Data for: Revealing hidden sources of uncertainty in biodiversity trend assessments

Idiosyncratic decisions during the biodiversity trend assessment process may limit reproducibility, whilst “hidden” uncertainty due to collection bias, taxonomic incompleteness, and variable taxonomic resolution may limit the reliability of reported trends. We model alternative decisions made during assessment of taxon-level abundance and distribution trends using an 18-year time series covering freshwater fish, invertebrates, and primary producers in England. Through three case studies, we test for collection bias and quantify uncertainty stemming from data preparation and model specification decisions, assess the risk of conflating trends for individual species when aggregating data to higher taxonomic ranks, and evaluate the potential uncertainty stemming from taxonomic incompleteness. Choice of optimizer algorithm and data filtering to obtain more complete time series explained 52.5% of the variation in trend estimates, obscuring the signal from taxon-specific trends. The use of Pen..., , , ## Data for: Revealing hidden dources of uncertainty in biodiversity trend assessments

[https://doi.org/10.5061/dryad.np5hqc034](https://doi.org/10.5061/dryad.np5hqc034)

## Description of the data and file structure

Raw data to run analyses presented in the paper. Comprised of csvs containing count or detection-nondetection data, survey level metadata, and taxonomic information for four taxonomic groups (diat, fish, invs, macp); a csv of model arguments passed to generalized linear models; and shapefiles and rasters of spatial environmental datasets.

### Files and variables

#### File: home.zip

**Description:** 

**home/mod_args.csv** is tabular CSV data containaing information on the task id used for grid job submission. It contains the following variables:

* *task.id*: The unique task identifier used for grid job submission
* *group*: The taxonomic group (abbreviated as Diatoms=diat, Fish=fish, Invertebrates=invs, Macrophytes=macp)
* *full*: A binary flag (TRUE if the full trend...,