Rely on measurement protocols and know your units
(based on Rule 3 of the article doi:10.1111/2041-210X.14033)
To ensure comparability, future data reuse and synthesis, relate primary measurements of your traits to the wider body of published trait data. Conform your measurement procedures to existing trait measurement protocols, or – if no such standard protocols exist – document with precision and build upon unambiguous concepts. Also, confusion and errors in terms of recording and reporting of units can be propagated through large trait compilations. Thus, define your units clearly; they are essential for harmonising different trait data sets, approximations and uncertainties.
Beware of ambiguities:
In most cases, researchers of a domain (e.g. plants) have adopted sufficiently specific trait definitions to allow comparison of widely used measurements and enable synthesis within the field. However, some difficulties in measurement remain. To illustrate, specific leaf area (SLA) is the ratio of the surface area to leaf biomass of an individual leaf. However, the application of the concept of SLA may differ between research contexts, because the value reported may relate to measurements of individual leaves or an average of all leaves on the shoot, for one or both sides of the leaf, including or excluding the petiole, and focus on the leaf or leaflet (example taken from Garnier et al., 2017). While fully justified in the specific research setting, identifying and dealing with semantic disambiguation is a major challenge in trait-based synthesis.
Adhere to existing standards:
Methodological handbooks for trait measurements have been proposed, e.g. for plant (Cornelissen et al., 2003; Pérez-Harguindeguy et al., 2013), macrofungi (Dawson et al., 2019) or terrestrial invertebrate functional traits (Moretti et al., 2017). These handbooks provide precise, domain-specific definitions and recommended methods for trait measurement, measurement precision and replication. They also provide considerations and warnings of misconception and error, and point to the key literature debating the methodology. Taking formalisation of trait concepts even one step further are thesauri of trait concepts (Garnier et al., 2016, 2017), e.g. the TOP thesaurus of plant characteristics (https://top-thesaurus.org/). The bottom line is: research that provides original trait measurements should consider existing measurement protocols, make an explicit choice, and describe any deviations from or additions to protocols. When such handbooks do not exist, it is good practice to report specific measurement protocols in the metadata (see Rule 4). For instance, how the length of a fish has been measured and if potential extensions of the tail fin were taken into account.
Understand your units:
Trait data are necessarily ‘rich in dimensions’. That is, trait data may require multiple SI (International System of Units) base units and may also be measured and reported in various alternative configurations of units. For example, photosynthetic rate involves three SI base units, e.g., mass per area per time, and is often reported in units of µmol CO2m-2s-1 an amount per unit area per unit time. Measures of size, area, and time are often reported in different units, though all can be related to more fundamental base units. All metric trait data can be reduced to the seven base units as defined by the SI standard (m, kg, s, K, A, cd, mole). Significant data management effort is needed to record units accurately, preserve them through metadata, and convert them correctly to avoid propagating errors (Calder, 1982).