Abstract

Key Words

Soil carbon, paired plots, grassland, shrubland

Introduction

New Zealand is obliged under the United Nations Framework Convention on Climate Change and the Kyoto Protocol to report on greenhouse gas emissions and removals arising from land use, land-use change and forestry activities. The NZCAS will use historic soil databases and new process based models to account for C fluxes in soils arising from these activities (Tate et al. 2003, Scott et al. 2002). Paired plot or chronosequence studies are required to verify the direction and magnitude of C fluxes from these estimates. Paired plots have the advantage of providing immediate results, unlike chronosequence studies that may require years or decades of monitoring to measure such changes.

The two major land-use changes to have occurred in New Zealand since the beginning of 1990, the baseline year for C accounting, are from grassland to shrubland and from grassland to planted forest. As part of the NZCAS it is proposed to install paired plots at 32 sites (16 in each land use change) during the 10-year period from 2003 until 2012 to verify estimates of C flux in soils arising from these land use changes. Although some paired plot data already exist for afforestation of grassland (Davis & Condron 2002) uncertainties are high and more data are needed, particularly for soil depths below 0.1 m. Few data are available for the effects on soil C of shrubland establishment and growth in grassland. The paired plots selected in the first few years of the project will address this particular issue. This note outlines plot selection and sampling procedures adopted for paired site establishment for the grassland-shrubland transition, and presents some initial results.

Methods

Suitability criteria based on Conteh (1999) and Davis et al. (2004) were followed for the selection of plot pairs. Potential locations for plot pairs were identified from field knowledge, discussion with landowners, or from satellite images, maps, and aerial photographs. The most promising sites were visited and a brief examination of the site and soils for their suitability was made. For selection, the site pair was required to have, as far as possible, similar soils (hence similar parent materials, climate, original vegetation, topography and age), slope, aspect and elevation, but contrasting land use. One 20 x 20 m plot of each pair was located on land that had been in long-term grassland, and the other on nearby land where shrubland had developed since about 1990. Following initial selection, similarity was verified (or rejected) after soil examination by an independent pedologist. This involved examination and pedological description of eight soil profiles evenly spaced around the perimeter of each plot of the selected pair.

Soil sampling procedures for soil C determination followed Davis et al. (2004). Briefly, plots were divided into four quarters, and one sampling point was randomly located in each quarter. At each sampling point litter (L) and ferment humus (FH) samples, where present, were collected from three 0.1 m² quadrats spaced 1 m apart across the slope, with the centre quadrat located at the randomly selected sample point. The three samples were bulked within plot quarters. Soil samples from 0–0.1 m, 0.1–0.2 m and 0.2–0.3 m depths were collected at the random point using 98 mm diameter by 100mm deep stainless steel sampling rings. Additional samples were collected, using a 25-mm diameter core sampler, in 0.1 m increments to 0.3 m depth at eight locations spaced 0.25 m apart across the slope and centred on the random sampling point. Within each quarter-plot, these 25 mm diameter core samples were bulked by depth.

To enable future location of the plots, plot corners were marked with flagging tape and wooden pegs. Additionally, 15-mm diameter steel pegs were driven into the soil at plot corners and the plot centre to enable exact future location using a metal detector. GPS co-ordinates at plot centres were also recorded together with their placement errors as shown by the GPS instrument (to the nearest 10 m).

Litter and FH layer material was oven dried (70^oC) and weighed, and a sub-sample was ground for determination of C and nitrogen concentration. Soil collected using 0.1 m diameter corers was air dried and passed through a 2 mm sieve, and the weight of <2mm soil and > 2mm fractions were determined after oven drying. Soil collected using 25 mm diameter corers was air dried and passed through a 2 mm sieve for analysis. Carbon and nitrogen concentrations (nitrogen concentrations not presented here) were determined in sub-samples of < 2 mm soil and L and FH layer materials using a LECO CNS analyser. Carbon mass was determined by applying the C concentration data to the appropriate mass data.

Initial Results

The three initial paired sites selected were all located on Orthic Brown Soils (New Zealand Soil Classification, Hewitt 1998), that fall into the IPCC (Intergovernmental Panel on Climate Change) High Clay Activity soil grouping (Table 1). Grassland plots were dominated by high fertility pasture species at the Waitotara site, and by low fertility species at the Puketoi and Balmoral sites. At the two higher-rainfall North Island sites the shrubland plots were dominated by a single species, the native myteraceous shrub manuka (Leptospermum scoparium). At the South Island site the shrubland plot was again dominated by a single species, broom (Cytisus scoparius), an exotic legume. From stem-ring counts the oldest plants at the Waitotara and Balmoral sites were about 9 and 12 years old respectively. Anecdotal evidence indicated that the oldest plants at Puketoi were also about 12 years, and stem diameters at 0.5 m height there (40-70 mm) were within the range of those at Waitotara (20-90 mm). Thus all three shrublands appeared to have developed since 1990.

Soil C mass varied greatly between the three sites (Fig 1). Site mean C mass to 0.3 m depth amounted to 62 Mg ha^-1 at Balmoral, 92 Mg ha^-1 at Waitotara, and 193 Mg ha^-1 at Puketoi. There were no significant differences between grassland and shrubland at any of the depths measured, nor did mean total soil C mass to 0.3 m differ significantly between grassland (112 Mg ha^-1) and shrubland (120 Mg ha^-1).

An FH layer was present at both manuka shrubland plots, but not in the broom shrubland plot. The C mass in this layer amounted to 1.3 and 0.4 Mg ha^-1 at the Puketoi and Waitotara sites respectively.

Table 1. Paired site soil, rainfall and vegetation characteristics.

¹ New Zealand Soil Classification
² Intergovernmental Panel on Climate Change
³ Ryegrass, Lolium perenne; white clover, Trifolium repens; browntop, Agrostis capillaris; lotus, Lotus pedunculatus; matagouri, Discaria toumatou.

Fig. 1. Soil carbon mass for three paired grassland and shrubland sites. Values are means of four subplots, bars show standard errors.

Conclusion

Results from sampling of the initial three paired sites showed no significant difference between grassland and shrubland soil C to a depth of 0.3 m. Between-site variance was large, however, and additional sampling is required to accurately determine the magnitude and direction of change in soil C with land use change from grassland to shrubland.

Acknowledgements

Graham Shepherd and Suzanne Lambie assisted with plot establishment and sampling in the North Island. Landcare Research Environmental Chemistry Laboratory carried out the soil analyses. Guy Forrester, Landcare Research, Lincoln, carried out the analysis of variance on the analytical data. Alan Palmer, Massey University, Palmerston North verified paired plot similarity.

References

Conteh A (1999) Evaluation of the paired site approach to estimating changes in soil carbon. Discussion Paper 3, of Appendix 6 to Technical Report No. 2 (Australian Greenhouse Office: Canberra).

Davis M, Condron L (2002) Impact of grassland afforestation on soil C in New Zealand: A review of paired site studies. Australian Journal of Soil Research 40, 675-690.

Davis M, Wilde H Garrett L Oliver G (2004) New Zealand carbon monitoring system soil data collection manual. (Ministry for the Environment: Wellington).

Hewitt A (1998) New Zealand Soil Classification. (Manaaki Whenua - Landcare Research: Lincoln).

Scott NA, Tate KR, Giltrap DJ, Smith CT Wilde RH Newsome PF Davis MR (2002) Monitoring land-use change effects on soil carbon in New Zealand: quantifying baseline soil carbon stocks. Environmental Pollution 116, S167–S186.

Tate KR, Scott NA, Saggar S, Giltrap DJ, Baisden WT, Newsome PF, Trotter CM, Wilde RH (2003) Land-use change alters New Zealand's terrestrial carbon budget: uncertainties associated with estimates of soil carbon change between 1990-2000. Tellus 55B, 364-377.