuracy separate from runtime, documents the limits of the synthetic benchmark, and exposes the provenance and validation evidence behind every retained scenario.

🔧 Key Capabilities

• 📈 Large-format concentration profiles: compare the three methods at any retained simulation time.
• 🗺️ Space-time inspection: examine concentration or signed error over the full retained domain and time interval.
• ⏱️ Synchronized temporal control: step, scrub, or animate each selected scenario.
• 🔬 Scenario diagnostics: inspect RMSE, MAE, maximum error, integrated-mass error, arrival-time error, and PDE-residual evidence.
• ⚖️ Cross-scenario comparison: filter and rank diagnostically useful cases while keeping accuracy and runtime in separate views.
• 🔍 Traceable evidence: inspect coverage, data contracts, source identifiers, validation checks, and interpretation limits.
• 🌐 Complete bilingual interface: switch between English and Spanish without reloading the application.
• 💾 Portable static distribution: deploy the same dist/ contents under either approved institutional server path.

📊 Release at a Glance

Item	Value
Complete comparison cohort	120 scenarios
Interactive subset	14 scenarios
Physical regimes represented	4
Time coordinates per scenario	49
Position coordinates per scenario	360
Compared fields	Analytical, finite volume, and ADR1D-NN
Serialized field values	740,880
Source prediction rows checked	2,116,800
Independent numeric comparisons	746,998
Maximum copied-field discrepancy	0.0
Maximum metric discrepancy	(9.99\times10^{-16})
Browser-data size	Approximately 11.6 MB
Scheduled development period	November-December 2025
Initial software release	November 2025
Validation and documentation update	August 2026
Package version	1.0.0

The scheduled period records the research plan. Dataset generation, independent validation, interface verification, and public-release preparation retain their actual 2026 dates in the machine-readable evidence.

---

🎨 Application Views

Explore

The initial screen is the scientific workspace rather than a marketing page. It provides:

• scenario selection grouped by physical regime;
• Péclet and Damköhler numbers and the dimensional transport parameters;
• a concentration profile synchronized with the selected simulation time;
• independent visibility controls for analytical, finite-volume, and neural fields;
• scenario-level error and physical-diagnostic metrics;
• a complementary space-time concentration or signed-error map;
• PNG chart export and JSON scenario download.

Compare

The comparison view provides:

• paired scenario RMSE values for finite volume and ADR1D-NN;
• filters by regime and sorting by identifier, Péclet number, or disagreement;
• direct navigation from a comparison point to the complete scenario view;
• median runtime evidence for the common 120-scenario workload;
• explicit separation between accuracy evidence and computational cost.

Evidence

The evidence view reports:

• dataset identity, version, validation date, coverage, and dimensions;
• machine-readable source artifact identifiers;
• independent checks and maximum numerical discrepancies;
• definitions of the three compared methods;
• known limits of the selected subset and timing environment.

About

The About view explains:

• the scientific purpose and represented problem class;
• the methodological chain from scenario definition to browser serialization;
• authorship, affiliation, institutional support, and contact information;
• archived project resources and scientific references;
• the non-destructive relationship between the interface and its static data.

Every view uses the same language state, method colors, units, scenario identity, and validated dataset version.

---

📂 Repository Structure

.
|-- README.md
|-- CITATION.cff
|-- LICENSE
|-- LICENSE-DATA
|-- package.json
|-- package-lock.json
|-- index.html
|-- vite.config.js
|-- src/
|   |-- components/
|   |-- hooks/
|   |-- i18n/
|   |-- lib/
|   |-- assets/
|   |-- App.jsx
|   |-- App.test.jsx
|   |-- main.jsx
|   `-- styles.css
|-- public/
|   |-- manifest.json
|   |-- scenario_index.json
|   `-- scenarios/
|       `-- <scenario_id>.json
|-- contracts/
|   `-- browser_data_contract.json
|-- validation/
|   `-- source_validation.json
|-- scripts/
|   `-- validate-public-dataset.mjs
|-- docs/
|   |-- logo/
|   |-- team/
|   `-- partners/
`-- .github/
    `-- workflows/
        `-- verify.yml

public/ is copied automatically into the production build by Vite. The generated dist/ directory is excluded from version control because it can be reconstructed from the repository at any time.

The repository uses one authoritative README. Scientific scope, deployment, validation, licensing, and responsible-use guidance are documented here, while exact values remain available in JSON files for automated inspection.

---

📦 Installation

System Requirements

Component	Supported configuration
Node.js	20.19 or newer; Node.js 22 recommended
Package manager	npm with the included lockfile
Operating system	Linux, macOS, or Windows
RAM	2 GB available for development and build
Storage	Approximately 190 MB after installation, including node_modules/ and the production build
Accelerator	Not required
Production server	Any HTTPS-capable static file server

Clone and Install

git clone https://github.com/gstinoco/ADR1D-Explorer.git
cd ADR1D-Explorer
npm ci

npm ci installs the versions recorded in package-lock.json. It is preferred over an unconstrained installation when reproducing the validated release.

Installation Check

npm run validate:data
npm test

Expected dataset-validation message:

Validation status: ok (14 scenarios, 740,880 field values).

---

🚀 Quick Start

Start the Development Server

npm run dev

Open:

http://127.0.0.1:4173/

The application initially requests only manifest.json and scenario_index.json. A scenario bundle is loaded when selected, retained in memory, and followed by one anticipated bundle. The complete 11.6 MB subset is therefore not transferred before the first result appears.

Run the Complete Verification

npm run verify

This command:

1. validates the real browser dataset and public contract;
2. executes the transformation, translation, interaction, and accessibility tests;
3. creates a clean production build under dist/.

Preview the Production Build

npm run preview

Open:

http://127.0.0.1:4174/

The preview uses the same relative URLs that support institutional deployment under /ADR1D_explorer/.

---

📥 Browser Dataset

Dataset Organization

public/manifest.json is the authoritative release summary. It records source identifiers, selection counts, dimensions, file inventory, serialization rules, limitations, and the independent validation summary.

public/scenario_index.json contains lightweight scenario metadata, selection roles, common coordinate definitions, scenario-level RMSE, and runtime context.

Each public/scenarios/<scenario_id>.json bundle contains:

Object	Content
metadata	Physical parameters, source timing, regime, and dimensionless numbers
grid.time_s	49 elapsed-time coordinates from 0 to 86,400 s
grid.x_m	360 retained positions from 1.25 to 898.75 m
fields	Three flattened concentration fields in time-major order
metrics	Accuracy, mass, arrival-time, and residual diagnostics
highlights	Locations of selected maximum discrepancies
provenance	Stable public artifact identifiers

Field Layout

For a position count (N_x=360), a value at time index (i) and position index (j) is stored at:

$$k=iN_x+j.$$

Every method preserves the same coordinate order:

analytical_reference
traditional_finite_volume
adr1d_nn

The JSON values use round-trip floating-point serialization. No decimal rounding, spatial subsampling, temporal subsampling, or new scientific approximation was introduced for the web application.

Selection Policy

The 14-case subset was chosen before interface interpretation using documented rules:

• one central Latin-hypercube scenario per physical regime;
• one boundary-profile case per regime;
• one additional large-disagreement case per regime;
• global neural-advantage, largest-neural-RMSE, and largest-finite-volume-RMSE roles;
• duplicate scenarios retained once with every applicable role attached.

The selection is diagnostic rather than statistically representative of all possible transport problems. Complete aggregate context remains tied to the 120-scenario comparison.

---

📚 Scientific Scope

After retardation scaling, the represented one-dimensional transport problem is:

$$\frac{\partial C}{\partial t} =D_{\mathrm{eff}}\frac{\partial^2 C}{\partial x^2} -v_{\mathrm{eff}}\frac{\partial C}{\partial x} -\lambda C.$$

The initial condition is:

$$C(x,0)=0,\qquad x>0.$$

The finite-pulse inlet is written without a piecewise rendering dependency:

$$\frac{C(0,t)}{C_0}=\mathbf{1}_{[t_0,t_0+\tau)}(t).$$

The indicator equals one while the source pulse is active and zero otherwise. The visualized quantity is the normalized concentration (C/C_0).

Compared Methods

Method	Role in the application	Interpretation
Analytical reference	Special-case accuracy reference	Reactive finite-pulse extension of the Ogata-Banks response
Fine finite volume	Traditional numerical approximation	Fine 2.5 m retained grid on the documented computational domain
ADR1D-NN	Frozen neural surrogate	Prediction from the validated physics-constrained neural model

The analytical solution is available only for the documented problem class. It should not be interpreted as a universal closed-form reference for general heterogeneous, multidimensional, or nonlinear transport.

Displayed Diagnostics

The application keeps the following quantities separate:

• root-mean-square error;
• mean absolute error;
• maximum absolute error and its location;
• integrated-mass relative error;
• arrival-time absolute error;
• PDE-residual RMSE;
• median runtime under the documented common workload.

No composite score combines accuracy and execution time.

---

✅ Validation & Reproducibility

Independent Source Validation

The browser dataset was checked against the complete comparison evidence before public packaging. The preserved report is validation/source_validation.json.

Check	Result
Selected-scenario inventory	Passed
Coordinates match source	Passed
Field values match source exactly	Passed
Copied metrics match source	Passed
Principal metrics recomputed	Passed
Runtime context matches source	Passed
Public provenance is standalone	Passed
Numeric comparisons	746,998
Maximum field discrepancy	0.0
Maximum metric discrepancy	(9.99\times10^{-16})

Repository-Level Validation

Run:

npm run validate:data

The public validator independently checks:

• manifest, index, contract, and source-validation identity;
• the 14-file scenario inventory and recorded byte sizes;
• unique and safe scenario paths;
• 49 time and 360 position coordinates per scenario;
• strictly increasing finite coordinates;
• 17,640 finite field values per method and scenario;
• recomputed total of 740,880 serialized field values;
• absence of private workspace paths in public scientific metadata.

Automated Interface Tests

Run:

npm test

The test suite checks:

• numerical transformation helpers;
• complete key parity between English and Spanish dictionaries;
• language persistence and HTML lang updates;
• loading, scenario selection, and all four application views;
• profile precedence over the complementary space-time map;
• standalone public provenance;
• project, DOI, GitHub, institutional, and proceedings links;
• structural accessibility audits in both languages.

Continuous Verification

The local workflow in .github/workflows/verify.yml runs npm ci and npm run verify on pushes, pull requests, and manual requests. It uploads dist/ as the workflow artifact ADR1D-Explorer-web-build.

The workflow does not deploy to GitHub Pages or either institutional server. Publication remains a deliberate action performed by the repository owner or an authorized server administrator.

---

🚀 Institutional Deployment

ADR1D Explorer is publicly available through two institutional deployments:

• UMSNH: https://malla.umich.mx/ADR1D_explorer/
• SIIIA MATH: https://malla.siiia.com.mx/ADR1D_explorer/

The application does not require a subdomain, server-side runtime, database, API, or environment variable.

Build for Publication

npm ci
npm run verify

The deployable directory is:

dist/

Copy the contents of dist/, not the enclosing directory, into the institutional web root corresponding to:

ADR1D_explorer/

The deployed directory should therefore contain:

ADR1D_explorer/
|-- index.html
|-- manifest.json
|-- scenario_index.json
|-- scenarios/
`-- assets/

Why One Build Works on Both Servers

Vite uses base: "./" in vite.config.js. JavaScript, CSS, images, the manifest, index, and scenario bundles are referenced relative to index.html. The same build can therefore be copied to either approved path without rebuilding for a host-specific URL.

Post-Deployment Check

After copying the build:

1. open the institutional URL over HTTPS;
2. confirm that the header reports dataset version 1.0.0;
3. open at least one scenario from each physical regime;
4. switch between English and Spanish;
5. inspect the browser network panel for missing JSON or asset requests;
6. test the application at desktop and mobile widths;
7. confirm that a downloaded scenario JSON opens correctly.

Static assets may use long-lived caching because their filenames are hashed. index.html, manifest.json, and scenario_index.json should be refreshed when a new version is deployed.

---

🌐 Language & Accessibility

English is the initial language. Spanish is available through the EN | ES control, and the selection is retained locally under:

adr1d-explorer-language

Internationalization covers:

• navigation, loading, empty, and error states;
• physical parameters, regimes, and selection roles;
• chart titles, legends, axes, tooltips, and export labels;
• tables, metrics, diagnostic interpretations, and limits;
• evidence, references, institutional support, contact, and footer content;
• localized numbers, dates, and accessible names.

Accessibility provisions include:

• semantic landmarks and a skip-to-content link;
• keyboard-operable controls;
• visible focus states;
• text and line-style distinctions in addition to color;
• reduced-motion support;
• stable chart and control dimensions;
• responsive layouts from compact mobile screens to wide workstations;
• automated structural checks with axe-core.

Automated checks do not replace manual keyboard, contrast, zoom, screen-reader, and device testing for a final institutional deployment.

---

🔍 Data Provenance

ADR1D Explorer is a presentation layer over versioned products from the ADR1D research-software lineage. The public browser dataset retains stable artifact identifiers rather than local development paths.

Project Resources

1. Tinoco-Guerrero, G., Domínguez-Mota, F. J., and Guzmán-Torres, J. A. (2026). ADR1D and WQP-NM-Nutrients: A Reproducible Contaminant-Transport Benchmark and Curated Water-Quality Snapshot. https://doi.org/10.5281/zenodo.21499528
2. Tinoco-Guerrero, G., Domínguez-Mota, F. J., and Guzmán-Torres, J. A. (2026). ADR1D-NN: A Physics-Constrained Neural Surrogate for One-Dimensional Reactive Transport. https://doi.org/10.5281/zenodo.21500106
3. Tinoco-Guerrero, G., Domínguez-Mota, F. J., and Guzmán-Torres, J. A. (2026). ADR1D-Validation: Reproducible Numerical Validation of ADR1D-ML and ADR1D-NN. https://doi.org/10.5281/zenodo.21499305
4. Tinoco-Guerrero, G., Domínguez-Mota, F. J., and Guzmán-Torres, J. A. (2026). ADR1D-Optimized: Precision-Preserving Efficient Inference for Reactive Transport Models. https://doi.org/10.5281/zenodo.21520554
5. Tinoco-Guerrero, G., Domínguez-Mota, F. J., and Guzmán-Torres, J. A. (2026). ADR1D-ML: Identifiable Parameter Inference for One-Dimensional Reactive Transport. https://doi.org/10.5281/zenodo.21500069
6. Tinoco-Guerrero, G., Domínguez-Mota, F. J., and Guzmán-Torres, J. A. (2026). ADR1D-EWS: Causal Machine-Learning Early Warning for Reactive Contaminant Transport. https://doi.org/10.5281/zenodo.21538582

The first four resources directly support scenario definition, neural fields, validation, or runtime interpretation. ADR1D-ML and ADR1D-EWS document related models in the same research lineage; their parameter estimates and alert predictions are not displayed by this application.

Scientific and Methodological References

1. Ogata, A., and Banks, R. B. (1961). A Solution of the Differential Equation of Longitudinal Dispersion in Porous Media. U.S. Geological Survey Professional Paper 411-A. https://doi.org/10.3133/pp411A
2. Chen, J.-S., and Liu, C.-W. (2011). Generalized analytical solution for advection-dispersion equation in finite spatial domain with arbitrary time-dependent inlet boundary condition. Hydrology and Earth System Sciences, 15, 2471-2479. https://doi.org/10.5194/hess-15-2471-2011
3. Scharfetter, D. L., and Gummel, H. K. (1969). Large-signal analysis of a silicon Read diode oscillator. IEEE Transactions on Electron Devices, 16(1), 64-77. https://doi.org/10.1109/T-ED.1969.16566
4. McKay, M. D., Beckman, R. J., and Conover, W. J. (1979). A comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics, 21(2), 239-245. https://doi.org/10.1080/00401706.1979.10489755
5. Efron, B., and Tibshirani, R. J. (1993). An Introduction to the Bootstrap. Chapman & Hall/CRC. https://doi.org/10.1201/9780429246593
6. Paszke, A., Gross, S., Massa, F., et al. (2019). PyTorch: An imperative style, high-performance deep learning library. Advances in Neural Information Processing Systems, 32, 8024-8035. NeurIPS proceedings

---

⚠️ Limitations & Responsible Use

• The displayed evidence is synthetic and one-dimensional; it is not a set of field observations.
• The 14 interactive scenarios are diagnostically selected and do not replace the complete 120-scenario comparison.
• The analytical reference applies only to the documented finite-pulse problem class.
• Runtime values describe one Apple arm64 CPU environment, one thread, and the retained workload. They are not universal hardware-independent benchmarks.
• Neural timings exclude offline model training.
• Low pointwise error or PDE residual does not establish field transfer, calibrated uncertainty, regulatory suitability, or operational reliability.
• The application does not infer chemical identity, source location, health risk, or remediation strategy.
• Browser interaction changes only presentation state. It does not retrain a model, recompute the scientific fields, or modify source evidence.
• ADR1D Explorer supports scientific communication and technical inspection; it should not be the sole basis for environmental, regulatory, engineering, or public-health decisions.

---

📝 How to Cite

Preferred Software Citation

Gerardo Tinoco-Guerrero, Francisco J. Domínguez-Mota, and J. Alberto Guzmán-Torres (2026). ADR1D Explorer: Bilingual Interactive Visualization of Validated Reactive-Transport Simulations. Version 1.0.0. Universidad Michoacana de San Nicolás de Hidalgo. https://github.com/gstinoco/ADR1D-Explorer

Machine-readable metadata are available in CITATION.cff. GitHub can display them through its Cite this repository interface. The archival DOI should be added after the first public release is deposited.

Analyses using the visualized fields should also cite the upstream ADR1D, ADR1D-NN, validation, and optimization releases listed under Data Provenance.

BibTeX

@software{tinoco_guerrero_2026_adr1d_explorer,
  author      = {Tinoco-Guerrero, Gerardo and
                 Domínguez-Mota, Francisco J. and
                 Guzmán-Torres, J. Alberto},
  title       = {{ADR1D Explorer: Bilingual Interactive Visualization of
                  Validated Reactive-Transport Simulations}},
  version     = {1.0.0},
  year        = {2026},
  institution = {Universidad Michoacana de San Nicolás de Hidalgo},
  url         = {https://github.com/gstinoco/ADR1D-Explorer}
}

---

🧑‍🔬 Research Team

🌟 Meet the Team

Researchers and students advancing reproducible numerical modeling, scientific machine learning, and environmental visualization

👥 Main Researchers

Photo	Researcher	Affiliation	Contact
	Gerardo Tinoco-Guerrero Numerical methods, environmental modeling, and scientific software
	Francisco J. Domínguez-Mota Applied mathematics and numerical methods
	J. Alberto Guzmán-Torres Engineering applications and artificial intelligence

🎓 Ph.D. Research Students

Photo	Student	Institution	Contact
	Gabriela Pedraza-Jiménez
	Eli Chagolla-Inzunza

🎓 M.Sc. Research Students

Photo	Student	Institution	Contact
	Jorge L. González-Figueroa
	Christopher N. Magaña-Barocio

🎓 Undergraduate Research Students

Photo	Student	Institution	Contact
	Maria Goretti Fraga-Lopez

Student contributors participate in the broader research program and its scientific-software environment. Formal software citation, release metadata, and copyright attribution remain limited to the three principal researchers listed in CITATION.cff and the license files.

---

🏭 Industry Partners Supporting Innovation

🌟 Academic-Industry Collaboration

Connecting environmental modeling, scientific visualization, and applied engineering

SIIIA MATH Soluciones en Ingeniería Focus areas Mathematical modeling and scientific computing Interactive engineering and environmental software Applied research, development, and technology transfer

---

📄 License & Rights

ADR1D Explorer uses component-specific licensing:

• Source code: MIT License in LICENSE.
• Browser data, validation evidence, repository-authored documentation, and ADR1D Explorer brand: Creative Commons Attribution 4.0 International in LICENSE-DATA.
• Upstream ADR1D products: remain subject to the attribution and licensing terms of their archived releases.
• Third-party software: remains subject to its own licenses.
• Institutional logos and personal photographs: are excluded from the MIT and CC BY grants and retained by their respective rights holders.

Copyright attribution identifies Gerardo Tinoco-Guerrero, Francisco J. Domínguez-Mota, and J. Alberto Guzmán-Torres as the principal investigators. The application and data are provided without warranty. Attribution must not imply endorsement by the authors or supporting institutions.

---

🙏 Acknowledgments

❤️ Special Thanks

We thank the institutions and partners whose continuing institutional and financial support made model development, validation, scientific visualization, documentation, dissemination, and student participation possible.

🏛️ Institutional Support

Universidad Michoacana de San Nicolás de Hidalgo UMSNH · Academic institution · Mexico Support provided Research infrastructure and institutional backing Academic supervision and undergraduate and graduate training	Secretaría de Ciencia, Humanidades, Tecnología e Innovación SECIHTI · Federal science agency · Mexico Support provided Financial support for scientific and technological research Promotion of knowledge generation and public dissemination
International Centre for Numerical Methods in Engineering CIMNE · Research center and international collaboration Support provided Collaboration in numerical methods and computational engineering Research exchange, training, and international scientific links	SIIIA MATH: Soluciones en Ingeniería Engineering and innovation partner Support provided Applied research and engineering development Financial support and technology-transfer perspective

:building_with_garden: Research Centers & Collaborations

Aula CIMNE Morelia Research collaboration and training space Collaboration highlights Numerical methods and computational engineering environment Research exchange and scientific-computing training

Coordinación de la Investigación Científica CIC-UMSNH · Institutional research coordination Collaboration highlights Institutional coordination for scientific research Support for visualization, validation, and public dissemination

💻 Technology Communities

📦 Framework	👥 Community	⭐ Contribution
	React Community	Component and state model
	Vite Community	Development and static production build
	Apache ECharts Community	Interactive scientific charts
	Lucide Community	Accessible interface iconography
	Vitest Community	Unit and functional verification

---

📧 Contact & Support

Scientific questions, reproducibility reports, interface problems, and institutional-deployment support

Primary Research Contact Scientific coordination and visualization methodology Gerardo Tinoco-Guerrero Universidad Michoacana de San Nicolás de Hidalgo Morelia, Michoacán, Mexico

Repository Support Reproducible software and deployment reports Use GitHub Issues for reproducible software or documentation problems. Include the operating system, Node.js version, exact command, and complete error message. Report whether npm run verify succeeds locally. For server problems, include the requested URL and HTTP status without sharing credentials.

---

💬 FAQ

Does ADR1D Explorer run a numerical or neural model in the browser?

No. It loads validated static fields and metrics. Browser interaction changes the selected scenario, time, method, language, filters, and presentation state only.

Why are only 14 scenarios interactive?

The subset was selected by a documented diagnostic policy to cover all four physical regimes, representative cases, boundary profiles, large method disagreements, and global error extremes. Aggregate context still refers to the complete 120-scenario comparison.

Can I deploy the build under a directory instead of a subdomain?

Yes. All runtime resources use relative URLs. Copy the contents of dist/ into the approved ADR1D_explorer/ directory on either institutional server.

Does the GitHub workflow publish the application automatically?

No. It validates, tests, builds, and uploads a downloadable workflow artifact. Only the repository owner or an authorized server administrator publishes files to the institutional servers.

Can I open dist/index.html directly from the file system?

Use a local HTTP server instead. Browser security policies can block JSON requests from file:// pages. npm run preview serves the production build correctly.

Are the displayed results suitable for field or regulatory decisions?

No. The current evidence is synthetic, one-dimensional, and limited to the documented parameter space and methods. Field use requires site-specific data, calibration, uncertainty assessment, and independent validation.

How should a new release be deployed?

Tag and archive the validated source release, run npm run verify, replace the complete contents of the institutional ADR1D_explorer/ directory with the new dist/ contents, and then repeat the post-deployment checks.

---

Making validated reactive-transport evidence easier to inspect, compare, and communicate

If ADR1D Explorer supports your research, please cite the software and its upstream scientific resources.