CUAHSI Summer Institute 2026 ·
The uncalibrated NOAA-OWP HAND synthetic rating curve is Manning (Q ∝ √S), so the river slope S is an
important control on the mapped flood extent. This repository asks whether replacing that slope with a
static-satellite product (IRIS-SWORD, SWOT) or a time-varying gauge-derived S(Q) measurably changes HAND-FIM
skill, and where (by hydraulic regime) it matters. One notebook runs the whole study end to end over six reaches
and scores every slope treatment against FIMBench.
code/TV_Slope_FIM.ipynb the study notebook (run this)
code/tvslope_src/engine/ analysis modules
code/tvslope_src/fimbox_ext/ build HAND + generate the FIM
data/ small derived data
code/get_data.py downloads the large datasets that are not provided (FIMBench, SWORD)
conda env create -f environment.yml && conda activate slope
python code/get_data.py # FIMBench + SWORD (large; not in the repo)
TV_Slope_FIM.ipynbThe notebooks also use three HAND-FIM tools from the SDML lab. Install the ones you need:
pip install fimeval # query + download FIMBench benchmarks (needed by study_area / timevary)
pip install fimserve # HAND-FIM staging and serving (needed to (re)generate FIM)
pip install "git+https://github.com/sdmlua/fimbox" # HAND-FIM generation (needed to (re)generate FIM)
Provided in data/
| File | What it is | Provenance |
|---|---|---|
data/slope_treatments.csv |
Per-reach slope for each treatment (hydrofabric, IRIS-SWORD, SWOT median/floodstage/maxWSE) | Derived in this study |
data/FIMHF_IRIS_new.csv, data/FIMHF_IRIS_v1.0.csv |
IRIS-SWORD static slope | Built from IRIS v3.3 + SWORD v17b, following Chen et al. (2025) |
data/paired_reach_SWOT_gage/gauge_latlon.csv |
USGS gauge coordinates (id, name, lat, lon) | USGS NWIS |
data/study_area_gauges.csv |
Same-river upstream/on-reach/downstream gauge triplets per reach | Derived in this study |
data/fimbox_bankfull_2yr_cms.parquet |
2-year recurrence (bankfull) discharge per NWM feature_id | NWM recurrence flows |
data/us_states.gpkg |
US state boundaries for the CONUS map | Public US state boundaries |
Fetched by code — Run python code/get_data.py.
| Data | How to get it | Source |
|---|---|---|
FIMBench benchmark flood maps (data/FIMBench/) |
python code/get_data.py (uses fimeval) |
FIMbench (https://tethys.ciroh.org/apps/fimbench-gui/) |
SWORD v17b river network (data/SWORD_v17b_gpkg/na_sword_reaches_v17b.gpkg) |
python code/get_data.py |
SWORD v17 (https://zenodo.org/records/15299138) |
USGS gauge discharge & stage (data/discharge/, data/twin_gauge/) |
cached on the first notebook run (dataretrieval) |
USGS NWIS |
NWM hydrofabric + 3DEP DEM + staged HAND (data/fimbox_out/) |
Rebuilt by the notebook when REGEN_FIM=1 (needs fimbox) |
NWM / USGS 3DEP, via FIMbox |
- IRIS v3.3 (ICESat-2 River Surface Slope) — Scherer, D., Schwatke, C., Dettmering, D., & Seitz, F. (2022). ICESat-2 based River Surface Slope and Its Impact on Water Level Time Series From Satellite Altimetry. Water Resources Research. https://doi.org/10.1029/2022WR032842 · data: https://zenodo.org/records/14616464
- SWORD v17b (SWOT River Database) — Elizabeth H. Altenau, Tamlin M. Pavelsky, Michael T. Durand, Xiao Yang, Renato P. d. M. Frasson& Liam Bendezu. (2025). SWOT River Database (SWORD) (Version v17b) [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.15299138 · data: https://doi.org/10.5281/zenodo.14727521 · https://www.swordexplorer.com/
- IRIS-SWORD slope — Chen, Y., Cohen, S., Baruah, A., Devi, D., Dhital, S., Tian, D., & Munasinghe, D. (2025). Merging Remote Sensing Derived River Slope Datasets with High-Resolution Hydrofabrics for the United States. Scientific Data, 12(1), 1657.
- National Water Model (retrospective + operational short-range forecast) — NOAA Office of Water Prediction.
- USGS gauge data and 3DEP 10 m DEM — U.S. Geological Survey.
- FIMBench benchmark flood maps — Surface Dynamics Modeling Lab, University of Alabama.
| Path | Origin | What we changed |
|---|---|---|
code/tvslope_src/engine/*.py |
Our own code | Written for this study. timevarying_slope.py adapts the TimeVariantSlope S(Q) method; fim_reach.py and fim_eval.py reimplement the RiverJoin river-matching and FIMeval scoring concepts. |
code/tvslope_src/fimbox_ext/*.py |
Built on FIMbox (github.com/sdmlua/fimbox) | We modified the code from FIMbox: observation-calibration disabled, per-treatment slope injection, a branch-0 tributary gap-filler, and NWM-forecast forcing. Each file carries a Built on FIMbox header noting the source and the change. |
| Tool | Owner | License | Repository |
|---|---|---|---|
| FIMbox | Surface Dynamics Modeling Lab (Univ. of Alabama) | GPL-3.0 | https://github.com/sdmlua/fimbox |
| FIMserv | SDML | see repo | https://github.com/sdmlua/FIMserv |
| FIMeval | SDML | see repo | https://github.com/sdmlua/fimeval |
| FIMbench | SDML | see repo | https://github.com/sdmlua/fimbench |
| RiverJoin | SDML | see repo | https://github.com/sdmlua/riverjoin_py |
| NOAA-OWP/inundation-mapping | NOAA Office of Water Prediction | see repo | https://github.com/NOAA-OWP/inundation-mapping |
License note. FIMbox is GPL-3.0. Because the drivers in
code/tvslope_src/fimbox_ext/build on FIMbox, they are likewise distributed under GPL-3.0. Add a top-levelLICENSEfile before publishing more widely.
| Name | Institution | GitHub |
|---|---|---|
| Zih-Syun Chen | University of Houston | @zixunn |
| Sebastian Marshall | Johns Hopkins University | @rushmarshall |
| Pitamber Wagle | Brigham Young University | @Pitamberwagle |
| Reza Jamshidi | Northeastern University | @Reza-Jamshidi |
Theme leads: Sagy Cohen and Anupal Baruah, University of Alabama
CUAHSI Summer Institute 2026 · Team Slippery Slope
University of Houston · Johns Hopkins University · Brigham Young University · Northeastern University