Effect of changes in forest water balance and inferred root reinforcement on landslide occurrence and sediment generation following Pinus radiata harvest on Tertiary terrain, eastern North Island, New Zealand

Main Article Content

Michael Marden
Donna Rowan
Alex Watson

Keywords

forest removal, hydrological change, duration of heightened slope vulnerability, alternative species options

Abstract

Background: The frequent occurrence of storm-initiated landslides following harvesting of Pinus radiata D.Don in steep, Tertiary terrain, East Coast region, North Island, New Zealand, is of increasing concern. This paper documents the influence of tree removal and of replacement plantings on the canopy water balance and soil moisture regime when slopes are at their most vulnerable to landslide occurrence.


Methods: At a previously established study site, rainfall, throughfall, and soil moisture data were collected before a mature stand of P. radiata was harvested. After harvesting, part of the study site was replanted with P. radiata at 1000 stems ha-1 and part with 500 stems ha-1. Relationships between hydrological changes and landslide occurrence are discussed in relation to planting density, site factors, root system development, silvicultural regimes, and alternative land use options for mitigating erosion in highly erodible hill country.


Results: Following harvesting, soil moisture levels remained higher for longer than under a mature forest until rainfall interception and evapotranspiration returned to pre-harvest levels. This coincided with canopy closure, irrespective of planting density. After thinning, interception and evapotranspiration decreased, then regained the equivalent of a closed canopy 2 years later. Landslide occurrence was highest on slopes >25° and with a NE aspect. Sediment generation rates were highest in 2─4-year-old plantings, then decreased markedly with increasing tree age.


Conclusions: Irrespective of planting density, P. radiata had little influence on the soil-water regime until canopy interception, evapotranspiration rates, soil-drying and recharge cycles returned to pre-harvest levels, coinciding with canopy closure. During this period, pore-water pressures at times of heavy or prolonged rainfall likely result in soil saturation and an increase in landslides. The progressive loss of root strength of the harvested trees had a secondary influence. The duration of the post-harvest period of heightened slope vulnerability to landslide initiation is a function of the combined influences of site factors on rates of tree growth and survival, and of the planting density regime on the canopy water balance and soil water content until the development of an effective live soil-root reinforcement system. For areas identified as high risk, the targeting of high-value timber species with longer rotation length, including consideration of coppicing species, would minimise the risk of slope failure at harvest. Very high-risk areas unsuited to rotational harvesting will ultimately require transitioning to a permanent indigenous forest cover.

Abstract 1312 | PDF Downloads 692