##############################################################################
This dataset contains QGIS project files and associated files such as photos and output tifs, csv files of measured grain diameters, river flow analysis and outputs of cluster analysis and txt files with short sections of code used in Google Earth Engine and Jupyter Notebook
The data was used to produce figures and results for a manuscript (2024WR039351R) submitted to the AGU journal, 
Water Resources Research (https://agupubs.onlinelibrary.wiley.com/journal/19447973) in November 2024. The research is part of a PhD project. 

The following files are included in this dataset.

1. Read_me.txt 
This document. 

2. GEE-code-binary-masks.txt
Text file of Google Earth Engine code to extract binary mask of wetted channel or active channel from compiled satellite images. 

3. Satellite_image_stats.csv. 
For each year of the study, this file has the low-flow months used, the average recorded daily flow (in m3/s) for the period, the number of satellite images available for those months and the number of images used after cloud filtering. Average momthly flows were taken from summary tables provided by the Department of Hydrology & Meteorology, Nepal. 

4. Binary_Masks_WRR.zip 
Zip file contains a QGIS project file (East and West Masks.qgz) used to generate east and west masks from the binary masks created in GEE. For each year there is a wetted channel tif, an east tif and a west tif. There are also four combined images showing 10 years of wetted channel positions. 

5. Monsoon_flow_analysis.csv
Output of the counted cells for the east and west binary masks for each year of the study, for wetted and active channels. This is 
used to find the percentage of cells in the west and east each year, and percentage change in wetted area from previous year. 
Lengths of bifuracation movements, maximum daily average flow, and total monsoon flow for each year of the study are also recored 
here. See also the notes below the table within the file. River flow data provided by the Department of Hydrology & Meteorology, Nepal. 

 
6. Karnali_grainsizes.csv
Grain size data for each survey site collected from 28/02/2024 to 04/03/2024. Grain sizes include those measured on site and from photo sieving. 
Grain size data for two sites (C9 and C24) collected by other researchers in 2016 included for comparison

7. Grainsize_locations.csv
Latitude, longitude and description for each fieldwork site with caculcated D50 and D85. 

8. Clusters_WRR.zip. 
Zip file contains a QGIS project file ("DBSCAN-clustering.qgz") used to find DBSCAN values for fieldwork sites.  
For each fieldwork location there is a merged photo (jpg) and sidecar file (jpgw).
At each site three sets of grains have been analysed using DBSCAN: a) from D50 to D85, b) all grains over D50, c) all grains over D85. 
DBSCAN was run using an internal function in QGIS for clusters with radius 100mm and 150mm.  

9. Python-Notebook-C12 Geostats Ripley.ipynb 
Python notebook code example used to find Ripley's values for grain clusters. 

10. Cluster_results_DBSCAN.csv
Summary of outputs of DBSCAN analysis. For each fieldwork site the size of the merged photo, number of clusters (at two radii) and 
clusters per square metre are collated.
 
11. Cluster_results_Ripley.csv
Summary of outputs of Ripley's analysis. The maximum H(r) (cross-Ripley) values are recorded for each fieldwork location and the radius
at which the max H(r) occured.