Recovery of topsoil characteristics after landslip erosion in dry hill country of New Zealand, and a test of the space-for-time hypothesis

The rate of development of topsoil is an important characteristic for soil resilience and sustainable use. We located a chronosequence (1-59 yr) of recovering landslip scars in erodible siltstone hill country under permanent pasture for sheep farming in New Zealand. We measured the rates of recovery in microbial C, respiration, catabolic diversity, phosphatase, sulphatase and invertase activities, pH, total C, total N, C/N ratio, potentially mineralizable N, total P, Olsen P, cation-exchange capacity, bulk density, particle density, porosity, available water and aggregate stability (0-10 cm depth). A subset of the same sites was sampled again after a 14-yr interval, enabling us to test whether rates of change estimated by resampling the same sites were the same as those estimated from a single time sample from the chronosequence (the space-for-time hypothesis). Most topsoil characteristics had recovered to 71-85% of those in the non-slipped sites after 59 yr. Exceptions were soil respiration, invertase and sulphatase activities, and bulk density, which recovered to 94-110% of the values of the non-slipped sites. There was little change in soil pH, total P, Olsen P, exchangeable cations and water storage along the chronosequence. An asymptote model fitted the patterns of recovery in biochemical characteristics, organic matter, bulk density and particle density. Recovery (to 90% of the asymptote value) was most rapid for the C/N ratio (5 yr) and longest for particle density (79 yr); most other characteristics fell in an 18-50 yr range. Overall, a single sampling of a chronosequence of matched landslip scars was as reliable to estimate rates of recovery as was resampling individual sites through time. Total C and N were as effective as more complicated biochemical measures to monitor the recovery of topsoil. (C) 2003 Elsevier Ltd. All rights reserved.