Patterns of avian diversity across a decreasing patch‐size gradient in a critically endangered subtropical forest system

Abstract

Aim:We investigated habitat heterogeneity and patterns of avian taxonomic and functional diversity change across a decreasing patch-size gradient in a critically endangered, fragmented forest-system to elucidate: (1) habitat patch-size and structural drivers of avian diversity change, (2) potential patch-size thresholds at which avian diversity and habitat heterogeneity changes occurred and (3) avian species/ communities within thresholds that indicated distinct patch-size categories.Location: Indian Ocean Coastal Belt, South Africa. Methods:We conduced habitat-structure and fixed-radius point-count surveys during the avian breeding season across 123 habitat patches (range 0.06-386.9 ha).We calculated taxonomic and functional diversity/patch (a) using a distance matrix of all traits present in the avian community (c) and described community change across the patch-size gradient (b). We compared linear regressions of patch-size effects on avian functional a-diversity and habitat-structural heterogeneity to segmented regression at five bands, determined by mean patch sizes responsible for provisioning avian functional a-diversity at <50%, 50%-62.5%, 62.5%-75%, 75%-87.5% and 87.5%-100% of avian functional c-diversity to explore patch-size thresholds, and tested for significant avian indicator species of categories identified.Results: Avian taxonomic and functional a-diversity significantly decreased with decreasing patch size and habitat heterogeneity; correlations between taxonomic and functional diversity declines and a positive significant influence of habitat heterogeneity on avian functional diversity indicated selective extinction pressure on niches and concomitant functional traits. Species turnover significantly increased in smaller patches. Segmented regressions of patch size-effect on avian functional diversity and habitat heterogeneity out-performed linear regressions at three patchsize thresholds (157, 96 and 9 ha). Forest-dependent specialists were significant indicators of high-diversity patches >90 ha. Species characterizing small patches <9 ha were generalists, but many were also present in larger patches, highlighting nested-community subsets of c-diversity.Main conclusions: Specialized niches and concomitant species became locally extinct at patch-size thresholds. We present replicable methodology and recommendations for conservation of forest systems using a critically endangered biome containing disproportionately high avifauna.

Predicted distributions of avian specialists: A framework for conservation of endangered forests under future climates

“…Locality data for Spotted Ground Thrushes were obtained from the following sources: standardized camera‐trap surveys (for survey methods, see Ehlers Smith, Ehlers Smith, & Downs, ; Ehlers Smith, Ehlers Smith, Ramesh, & Downs, ; Ehlers Smith, Ehlers Smith, Ramesh, & Downs, ) and standardized point‐count surveys (for survey methods, see Ehlers Smith, Si, Ehlers Smith, & Downs, ; Ehlers Smith, Si, Ehlers Smith, Kalle, et al, ) conducted in the wintering‐range habitat patches between Umtamvuna (31°04’46.69” S, 30°11’39.87” E) and Umkomasi Rivers (30°12’09.10” S, 30°48’12.10” E), southern KwaZulu‐Natal Province; capture localities during bird‐ringing surveys (SAFRING); habitat patches confirmed as presence localities from bird‐atlasing pentads (SABAP2; Underhill, Harebottle & Brooks, ) and georeferenced localities of birds recorded in bird‐atlasing using the BirdLasser application, a citizen science platform linked to SABAP2 (BirdLasser, Johannesburg). The resultant data set contained 89 presence localities.…”

“…Locality data for Orange Ground Thrushes were obtained from the following sources: standardized camera‐trap surveys (M. Sosibo, unpublished data, using the survey protocol of Ehlers Smith, Ehlers Smith, & Downs, et al, ; Ehlers Smith, Ehlers Smith, Ramesh, et al, ; Ehlers Smith, Ehlers Smith, Ramesh, & Downs, ) and standardized point‐count surveys (S. Ngcobo & T. Gumede, unpublished data, using the survey protocol of Ehlers Smith, Ehlers Smith, & Downs, ; Ehlers Smith, Si, Ehlers Smith, Kalle, et al, ; Ehlers Smith, Si, Ehlers Smith, & Downs, ) conducted from the Umtata Forest cluster of the east Eastern Cape (31°31′30.40″ S, 28°10′45.96″ E) and the Weza‐Ingeli Forest cluster of southern KwaZulu‐Natal (30°35′34.19″ S, 29°40′56.25″ E); capture localities during bird‐ringing surveys (SAFRING); habitat patches confirmed as presence localities from bird‐atlasing pentads (SABAP2); and georeferenced localities of birds recorded in bird‐atlasing using the BirdLasser application. The resultant data set contained 57 presence localities.…”

“…Locality data for Bush Blackcap were obtained from the following sources: standardized point‐count surveys (R. Colyn, S. Ngcobo & T. Gumede, unpublished data, using the survey protocol of Ehlers Smith Ehlers Smith, Downs, et al, ; Ehlers Smith, Si, Ehlers Smith, & Downs, ; Ehlers Smith, Si, Ehlers Smith, Kalle, et al, ) conducted along the eastern escarpment forest cluster straddling the provincial boundary between KwaZulu‐Natal and Free State Provinces (28°13′43.04″ S, 29°37′38.53″ E) to Poccolan Robinson's Bush Nature Reserve (28°34′34.05″ S, 29°02′48.54″ E); habitat patches confirmed as presence localities from bird‐atlasing pentads (SABAP2); and georeferenced localities of birds recorded in bird‐atlasing using the BirdLasser application. The resultant data set contained 94 presence localities.…”

“…Large, contiguous and intact forests that are well-connected are biologically more diverse and function at a higher level than smaller, degraded and isolated forest patches (Fahrig, 2003;Laurance, 2008;Murcia, 1995). This is attributed to a greater structural ecosystem diversity, more diverse resources and therefore more specialized species' niches (Ehlers Smith, Si, Ehlers Smith, Kalle, et al, 2018), improved dispersal and colonization opportunities affording healthy genetic transfer (Moilanen & Nieminen, 2002), and an increase in overall habitat availability (Fahrig, 2003).…”

“…exotic timber plantations, and selective logging of Afrocarpus and Podocarpus yellowwood trees in remnant native forest patches have significantly altered their extent, composition, structure and capacity for carbon sequestration (Adie, Rushworth, & Lawes, 2013;Symes & Woodborne, 2010). Much research has been conducted to quantify forest floral and faunal assemblages (e.g., Symes, Wirminghaus, & Downs, 2002;von Malitz et al, 2003) and to describe the effects of forest homogenization and reduction (e.g., Ehlers Smith, Si, Ehlers Smith, Kalle, et al, 2018). Few studies to date have focussed on the importance of forest connectivity for species' persistence (Ehlers Smith, Ehlers Smith & Downs, 2019); none have analysed the projected impacts of climate change on the forest biome, and the K E Y W O R D S core-habitat mapping, ecological circuit theory, ecological niche, ensemble modelling, global circulation models, habitat resistance, least-cost pathways, pinch-point mapping likely persistence or resilience of the forests under such scenarios.…”