Drainage to groundwater under a closed-canopy radiata pine plantation on the Canterbury Plains, South Island, New Zealand

The amount of soil water draining through the root zone to shallow groundwater was measured in two stands, Shellocks and Doyles, at two radiata pine (Pinus radiata D. Don) plantations on the Canterbury Plains, South Island, New Zealand. The soil at both sites is a free-draining, very stony silt loam, and is classified as a Pallie Firm Brown soil belonging to the Lismore Series. Rainfall and interception were recorded and used as input to a daily water balance model. The portion of throughfall available for groundwater recharge (soil water drainage) was measured using pairs of free-draining zero-tension lysimeters. This information, together with stand transpiration (sap flow) and soil moisture storage (neutron probe) data, was used for model validation. Rainfall for years 2001 (538 mm) and 2002 (711 mm) was below the long-term (1919-1980) mean of 801 mm. Total stand transpiration from 1 May 2001 to 31 December 2002 at Shellocks and Doyles was 480 ± 120 mm and 830 ± 180 mm respectively. Rainfall for the same period was 1139 mm and 1143 mm respectively. Throughfall recorded at Doyles from October 2001 to July 2002 was estimated to be 70% of rainfall. During this study rainfall rarely wetted the soil to depths much in excess of 1 m. Very little water was available for deeper drainage and there were few opportunities for groundwater recharge. Under tall woody vegetation, discharge from these soils is driven by sporadic intense and/or longduration rainfalls. In this environment, transpiration rapidly removes soil moisture and thereby reduces the potential for soil drainage. The wetting-up phases are most likely to produce soil water drainage capable of generating groundwater recharge. These phases tend to occur late autumn to early spring and occasionally during summer wet periods with high intensity and/or long duration rainfalls. For some years, when they supported a closedcanopy forest, there was no drainage at all from these soils. During the period measured, drainage was infrequent and limited to the winter months. The daily water balance model provided an effective estimate of soil moisture conditions and timing of when drainage to groundwater was likely to occur. The soil water drainage predicted for the longer-term, higherintensity rainfalls tended to be too high, particularly when pre-storm soil moisture conditions approached field capacity.