Jones, H. F. E., Özkundakci, D., Kochendoerfer, S., McBride, C. G., & Hamilton, D. P. (2014). Lake Rotokakahi water quality modelling. ERI report No. 32. Client report prepared for Bay of Plenty 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/10289/12788
Lake Rotokakahi is an iwi-owned lake, administered by the Lake Rotokakahi Board of Control. The lake has a surface area of 4.4 km2, a mean depth of 17.5 m, and a catchment of 19.7 km2 of which most is in exotic forest (46.5%), with some pasture (27.8%) and regenerating native forest and scrub (25.7%). Lake Rotokakahi has only one permanent surface inflow and it is assumed that the lake must be predominantly groundwater fed. The lake outflow, Te Wairoa Stream, flows into Lake Tarawera. Monitoring indicates that water quality has recently declined in the lake, with a shift from a mesotrophic state in the 1990's, to currently being classified as eutrophic. Increased algal biomass has reduced water clarity and in May 2011 the first ever recorded algae bloom of a toxic cyanobacterium species resulted in a fish kill in Te Wairoa Stream. With the majority of the catchment in either exotic or native forest, the cause for the decline in water quality is unclear. However there is some concern that forestry harvesting operations may lead to increased surface runoff of sediment and associated phosphorus in ephemeral streams that feed into the lake. The objective of this study was to setup a water quality model for Lake Rotokakahi, which in the future may provide a decisionsupport tool for lake managers. The model used, DYRESM-CAEDYM, is a one-dimensional (1D) coupled hydrodynamic-water quality model that has been widely used in New Zealand and overseas. Methodology for deriving model forcing data (e.g. meteorological data, inflow volumes and nutrient loads), including the application of catchment and lake water balances, is described in this report. The catchment water balance indicates that groundwater and rainfall account for 37% and 42%, respectively, of the total inflow volume to the lake, with surface and ephemeral inflows contributing a smaller proportion (c. 20% combined). Furthermore, it appears that groundwater contributes a significant proportion (c. 60 - 70%) of the total nutrient loads, although surface and ephemeral inflows also contribute a significant proportion (34%) of the total phosphorus load. It should be noted that the catchment water balance is based on a very limited dataset, and therefore subject to significant uncertainty. We simulated the period July 2009 to June 2012 with DYRESM-CAEDYM. Model performance statistics indicated reasonable simulation of water quality for the calibration period, but did not satisfactorily capture the magnitude and dynamics of chlorophyll a and some nutrient species over an independent model validation period. There may be a number of reasons for the poor model performance over the validation period. For example, it may be that zooplankton and/or freshwater mussels (kakahi), which were not included in the model configuration, may exert significant grazing pressure on the phytoplankton populations at certain times. Furthermore, the significant uncertainty in the catchment water and nutrient loads most likely will have affected model performance. Nutrient concentrations are routinely monitored only in the surface inflow. There are no measurements of nutrient and sediment loads in ephemeral streams, and there is very limited data on groundwater nutrient concentrations. In its current form, the DYRESM-CAEDYM model is not suitable for scenario testing of lake management options. It is recommended that effort is directed at quantification of ephemeral and groundwater inflows and their associated nutrient and (particularly for ephemeral inflows) sediment loads. This may be particularly important for ephemeral inflows located close to the lake that may be influenced by forestry harvesting operations. Recent increases in measured total phosphorus and phosphate concentrations in the surface inflow and in the lake, combined with the degradation of water quality that has already occurred over the last 20 years, provides further impetus for addressing this critical information gap. Were new data to become available, opportunities exist for improving the performance of the current model and for developing a three-dimensional coupled hydrodynamic-water quality model (e.g. ELCOM-CAEDYM) with representation of kakahi and/or zooplankton, which may be able to resolve questions pertinent to the functioning of the lake, and to lake and catchment management.
Environmental Research Institute, Faculty of Science and Engineering, The University of Waikato
© 2014 copyright with the author.