STRAVA_starter

A set of scripts to do some basic analysis with STRAVA data in Norway

Introduction

The aim of this repository is to provide reproducible scripts that allow NINA colleagues to do a basic analysis of the STRAVA data and its interaction with environmental data. Please contact Zander Venter (zander.venter@nina.no) if you need more details or you spot any issues with the code.

Outputs

A sorted and cleaned dataframe of STRAVA activities for your study area
A map of STRAVA activity intensity per trail segment over your study area
Plots of STRAVA activities aggregated over temporal (monthly) and spatial (ecosystem types) categories
Plots of STRVA activities against various environmental covariates

Scope of the scripts and analysis

NB!! - this is focussed on the "legacy" STRAVA dataset (anything before March 2020) which is stored on NINA R:\ server
- For data after that point, you need to download from the STRAVA Metro dashboard (https://metroview.strava.com/) and analyse differently
- The dashbaord solution has different column names, and temporal aggregation procedures, and is therefore difficult to harmonize with legacy data. It also includes sex, age etc.
Focuses on the monthly STRAVA rollups for 2019 (i.e. excludes annual rollups, hourly data, or origin-destination data)
Focuses on total activity for pedestrian and cycling categories (i.e. excludes athlete counts, leisure vs commute, activity times, revers vs forward direction)
Focuses on temporal aggregation (i.e. does not explore time series beyond 1 year long)
Focuses on relative activity over time and space (i.e. does not calibrate against counter station data to get absolute activity counts)

Prerequisites

You need to have access to the RStudio server: http://ninrstudio04.nina.no/
You need to be a registered Google Earth Engine (GEE) user: https://earthengine.google.com/
Ideally, to make maps in R, you need to have a Google Maps API key: https://developers.google.com/maps/documentation/javascript/get-api-key

Workflow

Download this entire repository and open up the "STRAVA_R_starter.Rproj" file from RStudio server
Create a shapefile outlining your area of interest (AOI) and upload to './DATA/' folder *I have placed a Shapefile to start with called "Oslo.shp" which you can use to test-run the scripts

Run the "Basic_analysis.R" script line-by-line, changing variable names or path directories to your specific work where necessary.

Upload the shapefile that was generated in './DATA/For_GEE/' to your GEE Asset

Run the "Environ_data_extract.js" in GEE JavaScript API, changing path directories to your AOI asset. You can copy and paste the script contents into the editor, or follow this code snapshot link: https://code.earthengine.google.com/32ca72b58ddff06c337769df9e5417f8

Run the export tasks in the GEE Javascript API. Download them from your Google Drive. Copy them to the './DATA/From_GEE/' folder