Application of ArcSWAT at 5 th Canadian Division Support Base - - PowerPoint PPT Presentation

Application of ArcSWAT at 5th Canadian Division Support Base Gagetown

Mike Hulley1, Kela Weber2, Nick Jewitt2, et Andy Smith3

1Departement of Civil Engineering, Royal Military College of Canada 2Environmental Sciences Group, Royal Military College of Canada 35th Canadian Division Support Base Gagetown, National Defence

2016 Real Property Institute of Canada (RPIC) Federal Contaminated Sites Workshop, Montréal (Québec)

Gagetown Training Area

Study Area

Background

• Located in south central N.B and established in 1950’s
• Home to the Army’s Combat Training Centre
• 21,000 ha of manoeuvre areas and 30,000 ha of impact

areas, 830 km of roads, 360 km of tracks, ~500 fords, ~1,200 in-stream culverts and bridges.

• 3,300 km of watercourses, ~160 lakes or ponds, ~6,500 ha
• f wetland. Atlantic salmon and brook trout are key

species.

Impacts

• 5 year, $50 million Sedimentation and Erosion Control

Program (SECP) focusing on the improvement and decommissioning of roads, tracks, fords, other water crossings and re-vegetation of barren soils. Associated works includes stream restoration and wetland creation.

Current Control Efforts

Model goals

• Model flow and suspended sediments in the

Nerepis River.

• Assess the effectiveness of Sedimentation and

Erosion Control Program and target restoration. Carried out in conjunction with a benthic monitoring program for Gagetown

Overview

• ArcSWAT model development
• Hydrologic flow and water quality calibration and

validation for the Nerepis River.

• Assessment of current mitigation strategies.
• Preliminary development and assessment of

alternative strategies.

ArcSWAT Model

• Based on Soil Water Assessment Tool.
• Basin‐scale, continuous model, is designed to

predict the impact of management on water, sediment, and agricultural chemical yields in ungauged watersheds.

• Comprehensive surface, subsurface, and in-stream

modelling routines included,

• Public domain, well supported and documented,

and fully integrated into ArcGIS.

Key Model Inputs

• Soil, land cover, and topography.
• Characteristics of Nerepis River and associated

tributaries

• Meteorological information.
• Flow and water quality monitoring to support

calibration and validation.

• Sub-basin delineation.

Major Subbasins

QB SB RG KB NR1 NR2

River George Queens Brook Mountain Brook Sucker Brook Kerr Brook

6.5 13 3.25 Kilometers

Wet weather monitoring Continuous flow and wet weather monitoring

KB QB

Previous Kerr Brook Model

Land Cover Map

Soil Classes Map

6.5 13 3.25 Kilometers

Gagetown Soil Classes

Classes

ZZ CanterburyShalyLoam CarletonSiltyClay DeedGravellyLoam GagetownGravellySandyLoam IntervaleSandyLoam KingsLoam LomondGravellyLoam LornevilleSiltyClayLoam MidlandGravellySandyLoam OromoctoSandyLoam ParleevilleGravellySandyLoam Peat QueensLoam RiverbankSandyLoam SunburyGravellySandyLoam TracyLoam WaasisClay Basin

General Approach

Sensitivity Analysis

• Vary typical sensitive parameters to determine

effect

• Determine sensitive parameters through p-value

based on Student’s t-distribution

• Calibrate sensitive parameters within acceptable

ranges

• Leave non-sensitive parameters as typical values

Priority ¡ Description ¡

1 Snow melt base temperature ¡ 2 Initial SCS runoff curve number for moisture condition II ¡ 3 Groundwater evaporation coefficient ¡ 4 Threshold groundwater depth for baseflow ¡ 5 Snow pack temperature lag factor ¡ 6 Saturated hydraulic conductivity ¡ 8 Plant ET coefficient ¡ 9 Surface runoff lag coefficient ¡ 10 Maximum canopy storage ¡ 11 Depth of soil layer ¡ 1 USLE “C” (Cover Factor) ¡ 2 USLE “K” (Soil Erodibility Factor) ¡

Sensitive Parameters

Hydraulic Calibration

Nash-Sutcliffe = 0.65

Hydraulic Validation I

Nash-Sutcliffe = 0.68

Hydraulic Validation II

Hydraulic Validation (daily)

Hydraulic Validation (daily)

Water Quality

TSS vs Turbidity

Water Quality Calibration

• Large variability in monitoring results for

synoptic TSS and Turbidity.

• Preliminary regression relation developed based
• n Nerepis River event monitoring results.
• USLE soil and land cover coefficients adjusted in
• rder to achieve a reasonable agreement.

Max ¡TSS ¡ (mg/L) TS ¡(Tons) Max ¡TSS ¡ (mg/L) TS ¡(Tons) 2009 184 11791 209 11798 12.2 2010 410 10634 281 13262 11.4 2011 205 9736 232 9684 10.6 2012 300 12787 219 8665 8.2 Monitoring ¡Results ArcSWAT Year AverageQ ¡ (m3/s)

Water Quality Calibration

• Reasonable

agreement on an event basis.

• Considerable

uncertainty remains.

Event ¡Date TSS ¡Observed ¡ (24-­‑hour ¡ average ¡mg/L) TSS ¡Predicted ¡ ArcSWAT ¡ (mg/L) 25-­‑Oct-­‑09 51 56 29-­‑Sep-­‑09 80 111 15-­‑Nov-­‑09 63 70 22-­‑Jul-­‑10 247 232 15-­‑Oct-­‑10 64 94 5-­‑Nov-­‑10 133 76 28-­‑Aug-­‑11 140 113 20-­‑Oct-­‑11 46 48 11-­‑Nov-­‑11 85 60 5-­‑Sep-­‑12 120 120 31-­‑Oct-­‑12 191 69

Preliminary Mitigation Assessment

• Two illustrative scenarios assessed:
• 1. 25% of Barren land cover replaced with

range land.

• 2. 100% of Barren land cover replaced with

range land.

Scenario Loading ¡ (Tons) Reduction ¡ (%) Baseline 5700 0% 25% ¡Control 5100 11% 100% ¡Control 3480 64% Annual ¡Average ¡Total ¡Solids ¡Loading ¡ (Tons) ¡12-­‑year ¡continuous ¡simulation ¡

Preliminary Mitigation Assessment

Next Steps

• Growing database of monitoring results addressing the

relationship between TSS and turbidity, however, more required.

• Assessment of seasonal and spatial variability in TSS and

turbidity relationships.

• Site specific mitigation strategies need to be developed

and critically assessed.

• Need to differentiate between stream bedload and
• verland erosion and sediment transport.
• Continue to refine and improve ArcSWAT assessment

tool.

Conclusions

• Preliminary model calibration and validation completed

for flow.

• Preliminary calibration for TSS achieved, for individual

events and monthly loads, however, additional work remains (seasonal and spatial variability).

• Long term model run demonstrates that recent mitigation

measures have not significantly changed average flow conditions.

• Preliminary assessment of broad land-cover changes

demonstrate that significant reductions in suspended solids loadings are achievable with targeted land cover improvements.

Acknowledgements

• 5th Canadian Division Support Base Gagetown

(project sponsor)

• Royal Military College of Canada- Environmental

sciences Group (scientific and administrative support)

• Environment and Climate Change Canada(Flow

and water quality data)

• Fisheries and Oceans Canada (Flow and water

quality data)