Soil organic carbon stocks and flows in New Zealand: System development, measurement and modelling

K. R. Tate¹, R. H. Wilde¹, D. J. Giltrap², W. T. Baisden¹, S. Saggar¹, N. A. Trustrum³, N. A. Scott⁴, and J. P. Barton⁵

¹Landcare Research, Private Bag 11052, Palmerston North, New Zealand; ²Small Office Systems Ltd, P.O.Box 46 024, Lower Hutt, New Zealand (deceased); ³Instiitute of Geological and Nuclear Sciences, P. O. Box 30368, Lower Hutt, New Zealand; ⁴Department of Geography, Queen’s University, Kingston, Ontario, Canada K7L 3N6; and ⁵Ministry for the Environment, Climate Change Office, P.O. Box 10362, Wellington, New Zealand

Can. J. Soil Sci. 85: 481-489

Full text pdf

An IPCC-based Carbon Monitoring System (CMS) was developed to monitor soil organic C stocks and flows to assist New Zealand to achieve its CO₂ emissions reduction target under the Kyoto Protocol. Geo-referenced soil C data from 1158 sites (0.3 m depth) were used to assign steady-state soil C stocks to various combinations of soil class, climate, and land use. Overall, CMS soil C stock estimates are consistent with detailed, stratified soil C measurements at specific sites and over larger regions. Soil C changes accompanying land-use changes were quantified using a national set of land-use effects (LUEs). These were derived using a General Linear Model to include the effects of numeric predictors (e.g., slope angle). Major uncertainties arise from estimates of changes in the areas involved, the assumption that soil C is at steady state for all land-cover types, and lack of soil C data for some LUEs. Total national soil organic C stocks estimated using the LUEs for 0-0.1, 0.1-0.3, and 0.3-1 m depths were 1300 ± 20, 1590 ± 30, and 1750 ± 70 Tg, respectively. Most soil C is stored in grazing lands (1480 ± 60 Tg to 0.3 m depth), which appear to be at or near steady state; their conversion to exotic forests and shrubland contributed most to the predicted national soil C loss of 0.6 ± 0.2 Tg C yr^-1 during 1990-2000. Predicted and measured soil C changes for the grazing-forestry conversion agreed closely. Other uncertainties in our current soil CMS include: spatially integrated annual changes in soil C for the major land-use changes, lack of soil C change estimates below 0.3 m, C losses from erosion, the contribution of agricultural management of organic soils, and a possible interaction between land use and our soil-climate classification. Our approach could be adapted for use by other countries with land-use-change issues that differ from those in the IPCC default methodology.

Key words: Soil organic carbon, land-use change, stocks, flows, measurement, modelling, IPCC

© 2005 Agricultural Institute of Canada