Aim: The Cape Fold Belt Mountains, underlying the Cape fynbos flora, facilitate widespread moisture collection and groundwater availability across the region, with importance for maintaining hydric habitat niches. We assessed the contribution of hydric habitat associated species (HH species) to the richness of this flora, and how this varies phylogenetically and spatially.

Location: The Cape Floristic Region (CFR), South Africa.

Methods: We compiled data describing habitat preference, and distributions for 3114 species in 23 fynbos clades. We used published habitat descriptions to identify putative HH species and tested how hydric habitat association is structured phylogenetically using both Ornstein-Uhlenbeck models and measures of phylogenetic signal. We used species’ distribution data to identify regional hotspots of HH species and boosted regression trees to identify associated environmental drivers.

Results: Preference for hydric habitats is associated with nearly a fifth of the species in our study, being most strongly represented in monocot clades. HH species are most strongly concentrated in the southwestern CFR, where they contribute to high overall richness. Boosted regression trees (BRTs) indicated that hotspots of HH species are generally associated with saturating, winter precipitation and/or elevated levels of groundwater discharge.

Main conclusions: Our findings indicate that the unique climate, topography and hydrology of the CFR facilitate the formation and maintenance of important hydric habitats, which play host to a large fraction of the flora despite much of the region being semi- or seasonally arid. These findings highlight the threat of climate change and underscore the need for extreme caution regarding activities like groundwater abstraction, which could disrupt the hydrological processes essential to sustaining these habitats and their unique flora.

https://doi.org/10.5061/dryad.kprr4xhcw

Provided files:

1) [Species_location_habitat_data.csv]

Our primary dataset consists of species location data paired with habitat preferences for 3114 species in 23 Cape fynbos clades. These data were used to quantify total species richness and hydric habitat associated species richness. The dataset contains columns, 'Genus' [e.g., 'Acmadenia'], 'Species' [e.g.,'patentifolia'], 'Decimal Latitude' [e.g., '-32.375'] , 'Decimal Longitude' [e.g., '19.375'], 'Clade' [e.g. 'Diosmeae'], 'Microhabitat' [e.g., 'marsh'...] and 'Hydric habitat associated' [e.g., 'Yes']

Habitat preference data was primarily sourced from (Manning and Goldblatt 2012) and more recent taxonomic updates. See manuscript for full list of references.
Species location data was sourced from (South African National Biodiversity Institute. (2016). Botanical Database of Southern Africa (BODATSA). https://posa.sanbi.org/)

2) [QDS_grid.shp]

Species richness and hydric habitat associated species richness was summarized across a quarter degree square grid for South Africa prior to analysis.

3) [Elevation_250m.tiff]

Derived from a digital elevation model sourced from (https ://earth explorer.usgs.gov; accessed May 2019). Spatial layers containing climatic, topographic and hydrological information were summarized at the QDS resolution and were used in boosted regression tree models to explain the spatial distribution of hydric habitat associated species.

4) [TRI_250m.tiff]

Terrain Ruggedness Index derived from the digital elevation model. Spatial layers containing climatic, topographic and hydrological information were summarized at the QDS resolution and were used in boosted regression tree models to explain the spatial distribution of hydric habitat associated species.

Other files required:

5) South African Vegetation Map.

Used to delineate the study area. Sourced from (South African National Biodiversity Institute (2006-2018). The Vegetation Map of South Africa, Lesotho and Swaziland, Mucina, L., Rutherford, M.C. and Powrie, L.W. (Editors), Online, http://bgis.sanbi.org/Projects/Detail/186, Version 2018.)

6) Angiosperm phylogeny.

Sourced from (Zanne, A. E., Tank, D. C., Cornwell, W. K., Eastman, J. M., Smith, S. A., FitzJohn, R. G., McGlinn, D. J., O’Meara, B. C., Moles, A. T., Reich, P. B., Royer, D. L., Soltis, D. E., Stevens, P. F., Westoby, M., Wright, I. J., Aarssen, L., Bertin, R. I., Calaminus, A., Govaerts, R., … Beaulieu, J. M. (2014). Three keys to the radiation of angiosperms into freezing environments. Nature, 506(7486), 89–92.)

The phylogeny was pruned to include a single representative from each focal clade. We used this pruned phylogeny to analyze the phylogenetic structure of hydric habitat association across clades.

7) Other spatial environmental layers used in our boosted regression tree models.

Bio1 (Mean Annual Temp - degrees Celsius), Bio10 (Mean Temp Warmest Quarter - degrees Celsius), Bio11 (Mean Temp Coldest Quarter - degrees Celsius), Bio12 (Mean Annual Precipitation - mm), Bio18 (Precipitation Warmest Quarter - mm), Bio19 (Precipitation Coldest Quarter - mm) sourced from (Fick, S.E. and R.J. Hijmans. (2017). Worldclim 2: New 1-km spatial resolution climate surfaces for global land areas. International Journal of Climatology. https://www.worldclim.org/).
Groundwater contribution to baseflow (m^3 km^-2) sourced from (Groundwater Resource Assessment II, Task 2C Groundwater Planning Potential Report (Project number 2003-150, 1 Final version 2006-02-24.)

Spatial layers containing climatic, topographic and hydrological information were summarized at the QDS resolution and were used in boosted regression tree models to explain the spatial distribution of hydric habitat associated species.

Data from: The contribution of hydric habitats to the richness of the Cape fynbos flora

Data files

Abstract

README: Data supporting the study: "The contribution of groundwater to the diversity of the Cape fynbos flora."