Assessing the Impact of Culvert Design on Three Ecosystem Functions in Northern

Wisconsin Streams James Olson1, Amy Marcarelli1, Sue Eggert2, Anne Timm2,

Randy Kolka2 1Michigan Technological University

2USDA Forest Service

Stream Restoration • The most common restorations try to rehabilitate

hydrologic connectivity or habitat heterogeneity • Restoration projects are most often evaluated using

ecosystem structure measurements

http://www.fws.gov/mountain-prairie/pfw/r6pfw2h.htm http://peakwater.org/tag/dam-removal/

Ecosystem Structure vs. Function

Ecosystem structures • Richness • Biomass • Abundance

Ecosystem functions • Production • Respiration • Nutrient uptake

“Ecosystem processes are the transfers of energy and materials from one pool to another” Chapin et al. 2011

Lindemann 1942

Culvert Issues • Culverts can be barriers to

aquatic organism movement

• Poorly designed culverts can alter natural stream flow and the balance between erosion and sedimentation

Photo Credits: S. Eggert

Culvert Designs

Full Stream Simulation Partial Stream Simulation

Non-Replaced

• Stream simulation culverts are designed to improve hydrologic characteristics, aquatic organism passage, and ecosystem functions

Objective

• To evaluate the effectiveness of stream simulation culvert designs for restoring or maintaining ecosystem functions

Two Comparisons • Full Stream Simulation vs. Partial Stream Simulation

– 6 streams (3 Full and 3 Partial) – 3 ecosystem functions

• Stream Simulations vs. Non-Replaced – 15 streams (5 Full, 6 Partial, 4 Non-replaced) – 2 ecosystem functions

Three Ecosystem Functions • Hydrologic Connectivity

• Nutrient Uptake • Coarse Particulate Organic

Matter (CPOM) Retention

Full Stream Sim vs. Partial Stream Sim Hypotheses: • Nutrient uptake velocity will be more similar upstream and

downstream of full stream simulation culverts compared to partial stream simulation culverts

• Hydrologic connectivity and CPOM retention will be more

similar upstream, through and downstream of full stream simulation culverts compared to partial stream simulation culverts

Hydrologic Connectivity

Runkel 1998

• Transient storage characteristics estimated with One Dimensional Transport with Inflow and Storage (OTIS) Model

α As

D A

Hydrologic Connectivity Average Velocity

Upstream Through Downstream

Ave

rage

Vel

ocity

(m/s

ec)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Full Stream SimPartial Stream Sim

P=0.23

Transient Storage Characteristics

Upstream Downstream

Dis

pers

ion

(m2 s

-1)

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Full Stream Sim Partial Stream Sim

P=0.03

Upstream Downstream

Stor

age

Exch

ange

C

oeffi

cien

t (

, s-1

)

-0.001

0.000

0.001

0.002

0.003

0.004

0.005

Full Stream SimPartial Stream Sim

P=0.13

Upstream Downstream

Stor

age

Zone

Cro

ss

Sect

ion

Are

a (m

2 )

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Full Stream SimPartial Stream Sim

P=0.17

Upstream Downstream

As/A

0.00

0.05

0.10

0.15

0.20

0.25

0.30

Full Stream SimPartial Stream Sim

P=0.14

Hydrologic Connectivity Dispersion

Upstream Downstream

Dis

pers

ion

(m2 s

-1)

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Full Stream Sim Partial Stream Sim

P=0.03

Nutrient Uptake

Newbold et al. 1982

Nutrient Uptake

Above CulvertBelow CulvertAv

g. U

ptak

e Ve

loci

ty (m

m/s

ec)

0.00

0.05

0.10

0.15

0.20

0.25

Full Stream SimPartial Stream Sim

P=0.38

Upstream Downstream

CPOM Retention

Upstream Through Downstream

Ret

entio

n Ve

loci

ty (m

m/s

ec)

0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

Full Stream SimPartial Stream Sim

CPOM Retention

P=0.003

Summary of First Comparison

• Limited differences between full and partial stream simulations (CPOM, Dispersion)

• Is either stream simulation design better than non-replaced culverts?

Stream Simulation vs. Non-Replaced Hypothesis: • Hydrologic connectivity and CPOM retention will be

most similar upstream and through full stream simulation culverts, less similar for partial stream simulation culverts and least similar for non-replaced culverts

Hydrologic Connectivity

Average Velocity

Upstream Through

Ave

rage

Vel

ocity

(m/s

ec)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Full Stream SimPartial Stream SimNon-Replaced

P=0.03

Upstream Through

Ret

entio

n Ve

loci

ty (m

m/s

ec)

-0.002

-0.001

0.000

0.001

0.002

0.003

0.004

0.005

0.006

Full Stream SimPartial Stream SimNon-Replaced

P=0.004

CPOM Retention

CPOM Retention

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

Upstream Through Above Through Upstream Through

Perc

ent o

f Rel

ease

d Le

af A

nalo

gs

% Debris Dam

% Wood

% Bank

% Backwater

% Roots

% Rocks

Non-replaced Partial Full

Summary • Hydrologic connectivity and nutrient uptake were

not different between culverts or between reaches (Dispersion*)

• CPOM retention decreased drastically through all designs but least through full stream simulations

Conclusions

• Major hydrologic issues are fixed by stream simulation designs but very little difference between them was observed in ecosystem functions

• Culverts with severe problems could not be

represented in this study

• Ecosystem functions that can be measured at small scales are more likely to respond to small culvert restorations (CPOM)

Acknowledgments Field and Lab Support • Ashley Coble • Jade Ortiz • Kellie Heiden • Brian Borowicz