Specific leaf area is lower on ultramafic than on neighbouring non-ultramafic soils

ABSTRACT

Background

Specific leaf area (SLA) is a core trait within the leaf economic spectrum that describes differences in plant performance and productivity. Research on the sources of variation in the leaf economic spectrum and SLA has primarily focused on climate. Much less is known about SLA variation across unusual edaphic environments, such as on ultramafic soils.

Aims

To determine the role of ultramafic soils as a driver of SLA variation.

Methods

We measured SLA for dominant species on paired ultramafic and non-ultramafic soils in five biogeographically distinct regions around the globe and compared mean SLA values to globally reported values.

Results

SLA was lower on ultramafic than on non-ultramafic soils in all regions, except Puerto Rico, and both climate and soil were important drivers of SLA. For three of the five regions, SLA values on ultramafic soils were lower than the global average.

Conclusions

Soils can be a major driver of SLA along with climate. Low SLA on ultramafic soil points to selection for stress resistance strategies. Furthermore, in some bioregions, SLA values on ultramafic soils were among the lowest on the planet and thus represent globally rare phenotypes that should be conserved within these unique edaphic habitats.

KEYWORDS:

• Edaphic
• extreme environments
• functional trait
• leaf economic spectrum
• serpentine syndrome
• stress resistance syndrome
• specific leaf area (SLA)

Acknowledgments

We thank four anonymous reviewers for their constructive feedback. We are grateful to Ulric J. Lund and John H. Walker at California Polytechnic State University for their help with statistical analyses in R.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Data used in this study is, in part, available through TRY (Kattge et al. 2020), BIEN (https://bien.nceas.ucsb.edu/bien/), and Dryad (Hulshof et al. 2021; Spasojevic and Harrison 2022). All code and raw data files are available on the Dryad Digital Repository at https://doi.org/10.5061/dryad.573n5tbbr.

Supplementary Information

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17550874.2022.2160673

Additional information

Funding

This material is based upon work supported by the National Science Foundation under Grant No. NSF MSB-ECA #1833358 and NSF CAREER #2042453. The authors thank Bill and Linda Frost for funding through the Frost Summer Undergraduate Research Program at California Polytechnic State University. GCA received funding from the programme MEDICUS, of the University of Patras. NR gratefully acknowledges funding from the Fulbright US Scholar Program (South Africa).