Tempero, G. W., & Hamilton, D. P. (2014). Final report: inflow monitoring of the Rotopiko lakes and Lake Mangakaware. ERI report No.49. Client report prepared for Waikato Regional Council. Hamilton, New Zealand: Environmental Research Institute, Faculty of Science and Engineering, The University of Waikato.
Permanent Research Commons link: https://hdl.handle.net/10289/12454
Farm drains flowing into the Rotopiko (Serpentine) North, Rotopiko South and Lake Mangakaware were monitored from April 2012 to April 2014. These drains included two tributaries to Rotopiko North (designated Rotopiko North 01 and 02), one tributary to Rotopiko South (designated Rotopiko South 02) and three tributaries to Lake Mangakaware (designated Mangakaware 01, 05 and 06). Surface-water grab samples were taken from each tributary at two-month intervals. Discharge was also measured on 13 separate occasions in each tributary. In June 2012, water-level loggers were installed in the Rotopiko South site and two of the Mangakaware sites. These readings were used together with direct-discharge measurements to provide continuous, instantaneous discharge. High-frequency sampling was also undertaken over 48 h on four separate occasions. Single point-discharge measurements and water quality samples were taken at the four sites up to 48 h following the start of rainfall. In addition, one of the four inflows was sampled automatically at 2-h intervals for up to 48 h. Both grab and automated water samples were analysed for total suspended solids (partitioned by volatile and non-volatile fractions), total nutrients (total nitrogen and total phosphorus) and dissolved inorganic nutrients (nitrate-N, ammonium-N and dissolved reactive phosphorus). All inflows showed high variability in water discharge and composition, driven primarily by weather and partly by drain maintenance. Droughts in the summers of 2012-13 and 2013-14 resulted in standing water with no measureable discharge in most of the inflows. In some cases, the inflow channel dried completely and no water samples were collected. Discharge to Lake Mangakaware captured with the high-frequency sampling responded to intense rainfall events with a 4 ̶ -6 h lag. Changes in drains due to plant growth and maintenance activities made it difficult to interpret some of the temporal responses of drains to rainfall. Drains became increasingly clogged with aquatic weeds and filamentous algae over summer. Periods of elevated discharge were characterised by increases in particulate phosphorus and suspended solids concentrations in all drains. It is surmised that much of the phosphorus transported into drains (and subsequently to the lake ecosystems) is derived from sediment erosion from the agricultural landscape. Management of losses of sediment from these agricultural landscapes may have an important influence on the rate of growth of aquatic plants and also the trophic state and rates of sedimentation in the receiving lake environment. We observed different responses of Mangakaware 05 and 06 sites in high-frequency measurements of storm events. Concentrations of suspended solids increased rapidly in Mangakaware 05 at the onset of increased discharge; suggesting a nearby source of sediment. The Mangakaware 06 storm event was much smaller (29.6 mm) than the event monitored at Mangakaware 05 (58.5 mm) and increases in suspended solids concentrations tended to more closely align with increases in discharge in this tributary inflow, suggesting a more diffuse source of sediment input. In contrast to the tributary inflows to Lake Mangakaware, many of the tributary inflows to the Rotopiko lakes were ephemeral, and measureable inflow ceased (or the drain dried up) for up to six months of the year. Without quantifying the contribution of groundwater, there may be a high degree of uncertainty in nutrient budgets or models constructed for these lakes on the basis of measured surface water inflows. Over the short-term (1-2 days), however, high-intensity rainfall events contribute strongly to variations in suspended sediment and nutrient loads to the peat lakes in this study. Catchments with primarily peat soils (e.g. Rotopiko Lakes) appear to be particularly susceptible to high-intensity rainfall events; suspended solids concentrations of 120 g m⁻³ occurred in tributary inflows following rainfall of c. 10 mm h⁻¹, compared to typical baseflow suspended solids concentration <3 g m⁻³. Our results suggest that farm drains are highly dynamic and strongly influenced by: (i) the time scale of rainfall events, ranging from seasonal to hourly, (ii) growth and decay of aquatic plants and filamentous algae within the drains at seasonal time scales, and (iii) drain maintenance programmes that may have beneficial effects of removing fine, nutrient-rich sediments in the bed of the drains while at the same time resulting in rapid mobilisation of a fraction of these sediments and providing an easier conduit for sediment and nutrient inputs to enter the lakes. The findings in this report indicate that storm events contribute to substantial variations in surface-water sediment and nutrient loads to the Rotopiko Lakes and Lake Mangakaware. Management efforts to reduce these loads are required as both systems, particularly the Rotopiko Lakes, support significant ecological communities and represent important remnant biodiversity. Within the agricultural landscapes of these lakes an important consideration should be identifying critical source areas of sediment and phosphorus and improving management of these areas to effect improvements in water quality and biodiversity. For agricultural drains it will be important to achieve some balance between drain access for maintenance, riparian planting of sufficient width and aspect to reduce sediment and nutrient loss, and in-drain plant establishment (to optimise opportunities for sediment and nutrient removal). Our results indicate a need for further research and consultation with the aim of developing consistent guidelines for drain maintenance and riparian planting, as well as supporting management of key areas where sediment and nutrient runoff is disproportionately large. Inflow monitoring programmes for farm drains should also be targeted towards storm events which are likely to make a disproportionately large contribution to surface-water sediment and nutrient loads compared with larger stream inflows which have a proportionately larger baseflow component. The monitoring could also be tailored specifically to support modelling required to construct lake water balances, identify groundwater contributions of water, sediment and nutrients to the lakes, and to calculate whole-lake nutrient and sediment budgets.
Environmental Research Institute, Faculty of Science and Engineering, The University of Waikato
© 2014 copyright with the authors.