Checking for non-preferred file/folder path names (may take a long time depending on the number of files/folders) ...

Geospatial Data Processing Pipeline for Hydrologic and Geomorphic Analysis and Modeling


Authors:
Owners: This resource does not have an owner who is an active HydroShare user. Contact CUAHSI (help@cuahsi.org) for information on this resource.
Type: Resource
Storage: The size of this resource is 156.8 MB
Created: Feb 26, 2026 at 9:23 p.m. (UTC)
Last updated: Feb 27, 2026 at 7:24 a.m. (UTC)
Citation: See how to cite this resource
Content types: Single File Content 
Sharing Status: Public
Views: 87
Downloads: 7
+1 Votes: 1 other +1 this
Comments: No comments (yet)

Abstract

This HydroShare resource archives a reproducible geospatial data processing pipeline designed for hydrologic and geomorphic analysis and modeling. The workflow standardizes heterogeneous raster datasets from sources such as OpenTopography, USDA soil databases, NLCD, MTBS, and satellite platforms including Landsat and MODIS into spatially aligned, modeling-ready grids. Input layers include digital elevation models, soil properties, burn severity, land cover, and satellite-derived ecohydrologic indicators. The pipeline performs coordinate reference system harmonization, resolution enforcement, raster resampling, grid alignment, and export to analysis-ready formats suitable for hydrologic and geomorphic modeling, feature engineering, and machine learning workflows. While example outputs demonstrate integration with terrain–hydrology simulations and landslide probability assessment, the primary contribution of this resource is the reproducible raster harmonization framework. The workflow is broadly applicable to watershed analysis, landscape evolution studies, hydro-geomorphic process modeling, and other geospatial preprocessing tasks requiring consistent spatial alignment and physically interpretable feature derivation.

Subject Keywords

Content

README.md

Geospatial Data Processing Pipeline for Hydrologic and Geomorphic Analysis and Modeling

Overview

This HydroShare resource archives a reproducible snapshot of the fire-debrisflow-ml workflow for raster preprocessing and downstream hydrologic and geomorphic modeling.

The focus of this archive is the data pipeline, which standardizes DEM, soils, burn severity, landcover, and derived ecohydrologic features into spatially aligned grids suitable for physics-based modeling and machine learning prediction.

A Landlab notebook is included for demonstration of probabilistic landslide estimation using the processed outputs.


Development Repository

Active development of the pipeline is maintained at:

https://github.com/Almehedi06/fire-debrisflow-ml

This HydroShare resource represents a versioned, archived snapshot for reproducible dissemination and scholarly reference.


Installation

Conda (Recommended)

conda env create -f environment.yml\ conda activate fire-debrisflow-ml

If using existing environment:

conda activate ml_debris

Known Compatible Snapshot

  • Python 3.10\
  • numpy\
  • rasterio\
  • fiona\
  • geopandas\
  • shapely\
  • pyproj\
  • landlab\
  • bmi-topography\
  • scikit-learn\
  • xgboost\
  • torch

Pip (if geospatial stack already installed)

pip install -r requirements.txt


Data Pipeline

The preprocessing pipeline:

python src/run_pipeline.py --config config/base.yaml --export-final-tifs

Pipeline stages:

  1. AOI ingestion and CRS standardization\
  2. DEM acquisition (USGS via bmi-topography)\
  3. Soil property harmonization (texture, bulk density, depth, etc.)\
  4. Burn severity and landcover alignment\
  5. Raster resampling and grid harmonization\
  6. Export of aligned .asc and .tif layers

Satellite-Derived Features

Ecohydrologic features may be derived from satellite platforms (e.g., Landsat), including:

  • NDVI / LAI\
  • Soil moisture indicators\
  • Snow water equivalent\
  • Snow cover\
  • Evapotranspiration proxies

Derived rasters are aligned to the pipeline grid prior to modeling.


Landlab Modeling

The included notebook:

Multi_Model_Probability.ipynb

Demonstrates:

  • Flow routing and slope computation\
  • Soil moisture dynamics\
  • Recharge aggregation\
  • Monte Carlo landslide probability estimation

The modeling component uses outputs generated by the data pipeline.


Funding

This work is supported in part by:

  • NSF 2303870 RAPID: Monitoring Postfire Geomorphic Response on Humid Slopes\
  • NSF 2530591 Collaborative Research: CAIG -- Framework for Artificial Intelligence-Enhanced Modeling

Intended Use

  • Raster harmonization for hydrologic and geomorphic analysis\
  • AI-ready feature generation\
  • Watershed-scale preprocessing\
  • Reproducible geospatial workflow archiving

Citation

Please cite this HydroShare resource using the citation provided on the resource page.

Credits

Funding Agencies

This resource was created using funding from the following sources:
Agency Name Award Title Award Number
NSF Collaborative Research: CAIG: Framework for Artificial Intelligence-Enhanced Modeling of Wildfire Geohazards (FAIM-WG) 2530591
NSF RAPID: Monitoring postfire geomorphic response on humid slopes of the North Cascade Range, Washington 2303870

Contributors

People or Organizations that contributed technically, materially, financially, or provided general support for the creation of the resource's content but are not considered authors.

Name Organization Address Phone Author Identifiers
Erkan Istanbulluoglu University of Washington WA, US
Hunter Jimenez University of Washington WA, US

How to Cite

Mehedi, M. A. A., H. Jimenez, E. Istanbulluoglu (2026). Geospatial Data Processing Pipeline for Hydrologic and Geomorphic Analysis and Modeling, HydroShare, http://www.hydroshare.org/resource/ec6ffad4d9594885bbc23e196e22949a

This resource is shared under the Creative Commons Attribution CC BY.

http://creativecommons.org/licenses/by/4.0/
CC-BY

Comments

There are currently no comments

New Comment

required