Introduction

Suspended sediments (SS) are organic and inorganic particles transported in suspension by water (Bilotta et al., 2009). Rainfall and surface runoff cause erosion, leading sedimentation in surface waters and causing aquatic ecosystem degradation by disrupting aquatic habitats and food webs (Howard-Williams, Davies-Colley, Rutherford, & Wilcock, 2010), and by delivering sediment-bound nutrients that contribute to eutrophication (Dare, 2018). Land use developments clearing native forests have accelerated the already naturally high rates of erosion across New Zealand and caused significant sedimentation in lakes and streams (Ministry for the Envrionment, 2019).
Pastoral agriculture in New Zealand is strongly associated with eutrophication and degraded freshwater ecosystems (Verburg, Hamill, Unwin, & Abell, 2010). Treading by grazing animals physically disrupts the soil, and increases the likelihood of surface runoff and erosion by decreasing infiltration rates and porosity, and impairing plant growth (Bilotta, Brazier, & Haygarth, 2007; R. W. McDowell et al., 2003; Ward, Talbot, Denne, & Abrahamson, 1985). Year-round grazing and high stocking rates used to graze crops are common practices in New Zealand and contribute to increased erosion (Monaghan et al., 2007). Addressing erosion is a challenge for pastoral farmers in New Zealand, particularly those on sloping landscapes, due to variable precipitation patterns with very wet winters, and dry summers interspersed with large storms (R. McDowell, Wilcock, & Hamilton, 2013). Erosion is likely to be intensified by climate change causing more dramatic hydrologic conditions (Ministry for the Envrironment, 2019; Ockenden et al., 2016).
Since the 1960’s water quality in Lake Rotorua, in the Bay of Plenty Region on the North Island of New Zealand, has declined due to nutrient inputs from residential, commercial, industrial and agricultural developments (Environment Bay of Plenty, 2009). An estimated 43% of the annual total phosphorus (P) delivered to the lake comes from pastoral dairy and drystock farms which cover ~48% of the 42,000 ha Lake Rotorua surface catchment (Bay of Plenty Regional Council, 2012). Between 71-79% of the anthropogenic P delivered to the lake is sediment bound (Hamill, 2018), and a portion of that may become biologically available under anoxic conditions which occur in Lake Rotorua and contributes to lake eutrophication (Abell & Hamilton, 2013). Lake sediments release an estimated 36 t P y-1, accounting for ~48% of the total annual P loading (Bay of Plenty Regional Council, 2012). Intensified anoxic conditions during periods of warmer temperatures have cause significantly greater contributions to annual P loads by lake sediments (Burns, McIntosh, & Scholes, 2005).
The 2012 Lake Rotorua Management Plan has set a target to reduce annual TP loads delivered from the catchment by 10 t P y-1 by 2022 from the estimated 39 t P y-1, in order to restore lake water quality (Bay of Plenty Regional Council, 2012). Achieving Lake Rotorua water quality targets by addressing P loading from pastoral agriculture will require combining multiple appropriate nutrient mitigation strategies and potentially developing new technologies (R. W. McDowell, 2010).
Mitigation strategies that increase stormflow residence time have been found to decrease surface runoff flows, leading to increased sediment deposition by decreasing the kinetic energy of flowing water (Dosskey, 2001; McKergow, Tanner, Monaghan, & Anderson, 2007; Stanley, 1996). However, the type of mitigation strategy affects the duration sediments are attenuated. Studies have found that sediment retention times are more brief (weeks to months) where sediments accumulate in more concentrated areas, such as narrow grass filter strips and constructed treatment wetlands, compared to strategies where sediments are blanketed over a wide area, which may have retention times of up to hundreds of years (McKergow et al., 2007).
Previous research has found that ponding surface runoff can decrease discharge concentrations and loads of sediments and particulate bound P by decreasing the kinetic energy of flowing water (Brown, Bondurant, & Brockway, 1981; Harper, Herr, Baker, & Livingston, 1999; Levine et al., 2019; R. W. McDowell et al., 2006; Stanley, 1996). Detainment bunds (DBs), are earthen storm water retention structures constructed on pastures across the flow path of targeted low-order ephemeral streams capable of impeding stormflow and temporarily ponding up to 10,000 m3 of surface runoff. The structures were implemented in the Lake Rotorua catchment in 2010 as a mitigation strategy to target P losses from pastures (Clarke, 2013). A concurrent paper by Levine et al., (In review) investigating the strategy’s effect on stormwater volumes leaving the same DB catchments during the same storm events as this current study presents further detail on DB design and function.
Preliminary studies of DBs in the Lake Rotorua catchment have found that P enriched sediments were deposited in DB ponding areas (Clarke, 2013), and that a DB effectively decreased runoff volumes, and sediment and P loads, discharged during 3 non-consecutive ponding events (Levine et al., 2019). A concurrent study investigating the same ponding events at the same DB sites as this present study reported 31 and 43% and noted that deposited sediments could be developing a less permeable surface soil layer and/or clogging soil pore spaces and causing infiltration rates to decline in the ponding areas (Levine et al., In review).
Although erosion is recognised for its potential impact to aquatic ecosystems, there is room to progress our understanding of the transport and fate of sediments from intensively managed pastures (Haygarth et al., 2006). To determine if DBs could provide a viable strategy for pastoral farmers to improve Lake Rotorua water quality, it is important to quantify their ability to decrease SS loads. The main objective of this study was to measure the effect of the DB strategy on SS concentrations and yields delivered to 2 DBs with 55 ha and 20 ha catchments, located on pastures in the Lake Rotorua catchment, and identify the factors influencing the results. Previous studies on DBs, and related mitigation strategies, suggest ponding surface runoff facilitates sedimentation, although there is currently no definitive research quantifying the impact of the DBs on annual sediment loads transported from pastures in the Lake Rotorua catchment. We hypothesised that ponding surface runoff will facilitate sedimentation and result in lower discharge concentrations, which, combined with soil infiltration lowering runoff discharge volumes reported in a concurrent study by Levine et al. (In Review), will decrease annual SS loads discharged from the treatment catchments.