Rnrfa: An R package to Retrieve, Filter and Visualize Data from the UK National River Flow Archive

Claudia Vitolo,Matthew Fry,Wouter Buytaert

doi:10.32614/rj-2016-036

Abstract

The UK National River Flow Archive (NRFA) stores several types of hydrological data and metadata: daily river flow and catchment rainfall time series, gauging station and catchment information. Data are served through the NRFA web services via experimental RESTful APIs. Obtaining NRFA data can be unwieldy due to complexities in handling HTTP GET requests and parsing responses in JSON and XML formats. The rnrfa package provides a set of functions to programmatically access, filter, and visualize NRFA data using simple R syntax. This paper describes the structure of the rnrfa package, including examples using the main functions gdf() and cmr() for flow and rainfall data, respectively. Visualization examples are also provided with a shiny web application and functions provided in the package. Although this package is regional specific, the general framework and structure could be applied to similar databases.

Highlights

The increasing volume of environmental data available online poses non-trivial challenges for efficient storage, access and share of this information (Vitolo et al, 2015)
The package provides functions to query the catalogue based on various criteria, retrieve and visualise flow and rainfall time series, convert coordinates and flow measurements, and plot basic seasonal trends grouped on user defined regions
Some of these capabilities are strongly linked to the particular content of the National River Flow Archive (NRFA) database and are not directly transferable/applicable to other data sources

Summary

Introduction

The increasing volume of environmental data available online poses non-trivial challenges for efficient storage, access and share of this information (Vitolo et al, 2015). The NRFA is a primary source of information for hydrologists, modellers, researchers and practitioners operating on UK catchments It stores several types of hydrological data and metadata: gauged daily flow and catchment mean rainfall time series as well as gauging station and catchment information. This section contains a gml-based identifier (station identification number) and additional information, such as ObservationMetadata (contact info, identification info, etc.), phenomenonTime (beginning and end of recordings), ObservationProcess (process type and reference). In order to improve access to NRFA’s public data and metadata, we implemented a set of functions to assemble HTTP GET requests and parse XML/JSON responses from/to the catalogue and WaterML2 services using simple R syntax. UK hydrometric area identification number, the related map is based on the Surface Water Survey designed in the 1930s and available at http://www.ceh.ac.uk/ data/nrfa/hydrometry/has.html. Feedbacks and contributions can be submitted through the GitHub issue tracking system (https://github.com/cvitolo/rnrfa/issues) and pull requests (https://github.com/cvitolo/rnrfa/pulls), respectively

Design and implementation

UKC 2 UKD 3 UKE 4 UKF 5 UKG 6 UKH 7 UKI 8 UKJ 9 UKK UKL UKM UKN

Summary