GLEAMS - Waiarohia Catchment

This study of predicted impacts of proposed earthworks focuses upon a 201-hectare parcel of land that lies to the south of the Whenuapai airbase.

Additional aims of the study were:

• to assess the effectiveness of specified sediment control measures associated with the proposed earthworks;
• generate information for feeding into the wider Upper Waitemata Harbour (UWH) contamination accumulation study;
• provide validation (or not) of the need for modelling sediment yield in the UWH study; and
• provide an assessment of the risk of excessive sedimentation occurring in the Upper Harbour  as a result of the proposed developed in the Waiarohia catchment

A physically based field scale computer model GLEAMS (Knisel 1993) is used to predict sediment generation for the existing pastoral landuse and the earthworks phase of the proposed development, both with and without sediment control measures. Three control measures are simulated, both singularly and in combination. These are; sediment retention ponds, a restriction of the earthworks season with earthworks stabilisation in the off-season, and riparian grass buffer strips. 

Methodology

The results are derived in a 2-stage process.  The predictions from stage 1 enable a direct comparison of the impact of control measures, whilst the results from stage 2, weighted to account for the relative proportions of pasture and earthworks, enable impact assessment over the development area as a whole. Results from both stages are expressed as kg/ha of sediment loss but it is important to note the difference in their derivation since they yield markedly different values. 

 Stage 1:

Simulations are conducted at a 1-hectare scale for the dominant combination of existing landuse and soil, and mean slope angle (pasture, silt loam and 3°, respectively) within the Waiarohia development area.  GLEAMS is then re-run using the same soil type (but with the topsoil removed reflecting earthworks practice) and slope angle, with i) bare earth and ii) bare earth with sediment control measures. 

The effectiveness of three sediment control measures is assessed both individually and in combination. This assessment is made with respect to both individual storm events and annual averages.  These are:

• Sediment retention ponds
• Seasonal earthwork restrictions
• Riparian grass buffer strips

Stage 2:

Predicted sediment yields from stage 1 are combined based on the relative proportions of pasture and bare earth (97.7% and 2.3% respectively) across the Waiarohia development area. 

Input data

• Daily rainfall data was obtained from NIWA’s climate database for the Whenuapai airbase site from 1965 to 1992.
• Soils information for the study area was obtained from a database held by Landcare Research and, where required, interpreted for use in the model by Landcare Research
• Landuse, currently dominated by pasture was derived from aerial photographs provided by Auckland Regional Council.
• A 30 m resolution digital elevation model held by NIWA showed that 65% and 35% of land within the study area falls within the 0-3° and 3-6° slope categories respectively.
• Waitakare City Council (WCC) provided information describing future development in the Waiarohia study area

 The 2-stage modelling method

Outputs

In stage 1, eight scenario’s were developed and the predicted mean annual sediment loss predicted (kg/ha/year):

 Predicted mean annual losses (kg/ha/yr) and sediment control efficiency (% relative to bare earth alone) under each scenario

In stage 2, the methodology involved weighting the mean annual pastoral and bare soil losses according to the relative area they encompass under the worstcase stage of development (97.7% and 2.3%, respectively). This provided predictions of sediment loss across the Waiarohia development area:

Predicted mean annual losses (kg/ha/yr) and % increase (relative to Pasture only) under 1-hectare deelopment

Recommendations

The proposed rural residential developments in the Waiarohia are predicted to increase sediment loss relative to that under the existing pastoral landuse.  A 72% increase in average annual sediment load is predicted across the development area under the worst-case (no controls) scenario. This figure is lowered, however, through the incorporation of sediment control measures, and simulation of all three controls in combination is predicted to increase sediment loss across the area by 12%.

Under only one scenario incorporating a control measure (grass buffer) is an increased sediment loss of > 50% simulated. Retention ponds are predicted to be the most effective single control measure (67% efficiency), whilst seasonal restrictions to earthworks (31%) and a grass buffer (20%) have less impact.

The results of this study are dependent upon the density of ongoing earthworks: During the worst-case stage, ongoing development of 7 houses and 1.65 km of roads is expected for the 201-hectare area. The Waiarohia development area is characterised by a mean slope angle of about 3°, and is therefore relatively flat compared to some areas in the wider UWH catchment. However, the modelling results are broadly applicable to areas of proposed rural residential development across the UWH catchment, particularly those upon alluvial soil and flat or gently sloping land. The sensitivity of some of the predictions to slope angle is illustrated in Table 4, whereby sediment losses upon a 6° slope are substantially higher than those upon 3°.

Associated Models

References

Predicting sediment loss under proposed development in the Waiarohia catchment, Auckland Regional Council, Technical Publication 218, August 2003.  Prepared for Auckland Regional Council by Rob Collins, NIWA