Temporal responses over 30 years to removal of grazing from a mid‐altitude snow tussock grassland reserve, Lammerlaw Ecological Region, New Zealand

Abstract

Monitoring of five representative sites in the 144‐ha Black Rock Scientific Reserve of mid‐altitude (690–770 m) narrow‐leaved snow tussock (Chionochloa rigida) grassland over the 30 years since its establishment has revealed, contrary to an early prediction, significant increases in both cover and height of snow tussock. By contrast, co‐dominant shrubs have shown only a slight, generally nonsignificant gain, with Dracophyllum longifolium rather than the predicted Hebe odora as the only significant increaser. Several sub‐dominant shrubs (Coprosma cheesemanii, Leucopogon colensoi, Gaultheria macro stigma) plus some mosses (Hypnum cupressiforme) and lichens (Cladia retipora, Stereocaulon ramulosum) have increased significantly while some rosette herbs (Brachyglottis bellidioides, Oreomyrrhis colensoi, Plantago novae‐zelandiae, and the adventive Hypochoeris radicata) have declined. The generally aggressive exotic flatweed Hieracium pilosella remains as yet a minor component. These changes in subcanopy cover probably reflect the obvious increase in shade and dampness of the micro‐habitat. The height‐frequency sampling indicates an overall decline in vascular species diversity since losses have significantly exceeded gains over the 30‐year period of monitoring. Our results confirm that low‐ to mid‐altitude snow tussock grassland ecosystems can be sustained for at least several decades, for their conservation, landscape, and water yield values. We question the interpretation of a general lack of tussock grassland below treeline in immediate pre‐human times, and its widespread downslope replacement of forest following Polynesian fires, since it is at variance with the known ecology of the dominant grass species, evidence from relevant pollen records, and results from the present study. Rather, we interpret the available evidence as indicative of succession to a vegetation mosaic of non‐woody and woody dominants related to physiography and disturbance, as currently being debated for north‐western Europe. We hypothesise that such a mosaic would more closely reflect the pre‐human situation below treeline which would have been moulded by periodic fire and avian and invertebrate herbivory, in the absence of land mammals.