Abstract: The deliberate use of fire in the pastoral management of tall-tussock (Chionochloa) grasslands has long been a contentious issue. Environmental advocates portray fire as damaging native biota, increasing opportunities for weed invasion and promoting soil erosion, while pastoralists see it as a means of improving access for stock, promoting palatable regrowth, and removing ‘woody weeds’. This paper describes the development of a study that aims to provide answers to three questions central to this debate, (i) Does fire cause long-term damage to the physical and biological resources of tall-tussock grasslands, (ii) Are accidental summer fires more damaging than prescribed burns in late winter or early spring and (iii) Are fuel availability (plant biomass) and local weather conditions good predictors of fire behaviour in tall-tussock grasslands?

In the eastern South Island, extensive tracts of tall-tussock (Chionochloa) grassland present at the time of European settlement have their origins in early Polynesian fires that allowed downslope migration of subalpine Chionochloa species (reviewed in McGlone 2001). The advent of pastoralism in the 1850s increased fire frequency, introduced mammalian herbivory, and precipitated widespread and dramatic changes in the stature and composition of these fire-induced grasslands (Connor 1964, O’Connor 1982). Change has been greatest in areas of lower rainfall, where the displacement of the dominant tall-tussocks by shorter tussock (Festuca, Poa) or mat (Hieracium, Raoulia) growth forms is exacerbated by a range of alien plant (e.g., Hieracium, Rosa) and animal (e.g., rabbit) pests. As the native forage resource declined, many montane grasslands were oversown with adventive pasture grasses and legumes, and since the introduction of aerial topdressing some have received intermittent applications of phosphate-based fertilisers (O’Connor 1987, O’Connor & Harris 1991). The net result is an extensive mosaic of modified grassland communities that retain varying degrees of native dominance and biodiversity.

The deliberate use of fire has long been a contentious issue in the South Island high country. While environmentalists frequently portray fire as damaging native flora and fauna, increasing opportunites for invasive weeds and promoting soil erosion, pastoralists see it as a means of improving access for stock, promoting palatable regrowth, and reducing or removing what are seen as ‘woody weeds’. Over recent decades, the retirement of pastoral leasehold land and predictions of reduced rainfall have raised concerns in some rural communities that the increased biomass (fuel load) of grasslands that are no longer intentionally burned or grazed may pose an increased fire risk during the dry summer and autumn months. From a conservation perspective there is also the issue of whether deliberate burning under damp spring conditions (the current pastoral management practice) is less damaging to the native biota than an accidental fire when weather conditions are hot and dry, and seasonal plant growth is well advanced.

Over the last 50 years a large number of studies have described the impacts of burning and grazing on tall-tussock grasslands, and have documented the effect of fire on the growth and forage quality of the dominant tussock species (reviewed in Basher et al. 1990, McKendry & O’Connor 1990). Most of these studies have examined only single aspects of the tall-tussock ecosystem, and none have established a pre-burn baseline and characterised the severity of the fire. In this paper we discuss the development of a study that aims to quantify the extent to which fire degrades the fertility of tall-tussock grasslands and depletes the native biota, and present some initial results on biomass and nutrient pools and fire behaviour.

Development of the project proposal
The study has its origins in a 1993 resource consent hearing which gave AC Sanders & Sons approval to burn red tussock (Chionochloa rubra) grassland on Little Valley Station, near Alexandra, and the subsequent outcry from many in the farming community when the then Minister of Conservation, Hon. Denis Marshall, intervened to overturn the approval. Central to the debate over whether the grassland should be burned and subsequently grazed was the question of whether red tussock (for which there was little documented information on fire effects) would respond in a similar manner to snow tussock (Chionochloa rigida), for which documented information on fire response was available.

In 1995/96 a consortium of three government agencies (Forest Research, Landcare Research, and the Department of Conservation) began developing a proposal to compare the effects of fire on red tussock and snow tussock grasslands. By early 1997 the group had failed to find areas of red tussock grassland (a relatively uncommon type of tall-tussock grassland in Otago) that were able to be burned safely, and were meeting opposition to using Conservation Land for their experimental sites. These issues came to a head at an April 1997 meeting of interested parties from conservation (Department of Conservation, Royal Forest & Bird Protection Society), farming (Federated Farmers), regional government (Otago Regional Council), university (Botany Department, Otago University) and government science (Forest Research, Landcare Research) agencies. That meeting changed the focus of the project to a comparison of fire effects in snow tussock grasslands from coastal (maritime) and inland (subcontinental) environments in Otago, and committed the researchers to seeking experimental sites outside the Conservation Estate.

The project is now a collaborative venture between the Department of Conservation and three government science agencies (AgResearch, Forest Research, and Landcare Research), and receives funding from the Foundation for Research, Science, and Technology, the Department of Conservation, the Hellaby Indigenous Grasslands Research Trust, and the National Rural Fire Authority.

Biomass losses associated with spring and summer burns

By contrast the summer fires at Deep Stream burned an average of 74% of the above-ground plant material, including most (89%) of the shrub, herb and lower plant biomass. Tussock tiller mortality was much higher than that after the spring burns, and some mature tussocks have been killed. The plant species least affected by the summer fires is the golden spaniard (Aciphylla aurea), which 6 weeks after the fires was regrowing vigorously from its blackened stumps (Figure 4).

In the absence of grazing by stock, tussock cover on the spring burn plots appears to be recovering well (Figure 5). Tussock recovery after the summer burns was much less advanced at the end of autumn, and the plots have gone into winter with little tussock cover and a high proportion of exposed soil (Figure 4).

Fire behaviour

Fire weather conditions were characterised using onsite climate data, and the fuel moisture codes of the Fire Weather Index (FWI) System¹ (Van Wagner 1987). The fuel moisture codes provide a measure of the dryness of available fuels and soil organic layers, based on the cumulative effects of temperature, humidity and rainfall. These data were supplemented by measurements of vegetation (height, cover, biomass, moisture) and soil (moisture) parameters (Tables 1&2).

The spring burns at Mt Benger were carried out in early November 2000. While this is later than pastoral burns are permitted, weather conditions were still characteristic of the spring-burning period. There was a good crisp frost on the morning of the burns, and snow blanketed the site two days later. The Deep Stream plots were burned in early March 2001, which was the first time the site had crossed our threshold for summer burns (FFMC > 75, DMC 10-30, DC > 200, BUI 20-50).

Table 2. Weather conditions, fire weather indicies, fuel characteristics, and fire behaviour for tall-tussock grassland fires at Mt Benger (spring burns) and Deep Stream (summer burns).

Among the fuel characteristics, the primary feature distinguishing spring and summer burns was the moisture content of the tussock bases and the upper soil layers (Table 2). Other parameters did not vary significantly between the seasonsMaximum fire temperatures measured using heat-sensitive paints also showed little variation between seasons. The highest temperatures were observed at the ground surface (> 1010 ºC), and at 1 m above-ground (>500 ºC).

Soil temperatures at depths of 2.5 and 5.0cm did not exceed 69 ºC (the temperature at which the first heat sensitive paint strip melts) during any of the spring or summer burns. Rates of fire spread (Figure 6, Table 2) for individual burns were influenced by wind speed, fuel moisture content and slope, and generally increased during the afternoon as wind speeds increased and fuels dried out. Despite quite different seasonal conditions, rates of fire spread were similar for spring and summer burns.

Figure 6. Recording the rate of fire spread for a spring burn at Mt Benger.

Acknowledgements:
We greatly appreciate the ongoing support of staff from the Department of Conservation (Otago Conservancy), particularly Jeff Connell, John Pearce, Marcus Simons and Dave White. Nigel Harwood (Dunedin City Council) sought and gained permission for us to use Deep Stream as one of our experimental sites, and Alan Michelle and John Deans allowed us to use land at Mt Benger. Ross Hamilton (Dunedin City Council) and Bob McNeil (Central Otago District Council) provided permits for the burns at Deep Stream and Mt Benger respectively. We also express our thanks to all those who shelved other priorities to help ensure the experimental fires stayed within their allotted bounds.

References:
Alexander, M.E. 1994. Proposed revision of fire danger class criteria for forest and rural fire areas in New Zealand. National Rural Fire Authority, Wellington. Circular 1994/2. 73 p.

Basher, L.R., Meurk, C.D., Tate, K.R. 1990. The effects of burning on soil properties and vegetation: a review of the scientific evidence relating to the sustainability of ecosystems and land use in the eastern South Island hill and high country. DSIR Land Resources Technical Record 18. Unpublished. 93p.

Connor, H.E. 1964. Tussock grassland communities in the Mackenzie Country, South Canterbury, New Zealand. New Zealand Journal of Botany 2: 325-351.

Fogarty, L.G.; Pearce, H.G.; Catchpole, W.R.; Alexander, M.E. 1998. Adoption vs. adaptation: lessons from applying the Canadian Forest Fire Danger Rating System in New Zealand. In Viegas, D.X. (editor). Proceedings, 3^rd International Conference on Forest Fire Research and 14^th Fire and Forest Meteorology Conference, Luso, Coimbra, Portugal, 16-20 November, 1998. pp 1011-1028.

McGlone, M.S. 2001. The origin of the indigenous grasslands of southeastern South Island in relation to pre-human woody ecosystems. New Zealand Journal of Ecology 25: 1-15.

McKendry, P.J., O’Connor, K.F. 1990. The ecology of tussock grasslands for production and protection. Unpublished report for the Department of Conservation. Centre for Resource Management, Lincoln University and University of Canterbury. 161p.

Molloy, B.P.J., Burrows, C.J., Cox, C.J., Johnston, J.A., Wardle, P. 1963. Distribution of subfossil forest remains, eastern South Island, New Zealand. New Zealand Journal of Botany 1: 68-77.

O’Connor, K.F. 1982. The implications of past exploitation and current developments to the conservation of South Island tussock grasslands. New Zealand Journal of Ecology 5: 97-107.

O’Connor, K.F. 1987. The sustainability of pastoralism. Pp. 161-188 in: Proceedings of the 1987 Hill and High Country Seminar, Lincoln College, Canterbury.

O’Connor, K.F., Harris, P.S. 199x. Biophysical and cultural factors affecting the sustainability of high country pastoral land uses. Pp. 304-313 in: Henriques, P.R. (ed.), Sustainable land management: the proceedings of the International Conference on Sustainable Land Management, Napier, Hawkes Bay, New Zealand.

Van Wagner, C.E. 1987. Development and structure of the Canadian Forest Fire Weather Index System. Government of Canada, Canadian Forestry Service, Ottawa, Ontario. Forestry Technical Report 35. 37 p.

Wiser, S.K., Rose, A.B. 1997. Two permanent plot methods for monitoring changes in grasslands: a field manual. Manaaki Whenua-Landcare Research, Christchurch. 51 p.