Quantifying the sampling error in tree census measurements by volunteers and its effect on carbon stock estimates

Butt, Nathalie; Slade, Eleanor; Thompson, Jill ORCID:; Malhi, Yadvinder; Riutta, Terhi. 2013 Quantifying the sampling error in tree census measurements by volunteers and its effect on carbon stock estimates. Ecological Applications, 23 (4). 936-943.

Before downloading, please read NORA policies.
Butt 2013 - Samp err tree cen=11-2059.pdf - Published Version

Download (281kB) | Preview


A typical way to quantify aboveground carbon in forests is to measure tree diameters and use species-specific allometric equations to estimate biomass and carbon stocks. Using "citizen scientists'' to collect data that are usually time-consuming and labor-intensive can play a valuable role in ecological research. However, data validation, such as establishing the sampling error in volunteer measurements, is a crucial, but little studied, part of utilizing citizen science data. The aims of this study were to (1) evaluate the quality of tree diameter and height measurements carried out by volunteers compared to expert scientists and (2) estimate how sensitive carbon stock estimates are to these measurement sampling errors. Using all diameter data measured with a diameter tape, the volunteer mean sampling error (difference between repeated measurements of the same stem) was 9.9 mm, and the expert sampling error was 1.8 mm. Excluding those sampling errors >1 cm, the mean sampling errors were 2.3 mm (volunteers) and 1.4 mm (experts) (this excluded 14% [volunteer] and 3% [expert] of the data). The sampling error in diameter measurements had a small effect on the biomass estimates of the plots: a volunteer (expert) diameter sampling error of 2.3 mm (1.4 mm) translated into 1.7% (0.9%) change in the biomass estimates calculated from species-specific allometric equations based upon diameter. Height sampling error had a dependent relationship with tree height. Including height measurements in biomass calculations compounded the sampling error markedly; the impact of volunteer sampling error on biomass estimates was +/- 15%, and the expert range was +/- 9%. Using dendrometer bands, used to measure growth rates, we calculated that the volunteer (vs. expert) sampling error was 0.6 mm (vs. 0.3 mm), which is equivalent to a difference in carbon storage of +/- 0.011 kg C/yr (vs. +/- 0.002 kg C/yr) per stem. Using a citizen science model for monitoring carbon stocks not only has benefits in educating and engaging the public in science, but as demonstrated here, can also provide accurate estimates of biomass or forest carbon stocks.

Item Type: Publication - Article
Digital Object Identifier (DOI):
UKCEH and CEH Sections/Science Areas: Watt
ISSN: 1051-0761
Additional Keywords: biomass estimate, carbon stocks, citizen science, data quality, forest monitoring, tree measurements; Wytham Woods, Oxfordshire, United Kingdom
Date made live: 13 Mar 2014 11:39 +0 (UTC)

Actions (login required)

View Item View Item

Document Downloads

Downloads for past 30 days

Downloads per month over past year

More statistics for this item...