UK-SCAPE Virtual Survey Lab

Science challenge

The ability to efficiently access and analyse datasets from many sources is essential to gain an holistic understanding of environmental processes and change. The Virtual Survey Lab supports environmental research by bringing together data collected both within the UK-SCAPE programme and by external partners.

Objectives

To develop the components of a Data Science Framework (DSF) required to produce integrated data products utilising the data collected within the UK-SCAPE program and from external sources.
To develop statistical routines to analyse data across different spatial and temporal scales and from different sources to produce a series of highly resolved spatio-temporal data products.
To make these products discoverable and accessible to the wider scientific community and prioritised to address questions and fill gaps in other areas of the UK-SCAPE programme.
To test the efficiency of ongoing monitoring and, in the context of answering specific questions, highlight where we should direct monitoring campaigns and prioritise effort.
To provide a platform for the integration of monitoring data provided by a range of partners from across terrestrial, atmospheric and freshwater habitats.

Methodological developments

Survey design

Power analysis

Users can simulate data under different survey designs and expected change scenarios and then use this to calculate the expected power (probability of detecting an effect) of the specified survey design in detecting the specified change.

App: UK-SCAPE Power Analysis Tool
Paper:
- Boyd, R J, Powney, G D, Burns, F, Danet, A, Duchenne, F, Grainger, M J, Jarvis, S G, Martin, G, Nilsen, E B, Porcher, E, Stewart, G B, Wilson, O J, Pescott, O L. 2022. ROBITT : A tool for assessing the risk‐of‐bias in studies of temporal trends in ecology. Methods Ecol. Evol. 13, 1497–1507.

Data quality assurance (QA)

State tagging

State tagging is an unsupervised machine learning method to improve contextual understanding of environmental time series by segmenting it into arbitrary states, which can be used as a tool to QA variables whose typical range changes with environmental conditions.

Code: State tagging application for environmental data quality assurance
App: System state tagging (generic version)
Paper:
- Tso, C-H M, Henrys, P, Rennie, S, Watkins, J. 2020. State Tagging for Improved Earth and Environmental Data Quality Assurance. Front. Environ. Sci. 8.

Analysis methods

Spatial upscaling

Point data to maps

The EcoMAPS R library is a spatial statistical model to upscale point data to GB-wide mapped datasets using detailed covariate information.

Code: Spatial-Upscaling-model
Shiny App: EcoMAPS - GB Mapping of Environmental Indicators
Example output (EIDC dataset): Topsoil pH estimates from the Countryside Survey of Great Britain, 2007 using a generalized additive model

Integration

Joint species distribution modelling

This is an R package for running joint Species Distribution Models (jSDM) in Stan.

Code: jsdmstan: joint Species Distribution Models in Stan
Data: vegtaxon

Spatial integration

This work provides a detailed analysis and review of data integration, especially for species distribution models. The app allows users to compare two hypothetical sampling design as well as integrating them to show the benefits or disbenefits of each scheme.

Code: IOFFsimwork
Papers:
- Isaac, N J B, Jarzyna, M A, Keil, P, Dambly, L I, Boersch-Supan, P H, Browning, E, Freeman, S N, Golding, N, Guillera-Arroita, G, Henrys, P A, Jarvis, S, Lahoz-Monfort, J, Pagel, J, Pescott, O L, Schmucki, R, Simmonds, E G, O’Hara, R B. 2020. Data Integration for Large-Scale Models of Species Distributions. Trends Ecol. Evol. 35, 56–67.
- Simmonds, E G, Jarvis, S G, Henrys, P A, Isaac, N J B, O’Hara, R B. 2020. Is more data always better? A simulation study of benefits and limitations of integrated distribution models. Ecography (Cop.). 43, 1413–1422.
- Jarvis, S G, Mackay, E B, Risser, H A, Feuchtmayr, H, Fry, M, Isaac, N J B, Thackeray, S J, Henrys, P A. 2023. Integrating freshwater biodiversity data sources: Key challenges and opportunities. Freshw. Biol. 68, 1479–1488.
App: NCEA Data Integration Exploration Tool
Example output (EIDC dataset): Gridded estimates of Ellenberg N, R, F, and L indicators for Great Britain, 1990 and 2015–2019

Spatial-temporal integration

This exemplar provides examples of spatiotemporal integration of multiple data sources to predict insect populations using R-INLA.

Code and paper draft: Spatio-Temporal Data Integration in R-INLA Paper
App: gaussian_process_kernel_app

Process model – Data integration

This exemplar demonstrates how uncertainty in model discrepancy in UK maps with a spatial dependence can be modelled using a nearest-neighbour Gaussian Process methodology.