Nitrate processes in the Pukekohe–Bombay area

Morgenstern, U.; Buckthought, L.; Gardner, P.; Stenger, R. 2024 Nitrate processes in the Pukekohe–Bombay area. Lower Hutt, NZ: GNS Science. GNS Science report 2024/31. 55 p.; doi: 10.21420/FPET-WT69

Abstract

High-intensity land use – mainly market gardening on the weathered basalts that form rich horticultural soils – has caused high nitrate concentrations in groundwater and streams discharging from the Pukekohe and Bombay volcanoes. These concentrations significantly exceed the national bottom line for nitrate toxicity in rivers from the National Policy Statement for Freshwater Management 2020 and maximum acceptable value for nitrate in drinking water in this area. The main nitrate load in the Pukekohe–Bombay area, from market-gardening activity, infiltrates into large groundwater stores in the basalt formation. It discharges mainly through three large springs: the Patumahoe and Hickey springs at Pukekohe, with a Mean Transit Time (MTT)of 17 years; and Hillview Spring at Bombay, with a MTT of 36 years. Within the basalt formation, no electron donors (for example, from organic matter) are present to facilitate microbial denitrification reactions. Without denitrification in this groundwater system, the entire nitrate load into the basalt formation eventually returns to the surface with a lag time of 17 and 36 years at the abovementioned springs and high nitrate-N concentrations up to 25 mg/L. This causes high nitrate-N concentrations in the receiving streams near the springs: 9 mg/L in Hingaia Stream, 14 mg/L at Whangapouri Creek and 19 mg/L in Whangamaire Stream. Because of the long MTTs of nitrate load through the basalt formation, the results of potential source-mitigation actions will be significantly delayed. For this study, several stream sites that had been comprehensively sampled during summer baseflow conditions were re-sampled during winter baseflow. This was done to improve understanding of the hydrologic system and its associated nitrate loads, including seasonal variability. Similar concentrations of radon in these streams between summer and winter indicate that increased flow contributions during winter baseflow are also maintained from a groundwater system, as opposed to shallow flow with MTTs of only days, and surface run-off, which would both have lower radon concentrations. Tritium data showed that all sampled streams contained younger water in winter compared to summer, indicating activation of shallower flow paths into the streams during the wet season. However, even at winter baseflow, the stream waters were still relatively old, with MTTs between 6 and 12.5 years. In stream catchments where shallow flow paths were activated during winter and contributed to basalt-discharge-dominated flow, nitrate concentrations were lower compared to summer, likely due to dilution. Where the discharge from a large basalt spring also dominated the flow at winter baseflow, nitrate concentrations were similar between summer and winter. In catchments without basalt discharge, nitrate concentrations were higher during winter baseflow compared to summer baseflow where shallow flow paths were activated during winter. This applies particularly to the area with Pleistocene geology, resulting in nitrate from pastural farming (dairy and drystock) being flushed into the streams via near-surface and shallow groundwater flow paths.While the passage of the high nitrate load through the basalt formation and the seasonal flushing of nitrate from the pastoral grazing land in the Pleistocene is reasonably well understood, little is known about nitrate sinks within this catchment. Nitrate loads significantly decreased along the course of most of the sampled streams. Some streams, mainly those in the Pleistocene formation, had near-zero nitrate concentrations despite pastural farming being the predominant land use in their catchments. This implies that there are significant nitrate sinks in these catchments. Better understanding of these nitrate sinks may enable enhancement of natural attenuation of existing high nitrate loads in these waterways. Anoxic groundwater discharges from the volcanics/unconsolidated formations, the presence of excess nitrogen (the decay product of denitrification) in groundwater that has evolved to complete denitrification, and potential dilution of stream nitrate loads by anoxic groundwater discharges indicate that denitrification in groundwater systems occurs in these formations. Unfortunately, it was not possible to access significant anoxic discharges to quantify their contributions to the overall nitrate sink. The largest nitrate sinks within the catchment were found within the surface waterways –in streams and ponds/wetlands. Nitrate stable isotopes, despite their science being somewhat ambiguous, indicate that the main process of nitrate removal from these waterways is through natural denitrification into nitrogen gases. This is likely to occur within the shallow layers ofthe sediments, with the nitrate transported into the sediments through hyporheic exchange. Denitrification, as opposed to uptake by aquatic plants, implies that the nitrate is being removed from the waterways permanently.In smaller streams (Ngakoroa, Oira) and a pond (Puk5), 80–100% of the nitrate from land-use activities was found to have been removed. In two of the largest streams (Whangamaire and Whangapouri), nitrate-load reductions of c. 30% were observed between the sampled sites along the courses through the Pleistocene formations. It is likely that denitrification also occurs above and below these sampled stream sections, implying that total nitrate load reduction over the entire stream length is significantly higher. For example, in the top part of Whangamaire Stream, between Patumahoe Spring and the first sampling site, nitrate concentrations had already been reduced by 30%, probably mainly within the pond/wetland at Patumahoe Spring. Ponds/wetlands and stream beds appear to be efficient nitrate sinks in the Pukekohe–Bombay area. In the Pleistocene formation, where land use is mostly pastural farming, nitrate is flushed out seasonally. This area discharges water and nitrate loads only via shallow flow paths, which are not active in summer. In Oira Creek, for example, nitrate had been completely removed from the water during the low summer flow. But even in winter, when discharges are active, it is likely that more than 80% of the nitrate is being removed from the water. This nitrate removal may partially occur within the stream bed but could also have significant contributions from within the soil and near the redox interface in this geologic formation (auths)