Using eBFE Models: Rio Hondo 1D Steady Collection¶

This notebook demonstrates the Rio Hondo (13060008) eBFE/BLE 1D steady model collection. The delivery is different from the large 2D eBFE examples: it contains hundreds of small 1D steady HEC-RAS projects, and results HDF files are generally absent until the steady plans are run.

The validation target for this collection is therefore straightforward: organize the projects, run each steady plan through ras-commander, read the detailed compute messages, and confirm every selected plan completed without blocking errors.

Workflow¶

Use RasEbfeModels.organize_model("rio-hondo") so downloads, extraction, and organization use the same ras-commander entry point as the audit scripts.
Discover valid HEC-RAS project folders with RasUtils.find_valid_ras_folders().
Run the sequential steady-plan validation batch script for the full collection.
Read the JSON/Markdown evidence report and inspect compute-message summaries.

The full steady-plan batch can take a while because it touches 253 projects, so it is intentionally opt-in inside the notebook.

Python

from pathlib import Path
import json
import os
import subprocess
import sys

import pandas as pd

try:
    from ras_commander import HdfResultsPlan, RasPrj, RasUtils, init_ras_project
    from ras_commander.results.ResultsParser import ResultsParser
    from ras_commander.sources.federal import RasEbfeModels
except ImportError:
    repo_root = Path.cwd().parent
    sys.path.insert(0, str(repo_root))
    from ras_commander import HdfResultsPlan, RasPrj, RasUtils, init_ras_project
    from ras_commander.results.ResultsParser import ResultsParser
    from ras_commander.sources.federal import RasEbfeModels

Configure the eBFE Workspace¶

The shared eBFE workspace is on H:\\Testing\\eBFE Model Organization. Set RAS_COMMANDER_EBFE_ROOT if you want to use a different local cache.

Python

EBFE_WORKSPACE = Path(os.environ.get("RAS_COMMANDER_EBFE_ROOT", r"H:\Testing\eBFE Model Organization"))
DOWNLOAD_ROOT = EBFE_WORKSPACE / "Downloads"
ORGANIZED_ROOT = EBFE_WORKSPACE / "Organized"
VALIDATION_ROOT = EBFE_WORKSPACE / "Validation" / "ebfe_delivery"
repo_candidates = [Path.cwd(), Path.cwd().parent]
VALIDATION_MATRIX = next(
    (candidate / "VALIDATION_MATRIX.md" for candidate in repo_candidates if (candidate / "VALIDATION_MATRIX.md").exists()),
    Path.cwd() / "VALIDATION_MATRIX.md",
)

organized_folder = ORGANIZED_ROOT / "RioHondo_13060008"
if not organized_folder.exists():
    organized_folder = RasEbfeModels.organize_model(
        "rio-hondo",
        download_root=DOWNLOAD_ROOT,
        output_root=ORGANIZED_ROOT,
    )

ras_model_root = organized_folder / "RAS Model"
print(f"Rio Hondo organized folder: {organized_folder}")
print(f"RAS Model root exists: {ras_model_root.exists()}")
print(f"Validation root: {VALIDATION_ROOT}")
print(f"Validation matrix: {VALIDATION_MATRIX}")

Text Only

Rio Hondo organized folder: H:\Testing\eBFE Model Organization\Organized\RioHondo_13060008
RAS Model root exists: True
Validation root: H:\Testing\eBFE Model Organization\Validation\ebfe_delivery
Validation matrix: C:\GH\ras-commander\VALIDATION_MATRIX.md

Discover the RAS Projects¶

Rio Hondo is a 1D steady collection, so the unit of validation is each individual HEC-RAS project folder.

Python

projects = RasUtils.find_valid_ras_folders(
    ras_model_root,
    max_depth=10,
    return_project_info=True,
)
project_df = pd.DataFrame(projects)
print(f"Discovered {len(project_df)} HEC-RAS project folders")
project_df.head()

Text Only

2026-04-28 23:37:26 - ras_commander.RasUtils - INFO - Searching for HEC-RAS projects in: H:\Testing\eBFE Model Organization\Organized\RioHondo_13060008\RAS Model


2026-04-28 23:37:34 - ras_commander.RasUtils - INFO - Found 253 valid HEC-RAS project folders


Discovered 253 HEC-RAS project folders

	folder	project_name	prj_file	plan_count	plan_numbers
0	H:\Testing\eBFE Model Organization\Organized\R...	BarnDraw	H:\Testing\eBFE Model Organization\Organized\R...	1	[01]
1	H:\Testing\eBFE Model Organization\Organized\R...	BarnDrawTrib1	H:\Testing\eBFE Model Organization\Organized\R...	1	[01]
2	H:\Testing\eBFE Model Organization\Organized\R...	BarnDrawTrib1A	H:\Testing\eBFE Model Organization\Organized\R...	1	[01]
3	H:\Testing\eBFE Model Organization\Organized\R...	BerrendoCreek	H:\Testing\eBFE Model Organization\Organized\R...	1	[01]
4	H:\Testing\eBFE Model Organization\Organized\R...	BerrendoCreekTrib	H:\Testing\eBFE Model Organization\Organized\R...	1	[01]

Inspect One Project¶

A quick initialization pass confirms that ras-commander can read the project metadata before running the full batch.

Python

sample_project = Path(project_df.iloc[0]["folder"])
sample_ras = RasPrj()
init_ras_project(
    sample_project,
    "6.6",
    ras_object=sample_ras,
    load_results_summary=False,
)

sample_ras.plan_df[["plan_number", "flow_type", "Geom File", "Flow File", "HDF_Results_Path"]]

Text Only

2026-04-28 23:37:34 - ras_commander.RasPrj - INFO - No unsteady flow files found in the project.


2026-04-28 23:37:34 - ras_commander.RasMap - WARNING - RASMapper file not found: H:\Testing\eBFE Model Organization\Organized\RioHondo_13060008\RAS Model\Berrendo Creek\BarnDraw\BarnDraw.rasmap


2026-04-28 23:37:34 - ras_commander.RasMap - WARNING - No .rasmap file found for this project. Creating empty rasmap_df.

	plan_number	flow_type	Geom File	Flow File	HDF_Results_Path
0	01	Steady	01	01	H:\Testing\eBFE Model Organization\Organized\R...

Run or Reuse the Full Steady-Plan Validation¶

Set RUN_FULL_STEADY_VALIDATION = True to run all Rio Hondo steady plans. The script uses RasCmdr.compute_plan(..., verify=True), then reads detailed compute messages and writes evidence under Validation/ebfe_delivery/steady_plan_validation.

The saved notebook run keeps the batch compute disabled by default and loads the latest existing validation report when one is present. That makes the committed example fast while still showing the evidence trail used for acceptance.

Python

RUN_FULL_STEADY_VALIDATION = False

repo_root = Path.cwd().parent if Path.cwd().name == "examples" else Path.cwd()
batch_script = repo_root / "scripts" / "ebfe_steady_plan_batch.py"
cmd = [
    sys.executable,
    str(batch_script),
    "--root",
    str(ras_model_root),
    "--ras-version",
    "6.6",
    "--output-dir",
    str(VALIDATION_ROOT / "steady_plan_validation"),
]

if RUN_FULL_STEADY_VALIDATION:
    subprocess.run(cmd, check=True)
else:
    print("Full validation is opt-in. Command to run:")
    print(" ".join(f'\"{part}\"' if " " in part else part for part in cmd))

Text Only

Full validation is opt-in. Command to run:
C:\Users\billk_clb\anaconda3\python.exe C:\GH\ras-commander\scripts\ebfe_steady_plan_batch.py --root "H:\Testing\eBFE Model Organization\Organized\RioHondo_13060008\RAS Model" --ras-version 6.6 --output-dir "H:\Testing\eBFE Model Organization\Validation\ebfe_delivery\steady_plan_validation"

Review Validation Evidence¶

After the batch run completes, this cell loads the latest JSON report and summarizes project pass/fail status.

Python

report_dir = VALIDATION_ROOT / "steady_plan_validation"
reports = sorted(report_dir.glob("steady_plan_validation_*.json"))

if not reports:
    print(f"No steady validation reports found yet in {report_dir}")
else:
    latest_report = reports[-1]
    payload = json.loads(latest_report.read_text(encoding="utf-8"))
    records = payload["records"]
    validation_df = pd.DataFrame(
        {
            "project": record["project_name"],
            "status": record["status"],
            "plans": len(record.get("plans", [])),
            "seconds": record.get("elapsed_seconds"),
        }
        for record in records
    )
    print(f"Loaded existing report: {latest_report}")
    if not RUN_FULL_STEADY_VALIDATION:
        print("Saved notebook run reused existing validation evidence; set RUN_FULL_STEADY_VALIDATION = True to regenerate it.")
    display(validation_df["status"].value_counts())
    validation_df.head()

Text Only

Loaded existing report: H:\Testing\eBFE Model Organization\Validation\ebfe_delivery\steady_plan_validation\steady_plan_validation_20260424_160022.json
Saved notebook run reused existing validation evidence; set RUN_FULL_STEADY_VALIDATION = True to regenerate it.



status
passed    253
Name: count, dtype: int64

Inspect Detailed Compute Messages¶

For this collection, results are considered resolved when the generated HDFs contain Complete Process and the parsed compute messages do not contain blocking errors.

Python

sample_hdf = next(ras_model_root.glob("**/*.p01.hdf"), None)

if sample_hdf is None:
    print("No plan HDF found yet. Run the steady validation batch first.")
else:
    messages = HdfResultsPlan.get_compute_messages_hdf_only(sample_hdf)
    parsed = ResultsParser.parse_compute_messages(messages)
    print(sample_hdf)
    parsed

Text Only

2026-04-28 23:37:34 - ras_commander.hdf.HdfResultsPlan - INFO - Using existing Path object HDF file: H:\Testing\eBFE Model Organization\Organized\RioHondo_13060008\RAS Model\Rio Ruidoso\BradyCanyon\BradyCanyon.p01.hdf


2026-04-28 23:37:34 - ras_commander.hdf.HdfResultsPlan - INFO - Final validated file path: H:\Testing\eBFE Model Organization\Organized\RioHondo_13060008\RAS Model\Rio Ruidoso\BradyCanyon\BradyCanyon.p01.hdf


2026-04-28 23:37:34 - ras_commander.hdf.HdfResultsPlan - INFO - Reading computation messages from HDF: BradyCanyon.p01.hdf


2026-04-28 23:37:34 - ras_commander.hdf.HdfResultsPlan - INFO - Successfully extracted 715 characters from HDF


H:\Testing\eBFE Model Organization\Organized\RioHondo_13060008\RAS Model\Rio Ruidoso\BradyCanyon\BradyCanyon.p01.hdf

Delivery-Format Interpretation¶

Unlike the larger 2D eBFE projects, Rio Hondo does not depend on terrain or land cover layers for validation. The key delivery-format proof is that each 1D steady project can initialize, run its steady plan, produce a plan HDF, and expose detailed compute messages confirming successful completion.

In the committed notebook output, the expensive batch run is opt-in and the notebook loads the latest existing validation report from the shared eBFE workspace. Regenerate that report with RUN_FULL_STEADY_VALIDATION = True whenever the organized source data or ras-commander execution path changes.