Integrated Flood Risk Assessment and Automated Meander Classification in the Southeastern United States (FloodMMAT)
| This repository provides an R package and source code for integrated flood risk assessment, incorporating automated meander classification using deep learning and hydraulic modeling, specifically tailored for the Southeastern United States. It is developed by [Pawan Thapa/University of Alabama], [Department of Geography and the Environment], United States. |
Accurate flood risk assessment in complex fluvial environments like the Southeastern United States requires integrated approaches that capture the interplay between meander morphology and hydrological processes. Traditional methods often lack the automation and precision needed to effectively analyze these dynamic systems.
To address these challenges, we developed the FloodMMAT R package. This package combines deep learning-based automated meander classification with hydraulic modeling to enhance flood risk analysis. It leverages geometric properties and integrates various geospatial datasets to provide a comprehensive assessment framework.
The architecture of FloodMMAT integrates key functions for deep learning-based meander classification, hydraulic modeling, and flood risk assessment. All function codes are in the R folder.
FloodMMAT/
├── R/
│ ├── meander_classification.R
│ ├── hydraulic_modeling.R
│ ├── flood_risk_assessment.R
├── man/
│ ├── meander_classification.Rd
│ ├── hydraulic_modeling.Rd
│ ├── flood_risk_assessment.Rd
├── data/
│ ├── meander_images/ # Meander image dataset
│ ├── geospatial_data/ # SRTM DEM, JRC Water, CHIRPS, NLCD, OpenLandMap
├── inst/
│ └── doc/
│ └── FloodMMAT_usage.Rmd
├── DESCRIPTION
├── NAMESPACE
├── README.md
└── LICENSE
The workflow of the FloodMMAT package can be summarized as follows:
Meander Classification: Use meander_classification() to classify meanders based on geometric properties using deep learning models.
Hydraulic Modeling: Use hydraulic_modeling() to estimate stream flow characteristics using Manning's equation, incorporating meander classifications.
Flood Risk Assessment: Use flood_risk_assessment() to assess flood risk by integrating meander morphology, flow dynamics, and geospatial data.
Output: Obtain flood risk assessments, meander classifications, and hydraulic modeling results as spatial data frames or data frames.
Package Installation and Usage
This package can be installed directly from GitHub.
Code snippet
# Install from GitHub
if (!requireNamespace("remotes", quietly = TRUE)) {
install.packages("remotes")
}
remotes::install_github("yourusername/FloodMMAT")
# Load the package
library(FloodMMAT)
Note: Detailed usage instructions and examples are provided in the package documentation and vignettes (accessible with vignette("FloodMMAT_usage")).
Data Input
The package requires:
A dataset of meander images for deep learning-based classification.
Geospatial data from SRTM DEM, JRC Global Surface Water, CHIRPS rainfall, NLCD land cover, and OpenLandMap soil data.
Executing the FloodMMAT Functions
The FloodMMAT package provides functions for meander classification, hydraulic modeling, and flood risk assessment.
Code snippet
# Example usage:
meander_classifications <- meander_classification(meander_images_path, model = "ResNet152") # Example path and model choice
hydraulic_results <- hydraulic_modeling(meander_classifications, dem_path, water_surface_path) # Example paths
flood_risk_results <- flood_risk_assessment(hydraulic_results, land_cover_path, soil_data_path, rainfall_path) #Example paths
Functions in FloodMMAT
| Function Name | Objective 3. # Include the necessary libraries.
4 import os
5 import rasterio
6 from rasterio.mask import mask
7 from shapely.geometry import mapping
8 import geopandas as gpd
9 import numpy as np
10 import matplotlib.pyplot as plt
11 import contextily as ctx
12
13 # Function to extract river centerline and estimate width
14 def extract_river_data(raster_path, output_dir, spatial_resolution=10):
15 """
16 Extracts river centerline and estimates width from a raster image.
17
18 Args:
19 raster_path (str): Path to the raster image.
20 output_dir (str): Directory to save the output shapefiles.
21 spatial_resolution (int): Spatial resolution of the raster.
22
23 Returns:
24 tuple: Paths to the centerline and width shapefiles.
25 """
26 # Placeholder for centerline extraction and width estimation functions
27 # Replace with your actual functions or library calls
28 print(f"Extracting centerline and estimating width from {raster_path}")
29 centerline_gdf = gpd.GeoDataFrame() # Replace with actual centerline GeoDataFrame
30 width_df = pd.DataFrame() # Replace with actual width DataFrame
31
32 # Save centerline and width to shapefiles
33 centerline_path = os.path.join(output_dir, "centerline.shp")
34 width_path = os.path.join(output_dir, "width.shp")
35 centerline_gdf.to_file(centerline_path)
36 width_df.to_file(width_path)
37
38 return centerline_path, width_path
39
40 # Function to perform flood risk assessment
41 def flood_risk_assessment(centerline_path, width_path, dem_path, water_surface_path, land_cover_path, soil_data_path, rainfall_path, output_dir):
42 """
43 Performs flood risk assessment by integrating meander morphology, flow dynamics, and geospatial data.
44
45 Args:
46 centerline_path (str): Path to the centerline shapefile.
47 width_path (str): Path to the width shapefile.
48 dem_path (str): Path to the DEM raster.
49 water_surface_path (str): Path to the water surface raster.
50 land_cover_path (str): Path to the land cover raster.
51 soil_data_path (str): Path to the soil data raster.
52 rainfall_path (str): Path to the rainfall raster.
53 output_dir (str): Directory to save the flood risk assessment results.
54
55 Returns:
56 str: Path to the flood risk assessment shapefile.
57 """
58 # Placeholder for flood risk assessment function
59 # Replace with your actual flood risk assessment function or library calls
60 print("Performing flood risk assessment...")
61 flood_risk_gdf = gpd.GeoDataFrame() # Replace with actual flood risk GeoDataFrame
62
63 # Save flood risk assessment results to shapefile
64 flood_risk_path = os.path.join(output_dir, "flood_risk.shp")
65 flood_risk_gdf.to_file(flood_risk_path)
66
67 return flood_risk_path
68
69 # Function to visualize flood risk results
70 def visualize_flood_risk(flood_risk_path):
71 """
72 Visualizes the flood risk assessment results.
73
74 Args:
75 flood_risk_path (str): Path to the flood risk assessment shapefile.
76 """
77 # Placeholder for flood risk visualization function
78 # Replace with your actual visualization function or library calls
79 print("Visualizing flood risk results...")
80 flood_risk_gdf = gpd.read_file(flood_risk_path)
81
82 fig, ax = plt.subplots(figsize=(12, 10))
83 flood_risk_gdf.plot(ax=ax, column='flood_risk', legend=True) # Assuming 'flood_risk' column contains risk values
84 ctx.add_basemap(ax, crs=flood_risk_gdf.crs, source=ctx.providers.CartoDB.Positron)
85 plt.title("Flood Risk Assessment")
86 plt.show()
87
88 # Main script
89 if name == "main":
90 # Example usage:
91 raster_path = "/path/to/your/raster.tif" # Replace with your raster path
92 output_dir = "/path/to/output/directory" # Replace with your output directory
93 dem_path = "/path/to/your/dem.tif" # Replace with your DEM path
94 water_surface_path = "/path/to/your/water_surface.tif" # Replace with your water surface path
95 land_cover_path = "/path/to/your/land_cover.tif" # Replace with your land cover path
96 soil_data_path = "/path/to/your/soil_data.tif" # Replace with your soil data path
97 rainfall_path = "/path/to/your/rainfall.tif" # Replace with your rainfall path
98
99 # Extract river data
100 centerline_path, width_path = extract_river_data(raster_path, output_dir)
101
102 # Perform flood risk assessment
103 flood_risk_path = flood_risk_assessment(centerline_path, width_path, dem_path, water_surface_path, land_cover_path, soil_data_path, rainfall_path, output_dir)
104
105 # Visualize flood risk results
106 visualize_flood_risk(flood_risk_path)