The Impact of Dust and Increased Road Use on Wetlands in Western North Dakota
Jessica Creuzer, Christina Hargiss, Jack Norland, Tom DeSutter, Shawn DeKeyser, Frank Casey, Mike Ell
Objective • Compare wetlands in areas of increased travel
due to energy development with typical gravel road traffic (no energy development)
– Dust loading
– Water quality
– Trace element changes
in soil
– Vegetation differences
How Sites Were Chosen
• Restricted Randomization
• 10 High Impact sites – Road traffic mainly
from energy development
• 10 Low Impact sites – Typical road traffic
• All sites classified as seasonal
Selected Sites
High Impact Low Impact Weather Stations
Quantifying Dust
Quantifying Dust • Dust collectors set 10m, 40m and 80m from road
placed at cardinal directions
• Collected monthly
― 2012 July-October
― 2013 May-October
• Samples dried on NDSU campus
• Samples weighed
Water Quality • ND Department of Health protocols • Sampled monthly
– 2012 – July-September – 2013 – May-September
• On site – pH, dissolved oxygen, temperature, conductivity • Lab analysis – Major
cations/anions, total suspended solids, total organic carbon, Chlorophyll A/B, trace elements
Soil Sampling
• Soil sampled once a year (2012, 2013)
• Samples collected from 0-0.5 cm and 5-6 cm
• NDSU soil lab for bulk density (BD), pH and electrical conductivity (EC)
• All samples sent to ACME lab for 53 trace element analysis
• Developed by DeKeyser et al. (2003) and Hargiss et al. (2008)
• Evaluates health of Prairie Pothole Region (PPR) wetlands based on the plant community
• Developed for temporary, seasonal, semi-permanent wetlands in the PPR
Index of Plant Community Integrity (IPCI)
Low Prairie Wet Meadow Shallow Marsh
IPCI
• Creates a comprehensive species list
• 9 metric system
• Total score between 0-99
• Condition categories based on final score
North Dakota Rapid Assessment Model (NDRAM)
• Quickly assesses PPR wetlands based on plant and landscape characteristics (Hargiss 2009)
• Approximately 20 minutes to conduct survey
• Final scores on a scale of 0-100
• Groups wetlands based on final score
Hydrogeomorphic Model (HGM)
• Assesses the physical traits and functional characteristics of each wetland – Incorporates physical
characteristics, land-use information, soil data, biological data, and GPS and GIS information
– Calculates six Functional Capacity Indices (FCI) for each wetland
(Gilbert et al. 2006)
Statistical Analysis of Soil and Water
• Non-metric Multidimensional Scaling (NMS) as the ordination procedure
– Euclidian for water data
– Relative Euclidian for soil data
• Multi-Response Permutation Procedure (MRPP) for both water and soil samples
– Used to test if high impact and low impact and years were different
Plaza
Ross
Minot
Rainfall Totals (cm) for collection
months
0.00
5.00
10.00
15.00
20.00
25.00
April May June July Aug Sept Oct
2012
2013
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
April May June July Aug Sept Oct
2012
2013
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
April May June July Aug Sept Oct
2012
2013
Preliminary Results: Dust (2012 only)
• Loading at 10m is significantly different (p=0.04)
• Deposition rate ~0.6 lbs/m² per year
• 40m and 80m not significantly different
0
1
2
3
4
5
6
7
8
10m 40m 80m
Low Impact
High Impact
p=0.004
p=0.136 p=0.136
• One axis significant
― Represents 99% of variation in data
• Unclear if dust is affecting water quality
Preliminary Results: Water
-6 -4 -2 0 2
2012 HI 2012 LI 2013 HI 2013 LI
Total Dissolved Solids, Conductivity, SO₄ Hardness, Mg, Ca, Na
Preliminary Results: Soil
• No significant difference in depth
• Difference in EC and Sulfur
– Rainfall and landscape position likely most important factors
• Year impact is more telling than high or low impact or depth
Preliminary Results: IPCI
0
10
20
30
40
50
60
70
80
90
Low Impact
High Impact
p=0.246
• There are differences in condition at sites
– Not significant
Preliminary Results: NDRAM
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Low Impact
High Impact
p=0.001 • There are differences in
condition at sites
• NDRAM a more subjective measurement of condition
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Low Impact
High Impact
p=0.202 FCI 1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Low Impact
High Impact
p=0.721 FCI 2
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Low Impact
High Impact
FCI 3 p=0.756
Preliminary Results: HGM
•Low impact sites function better than high impact sites but not significant •More an effect of sites chosen, not necessarily from dust
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Low Impact
High Impact
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Low Impact
High Impact
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Low Impact
High Impact
FCI 4 p=0.054
FCI 5 p=0.078
FCI 6 p=0.014
Preliminary Results: HGM
•Low impact sites function better than high impact sites but not significant •More an effect of sites chosen, not necessarily from dust
Conclusion
• Preliminary results indicate:
– Dust is significant only at 10m
– Water quality and soil data most affected by
rainfall and landscape position
– Difference of condition and function, but not
significant between high and low impact sites
• Impact of site selection not dust
Moving Forward
• Still analyzing data
– Final results expected Fall of 2014
• Future research
– This information provides baseline data
• Future analysis could indicate change over time
– Other impacts (dust particle size, element analysis, etc)
• Focus within 40m distance from road
Acknowledgements
• NDDoH/EPA
• Mike Ell (NDDoH)
• Mike Hargiss (NDDoH)
• Kevin Horsager
• Larry Settler
• Landowners
• State of ND and ND Game & Fish
• Felix Fernando, Zach Sager, Kory Bonnell
Questions???