Monitoring Soil Carbon Stock Change Following Harvesting In a Pinus Radiata Plantation Forest In New Zealand.

In New Zealand, soil C stocks and stock changes are currently modelled using the national Soil C Monitoring System for the purposes of international reporting.  The system assumes stocks are at steady state after 20 years of constant land use.  Although the cyclical nature of potentially soil-disturbing management practices in planted forests has long been recognised, available data on changes in soil C stocks with time over multiple rotations in New Zealand is limited.  Soils were sampled before (in 1995/96) and after (in 2000, 2006, and 2010) forest harvesting at the Puruki experimental catchment, located in the central North Island, New Zealand.  The aim of the work was to investigate the change in C stocks and concentrations with time following harvesting.  Topsoil (0-0.1 m) stocks were measured across the Puruki catchment at 30 randomly selected 0.04 ha permanent sample plots (PSPs) in all years.  Deeper sampling was also undertaken within a sub-catchment at Puruki to account for possible C redistribution down the profile due to harvesting disturbance.  Soil C concentrations in the 0-0.1, 0.1-0.2, 0.2-0.5, and 0.5-1.0 m depth ranges were measured at 10 PSPs in 1996, 2000, and 2010.  Profile (0-1.0 m) C stocks were calculated for 2000 and 2010.  Topsoil C stocks were significantly decreased after ground-based harvesting in 1997 and had not recovered 13 years after harvest, suggesting a relatively long-lasting impact.  Hauler-based harvesting had no significant impact on topsoil stocks over the monitoring period.  No evidence for C loss from the top 1.0 m of mineral soil was found over the monitoring period under ground-based harvesting.  Continued monitoring of the site will be required to establish the nature of future C stock changes throughout the second rotation and into the third.  The comparative effectiveness and advantages/disadvantages of deep sampling using a motor-auger will also be discussed.