Determining drivers of dragonfly diversity patterns and the implications for conservation in

South Africa

Summary
The article entitled “Determining drivers of dragonfly diversity patterns and the implications for
conservation in South Africa” was published on April 13, 2020, by Charl Deacon, Michael J.
Samways, James S. Pryke in the biological conservation journal. With the use of generalized linear
mixed models and generalized dissimilarity models— the researchers have determined the climatic
and spatial factors driving regional trends in overall dragonfly species richness, local endemism, and
assemblage turnover, identified assemblage turnover boundaries across the coastal and interior
regions of the country, and provided recommendations for the conservation of local freshwater
insects.

In South Africa, dragonflies and damselflies (Odonata) are taxonomically well-known, and their
distribution is well established (Samways and Simaika, 2016). With that being said, dragonflies are
excellent model organisms for detecting broad-scale biodiversity patterns and identifying localities of
conservation concern across South Africa (Simaika and Samways, 2009). To determine the species
richness, diversity, and assemblage of dragonflies and damselflies— regional factors and local
factors are considered as important drivers of diversity patterns. Moreover, Deacon et al. (2020) have
selected eight important variables in driving dragonfly species richness metrics across South Africa,
namely: latitude, longitude, elevation, average spring/summer and autumn/winter rainfall, average
spring/summer and autumn/winter solar radiation, and yearly average temperature. The results are
shown in the table and figure below.
The researchers found out that from south to north and west to east, the species richness increased,
and was highest in the north-eastern region (Table 1, Figure 1a)(Deacon et al., 2020). Results has
shown that endemic species richness was concentrated in the south-western region, and decreased
from south to north and west to east (Figure 1b). Moreover, red-Listed species richness decreased
from south to north and was lowest at high elevation, coinciding with areas of high species richness
and endemism (Figure 1c). The proportion of lentic species increased from west to east overall (Table
1, Figure 1d). The proportion of lotic species decreased from west to east, and higher proportions of
lotic species richness was associated with areas with lower autumn/winter rainfall (Table 1, Figure
1e). The proportion of lentic/lotic species richness was lowest in areas with low average
autumn/winter rainfall (Table 1, Figure 1f). Overall, lentic species were well-represented by the
existing conservation network, yet mountainous localities with proportionally higher lotic species
richness were found outside of formally protected areas along the east coast (Figure 1d and e). A
high proportion of lentic/lotic species richness was also outside formally protected areas throughout
South Africa (Figure 1f).

Deacon et al. (2020) used backward selection in identifying four essential environmental variables in
driving dragonfly assemblage-turnover at the national scale, namely: average autumn/winter solar
radiation, average soil drain rate, elevation, and geographical distance between cells. The results are
shown in the figure below.

The researchers found out that the dragonfly assemblage-turnover rate increased rapidly with
increasing soil drain rate until it reached 6 mm/day, above which assemblage-turnover rate slowed
(Figure 3a)(Deacon et al., 2020). Throughout South Africa, the effects of soil drain rate on
assemblage-turnover were relatively even but lower in the north-western coastal region, and north-
eastern mountainous region (Figure 3b). Assemblage-turnover rate remained constant with
increasing autumn/winter solar radiation until 23.2 MJ/m2/day was reached, above which
assemblage-turnover rate increased rapidly (Figure 3c). The effect of solar radiation on dragonfly
assemblage-turnover was highest in the arid, north-western regions (Figure 3d). Assemblage-
turnover rate increased steadily between 0 m a.s.l. and 250 m a.s.l., and there was a rapid increase
in assemblage-turnover rate between 250 m a.s.l. and 3000 m a.s.l (Figure 3e). Moreover, the effects
of elevation were highest in areas of sharp topographical gradients, specifically in the areas
surrounding the Drakensberg Mountains and the Cape Fold Mountains. Elevation also had an effect
on the assemblage-turnover rate on the High-veld (high-elevation grassland region) and parts of the
Northern Cape. (Figure 3f). Dragonfly assemblage-turnover rate increased rapidly throughout for
geographical distance, but slowed above 500 km and slowed further above 1500 km (Figure 3g).

Across South Africa, generalized dissimilarity models performed well in extrapolating assemblages
(Figure 3h), and landscape visualizations produced for observed and predicted assemblage turnover
showed similar spatial patterns (Figure 4a and b). Both approaches showed an overall difference in
species assemblage composition across the latitude gradient and across the longitude gradient.
Hence, assemblage composition varies substantially across the region— species assemblage
composition was different between the north-eastern and south-eastern regions, and between the
north-western and south-western regions (Deacon et al., 2020).

Dragonfly species diversity patterns were known to be complex across Africa's southernmost

point, yet the results of the study indicated strong latitudinal and longitudinal gradients in species
richness, endemism, and assemblage composition across the region. Furthermore, Deacon et al.
(2020) stated that other regional factors related to climate (i.e., average seasonal rainfall, average
seasonal solar radiation, and average temperature), underlying geology (i.e., soil type and drain rate),
and topography (i.e., elevation gradients and geographical distance between cells) significantly
determined overall dragonfly species diversity patterns. Dragonfly assemblage-turnover boundaries
were gradual at the regional scale, and most of these boundaries are determined by topography and
spatial climatic variation. 

Dragonflies are frequently recognized in global conservation efforts and are a valuable umbrella taxon
for other aquatic insects (Bried et al., 2007; Samways et al., 2011; Kietzka et al., 2019). The existing
national conservation network in South Africa represents areas of high dragonfly species richness,
endemism, and red-listed species richness well. With that, the researchers recommend that in areas
outside of protected areas with high species richness and endemism levels, and areas with high
assemblage-turnover rates— additional conservation efforts should be exerted to help protect many
other species as well. Finally, the researchers encouraged other experts to conduct further field
investigations searches in under-represented areas to improve distribution data for known species
and propose more researches in areas with high endemism and high assemblage-turnover for
possible discovery of unknown species (Deacon et al., 2020).
References

Bried, J.T., Herman, B.D., Ervin, G.N., 2007. Umbrella potential of plants and dragonflies for wetland
conservation: a quantitative case study using the umbrella index. J. Appl. Ecol. 44, 833-842

Deacon, C., Samways, M., & Pryke, J. (2020). Determining drivers of dragonfly diversity patterns and
the implications for conservation in South Africa. Biological Conservation, 245.
doi:https://doi.org/10.1016/j.biocon.2020.108548

Kietzka, G.J., Pryke, J.S., Gaigher, R., Samways, M.J., 2019. Applying the umbrella index across
aquatic taxon sets for freshwater assessment. Ecol. Indic. 107, 105655. https://
doi.org/10.1016/j.ecolind.2019.105655.

Samways, M.J., Simaika, J.P., 2016. Manual of Freshwater Assessment for South Africa: Dragonfly
Biotic Index. Suricata 2. South African National Biodiversity Institute, Pretoria, South Africa.

Samways, M.J., Sharratt, N.J., Simaika, J.P., 2011. Effect of alien riparian vegetation and its removal
on highly endemic macroinvertebrate community. Biol. Invasions 13, 1305-1324

Simaika, J.P., Samways, M.J., 2009. Reserve selection using red listed taxa in three biodiversity
hotspots: dragonflies in South Africa. Biol. Conserv. 142, 638-651