Landslide prediction is important. There are negative economic, transportation, and habitat impacts resulting from landslides. There are climate change implications where drought or heavier rains affect the number and severity of landslides. The goal of this project is to examine the relationship of soil moisture and precipitation over Colorado, USA. Colorado experiences many landslides each year because of its steep terrain. Some of them occur in remote areas that are difficult to monitor, with most occurring west of the Front Range. Damage from landslides in Colorado is estimated to be millions of dollars per year. A significant fraction of the Colorado population lives in the Rockies. It is also a popular tourist destination year-round and people are moving into the Rockies - it’s a popular area, to say the least.

• Soil moisture can be a potential indicator of the type of rainfall induced landslide. We can use soil moisture data to group landslides into two types (1) shallow slope failures - saturation induced by rainfall infiltration and (2) run off driven landslides - triggered by intense storms. This informs the forecaster which models and precipitation products are optimal to use to better predict a landslide event.
• Code presented here is helpful to those who want to know how to read/extract soil moisture parameters measured from NASA, USGS, and ESA satellites.
• The workflow should perform with any state who's landslide data is archived in the NASA Global Landslide Catalog (GLC).

• os
• glob
• datetime
• earthpy
• geopandas
• folium
• h5py (pip install h5py)
• matplotlib
• numpy
• pandas
• re
• rioxarray
• scipy
• seaborn
• shapely
• xarray

To install packages separately, using conda, type `conda install -c conda-forge name-of-package-listed-above`

This repository contains a file called environment.yml that contains the python packages to install the capstone-landslides-soilmoisture environment. Included is installation of Jupyter notebook and it's dependencies.

• Clone/Fork this repository.
• Once you are in the capstone-landslides-soilmoisture directory, you can create the environment. To do this run: conda env create -f environment.yml.
• Once the environment is installed you can activate it using: conda activate capstone-landslides-soilmoisture.
• To view a list of all conda environments available on your machine run: conda info --envs.
• To view a list of all conda packages installed in capstone-landslides-soilmoisture run: conda list.

• Soil moisture and precipitation satellite data are co-located with 2015-2020 landslide events in Colorado and other US states affected by landslides using the Global Landslide Catalog (GLC) (refer to "Data & Formats" below).
• CSV files have been created and stored in the data directory that ties all the various soil and precipitation data together and can be read using the pandas package.
• If users simply want to work with the analysis portion, the can clone this repo and work with the CSV data files contained in the data folder

1. capstone-study-area-final.ipynb

   • An introduction the study region and GLC statistics over Colorado.

2. landslide_precip_soilm_DataExport_final.ipynb

   • Co-located SMAP and ESA CCI soil moistures with GPM precipitation products (daily and 30 min resolutions)

   • Data Directories

     • To run landslide_precip_soilm_DataExport_final.ipynb notebook, you need to download daily 2015-2020 data listed in the "Data & Formats" section below.
     • Note: The data files are large, particularly for SMAP which ignore subsetting.
     • Links to the data and instructions on what specifically to download are provided used in this study are provided in the "Data & Formats" section below

   • Data directory structure under your home space (Mac/Linux syntax shown below)

     • GLC data directory: earth-analytics/data/capstone/landslide
     • GPM daily precipitation directory: earth-analytics/data/capstone/gpm_westernUS
     • GPM IMERG 30min precipitation directory: earth-analytics/data/capstone/precip_imerg
     • SMAP daily soil moisture directory: earth-analytics/data/capstone/smap_9km
     • ESA CCI soil moisture directory: earth-analytics/data/capstone/esa_soil_moisture

   • Users can choose a US state where landslides are archived in the GLC. Currently, the following states are included:

     • Colorado
     • Idaho
     • Utah
     • California
     • Oregon
     • Washington

   • The result are two a pandas dataFrames that are exported as CSV files to the data folder. For example,

     • glc_smap_esa_gpm_2015-2020_Colorado.csv contains the 7-day accummulated precipitation and maximum soil moisture co-located to a GLC landslide.

     • glc_smap_esa_gpm_2015-2020_7day_Colorado.csv contains the precipitation and soil moisture co-located to a GLC landslide going back for 7-days prior to the landslide.

3. landslide_precip_soilm_DataAnalyses_final.ipynb

   • Plot analysis that reads the exported CSV files created by landslide_precip_soilm_DataExport_final.ipynb above.
   • Output are plots inline with the notebooks. Noteworthy plots are exported as PNG files saved under the plots folder.

1. NASA Global Landslide Catalog (2007-2020)
   • LINK: https://maps.nccs.nasa.gov/arcgis/apps/MapAndAppGallery/index.html?appid=574f26408683485799d02e857e5d9521
   • Instructions: From the list of downloadables, choose `NASA Global Landslide Catalog Points (CSV)`
   • Citation: https://doi.org/10.1007/s11069-009-9401-4
   • Format: CSV format
2. NASA SMAP Enhanced L3 Radiometer Global Daily 9 km EASE-Grid Soil Moisture V004
   • LINK: Data downloadable via Earthdata: https://search.earthdata.nasa.gov
   • Instructions: Search under 'SMAP soil moisture' and find the dataset that matches the title above. NOTE: Subsetting is ignored so global files are downloaded.
   • Citation: https://doi.org/10.5067/NJ34TQ2LFE90
   • Format: HDF5 format
3. ESA Climate Change Initiative (CCI) ACTIVE soil moisture 0.25 degree x 0.25 degree V03.3
   • LINK: https://cds.climate.copernicus.eu/cdsapp#!/dataset/satellite-soil-moisture?tab=overview
   • Instructions: Choose the 'ACTIVE' dataset which calculates the percent saturation soil moisture.
   • Citation: https://doi.org/10.1016/j.rse.2012.03.014
   • Format: netCDF3 format
4. GPM IMERG Late Precipitation L3 1 day 0.1 degree x 0.1 degree V06
   • LINK: https://disc.gsfc.nasa.gov/datasets/GPM_3IMERGDL_06/summary?keywords=%22IMERG%20late%22
   • Instructions: Subsetting available via GES DISC over the globe. You can choose Colorado or any state(s) of interest.
   • Nearest neighbor remapping of fields between grids in spherical coordinates.
   • Citation: https://doi.org/10.5067/GPM/IMERGDL/DAY/06
   • Format: netCDF4 format
5. GPM IMERG Final Precipitation L3 Half Hourly 0.1 degree x 0.1 degree V06
   • This capstone uses 30 min GPM IMERG data pre-assembled and saved in CSV format by Dr. Elsa Culler (CU-Boulder). The beauty of this data set is that precipitation values have been subset and co-located with GLC landslides by their ID, latitude, and longitude.
   • Instructions: Request this data set by emailing [email protected]
   • Citation: https://doi.org/10.5067/GPM/IMERG/3B-HH/06
   • Format: CSV format
6. POLARIS 30m Probabilistic Soil Properties US
   • LINK: http://hydrology.cee.duke.edu/POLARIS/PROPERTIES/v1.0/ksat/mode/5_15/
   • Instructions: See function get_polaris_ksat() in landslide_precip_soilm_DataExport_final.ipynb on how to read these files from the url above.
   • Citation: https://doi.org/10.1029/2018WR022797
   • Format: tif files