Hydrogeology and revegetation of the Wangaloa opencast coal mine, South-east Otago, New Zealand

Todd, Andrea Jane.

The Wangaloa opencast coal mine was operational from 1945 to 1989. Initial rehabilitation to Pinus radiata was unsuccessful and an alternative programme was adopted in 200 1, which involved the clearing of exotic vegetation, re-contouring of overburden waste piles and the planting of approximately 70,000 native seedlings. The overall aims of this thesis were to investigate the physical hydrogeology of Wangaloa, and to gain an understanding of the factors influencing native plant revegetation at the site. A steady state hydrogeological model was constructed for Wangaloa using Visual MODFLOW. The modelled total flux of 344-351 m3/day and drain outflow of 117m3/day met the calibration criteria. There was a discrepancy of 2% between the modelled inflow and outflow. The shape of the modelled potentiometric surface was similar to contours manually interpreted from piezometer heads. However, the calculated head values poorly matched the observed values with a normalised RMS of 24%. The model was sensitive to small changes in hydraulic conductivities, recharge, evapotranspiration and boundary conditions. Overall, the model did not adequately represent the hydrogeological system at Wangaloa, and manual methods may be more appropriate. To further develop the model additional information is required, especially on the complex geology and hydraulic conductivities at the site. Substrate types at the mine site included soils and overburden with varying amounts of quartz, silt and coal. Organic matter content, moisture content and bulk density were characterised for the site based on 68 sample locations. Substrate pH was measured in December 2003, March, June and September 2004. Average pH ranged from 2.6 to 5.4 over the site. Overburden pH (2.6-5.4) was more variable than pH of the soils (4.8 to 4.7). The relatively high pH values on overburden were attributed to the application of topsoil to some areas, while the low values are due to acid mine drainage. The performance of 470 native plantings in 54 plots was examined by assigning survival ratings and measuring their size in December 2003, March, June and September 2004. Kohiihii, koromiko, manuka and wineberry were investigated to ascertain variances in performance between species. The seedlings were planted in the autumn and spring of 2003, and by September 2004 the mortality rate was 22-29%. Wineberry had a significantly higher death rate than other species, at almost 50%. Dead plants were distributed across the site in all substrate types, although, eastern mine soil plots generally had higher mortality rates. Group 1 (autumn) plantings had an average growth rate of 110% in September 2004, while Group 2 (spring) had 26%. Of the selected species koromiko grew best for both Group 1 (123%) and 2 (85%). Wineberry had the lowest growth rate for Group 1 (29%), while kohiihU had the lowest for Group 2 (8%). Spatial patterns of planting growth were more distinct than planting mortality. Overall, growth on overburden was lower than on mine soils, and eastern mine soil plots had less growth than those on western mine soils. Planting performance data and substrate properties (pH, bulk density, organic matter and moisture content) were compared using a geographical information system (GIS) and geographically weighted regression (GWR) to identify relationships between these variables. Results showed that there were no clear relationships between these substrate properties and planting mortality. In contrast, the substrate data explained on average 80% of the variation in planting growth. Other factors that were identified in this study as likely to have had some effect on planting performance included overburden versus non-overburden substrates, species selection, browsing and time of planting. It was also suggested that the level of land disturbance, distance from remnant native bush, landforms/microclimates, nutrient toxicity and/or deficiency, planting stock genetics and planting technique may have been of influence. From this study, it was apparent that there are limitations in the use of standard soil analysis techniques for mine substrates. This study highlighted the importance of considering environmental variables in combination, rather than individually, when assessing revegetation success.

x, 161, [40] p. : ill., maps ; 30 cm.

2005Todd