Expanding an existing classification of New Zealand vegetation to include non-forested vegetation

Abstract

Abstract:We produced the first national-scale quantitative classification of non-forest vegetation types, including shrubland, based on vegetation plot data from the National Vegetation Survey Databank. Semisupervised clustering with the fuzzy classification algorithm Noise Clustering was used to incorporate these new data into a pre-existing quantitative classification of New Zealand's woody vegetation. Fuzzy classification allows plots to be designated as transitional when they are similar to multiple vegetation types; the Noise Clustering algorithm allows plots having unique composition to be designated as outliers. We combined plot data collected using two different methods by transforming abundances to relative ranks and showed our classification results were robust to this. Of the 6362 plots analysed, 505 were assigned to previously defined woody vegetation types. Using the remaining 5857 plots, we defined vegetation types at two hierarchical levels comprising 25 alliances and 56 associations. Ten of the alliances are tussocklands, six are grasslands, four are stonefields or gravelfields, two are herbfields, one is rushland, and two are newly defined woody alliances. The classification defined compositional differences among well-known widely distributed short and tall tussock grasslands of the South Island. Notably it distinguished Chionochloa pallens, C. crassiuscula and C. oreophila tussocklands in wetter western regions from those dominated by C. rigida and C. macra in the east, and the domination of eastern South Island short tussock grasslands by Festuca novae-zelandiae and Poa colensoi. We demonstrate the distinctiveness of the vegetation of four naturally uncommon ecosystems -coastal turfs, northern gumlands, granite sand plains and braided riverbeds. Insufficient plot data precluded the definition of North Island Chionochloa rubra grassland types and many wetland and coastal communities. The 1846 plots designated as outliers mainly occur on warmer, wetter and less invaded sites than classified plots. Semi-supervised clustering allowed us to progress the development of an extendable, plot-based, quantitative classification of all New Zealand's vegetation despite data gaps.

http://newzealandecology.org/nzje/3276

“…While these were national scale classifications, relevant comparisons can still be made with our classification because of the similarity in the sampling approach. Several of the vegetation types identified in our Cass classification are similar to those identified by Wiser et al (2011Wiser et al ( , 2016, which they term 'alliances'. Our mountain beech forest type and mānuka shrubland type align well with Wiser et al (2011) Alliance 5 and 8 respectively.…”

“…The comprehensive grid-based sampling system used in this study enabled us to objectively describe and classify the vegetation and its relative abundance in the landscape. Such an approach allows for an unbiased assessment of what drives vegetation patterns without being constrained by preconceptions of the major underlying environmental gradients (Allen et al 2003;Wiser et al 2011Wiser et al , 2016. Despite the objectivity of our sampling approach, the ability to sample all of the floristic variation present will always be compromised by the scale of sampling.…”

“…Notwithstanding the under-sampling of rarer vegetation types, our grid-based sample still detected 80% of the vascular plant flora known from Cass (see Appendix S2 in Supplementary Material). Wiser et al (2011Wiser et al ( , 2016 used an objective, grid-based plot network to classify forest and shrubland communities, and non-forest communities based on vegetation compositional data. While these were national scale classifications, relevant comparisons can still be made with our classification because of the similarity in the sampling approach.…”

Mapping Physiognomic Types of Indigenous Forest using Space-Borne SAR, Optical Imagery and Air-borne LiDAR

“…Both the New Zealand Forest Class Maps (compiled at a scale of 1:250,000 from black and white aerial photographs, dating from 1948 to 1955; available online [6]) and the Vegetative Cover Map of New Zealand (compiled at a scale of 1:1,000,000 [7]) provide national coverage of broad forest types, but are now 50+ years out of date, and the scale is not fine enough for many desired applications. Wiser et al [8,9] used data from > 13,000 ground-based plots to classify New Zealand woody vegetation into forest types (at two levels of hierarchy termed alliances and associations). They then grouped these into more broadly defined physiognomic types as Beech, Beech-broadleaved, Beech-broadleaved-podocarp, Broadleaved-podocarp, Podocarp, and other forests.…”

Mapping Physiognomic Types of Indigenous Forest in New Zealand Using Space-Borne SAR, Optical Imagery and Air-Borne LiDAR

“…Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 16 July 2019 doi:10.20944/preprints201907.0191.v1 Table 1. Five-fold cross-validation accuracy (100 repeats) of predicted forest physiognomic types from support vector machine classification by leaving out one of the (i) Sentinel2 spectral bands (2,3,4,5,8,11,12), (ii) mean and 97 th percentile of canopy height model, and (iii) Sentinel1 VH/VV and PALSAR HH bands. Grey shade shows which bands were used.…”