Assessing Constructed Forested Wetland Development Using Successional (Performance) Trajectories Susan M. Carstenn Hawaii Pacific University Kaneohe, Hawaii.

Assessing Constructed Forested Wetland Development

Using Successional (Performance) Trajectories

Susan M. Carstenn Hawaii Pacific University Kaneohe, Hawaii

Why Reference Wetlands?

• Restoration Strategies• Endpoint…target….goal• Assessment metrics

• Reference Wetlands• minimally adversely affected by

anthropogenic activities• the wetland being destroyed as in the

case of mitigation• old restoration project that has been

deemed successful

Constructed Forested Wetland

Constructed Forested Wetland

Natural Forested Wetlands

Why Trajectories?

• Wetlands are dynamic ecosystems; therefore, static metrics are inappropriate assessment tools.

• Reference wetlands serve as the target, but can not assess incremental progress towards the target.

• Trajectories provide incremental targets.

Reference Wetland Approach

• The performance of a constructed wetland (stars) is compared to the mean ( standard error(s)) of a set of reference wetlands (green).

Time

Qua

lity

or Q

uant

ity

Reference Wetland Approach

• Success is declared when the constructed wetland (stars) meets or exceeds the reference wetlands.

• Wetlands developing at different rates are assessed with the same criteria.

• The time required to replace wetland function is inconsequential. Time

Qua

lity

or Q

uant

ity

Trajectory Approach

• A constructed wetland (star) is compared to the mean of all previously constructed wetlands (green).

• Success is declared when a newly constructed wetland (stars) falls within the 95% predication intervals around the mean of all previously constructed wetlands (green).

Qua

lity

or Q

uant

ity

Time

Trajectory Approach

• A wetland demonstrates suboptimal development…then what?

• Success is declared when a newly constructed wetland (stars) falls within the 95% predication intervals around the mean of all previously constructed wetlands (green).

Time

Qua

lity

or Q

uant

ity

The orange line represents the timing of a redemediation action e.g., supplemental planting, fertilization, or understory seeding.

Trajectory Construction

• Space for Time Substitution• A chronosequence of wetlands• Array of metrics measured once

• Individual Wetland Trajectories• Many newly constructed wetlands• Metrics monitored over time

Assessment Metrics

• Canopy Trees• Height• Diameter at breast height

(dbh)• Size class distributions

(dbh)• Community basal area• Canopy cover• Stem density• Species richness• Species diversity

• Subcanopy Trees and Shrubs• Stem density• Diameter at breast height• Species richness• Species diversity

• Herbaceous Species• Species richness• Species diversity

• Understory Species• Canopy and subcanopy

seedling richness and frequency of occurrence

• Vine species richness and frequency of occurrence

• Functional group richness and frequency of occurrence

• Soils• Soil water content• Bulk density• Organic Matter• Particle size analysis

Assessment Metrics

• Emerging Properties• Hierarchical size class frequency distributions of tree diameters• Changes in frequency of occurrence of vegetation structural categories with age• Increasing organic matter associated with increases in soil water content• Decreasing bulk density associated with increasing organic matter content

Assessment Metrics

• Emerging Properties• Hierarchical size class frequency distributions of tree diameters• Changes in frequency of occurrence of vegetation structural categories with age• Increasing organic matter associated with increases in soil water content• Decreasing bulk density associated with increasing organic matter content

0

2

4

6

8

0 5 10

Soil Water Content

Org

an

ic M

atte

r C

on

ten

t

0

2

4

6

8

0 5 10

Soil Water ContentO

rga

nic

Ma

tter

Co

nte

nt

0

2

4

6

8

0 5 10

Soil Water Content

Org

an

ic M

atte

r C

on

ten

t

Results – Individual Metrics

• Canopy Trees• height (r2 = 0.81; p < 0.05)• diameter at breast height (r2 = 0.80; p <

0.05)• stand basal area (r2 = 0.75; p < 0.05)• canopy cover (r2 = 0.77; p < 0.05)

• Subcanopy Trees (Ilex cassine)• diameter at breast height (r2 = 0.64; p <

0.05)• stand basal area (r2 = 0.75; p < 0.05)

• Shrubs and Understory• no significant trajectories

Hydrogeomorphic Classes

• Depressional• isolated• riparian• headwater

• Lake Fringing• littoral

• Stream Floodplain• bordering an

incised channel www.geog.psu.edu/wetlands/manual/image13.gif

Results – Individual Metrics by

Hydrogeomorphic Class

Metric Combined DepressionalLake

FringingSteam

Floodplain

Canopyheight 0.81 0.88 0.92 0.94

dbh 0.80 0.89 0.96 0.93basal area 0.75 0.62 0.92 0.83

cover 0.77 0.84 0.77 0.93

y = 0.22x1.29

r2 = 0.94

02468

1012

0 5 10 15 20

Age (years)

Heigh

t (m)

y = 1.02e0.14x

r2 = 0.81

0

2

4

6

8

10

12

14

16

0 5 10 15 20

Age (years)

Hei

ght (

m)

Results – Individual Metrics by

Hydrogeomorphic Class

Metric Combined DepressionalLake

FringingSteam

Floodplain

Understoryunderstory richness 0.49 0.76 0.71 0.31understory diversity 0.57 0.94 0.59 0.31

shrub richness 0.38 0.48 0.80 0.30vine richness NS NS NS 0.90

grass, sedge, rush 0.62 0.88 0.54 0.50

herbaceous richness 0.57 0.82 0.79 0.43woody richness NS NS NS 0.69

woody:herbaceous 0.57 0.31 0.57 0.73% vine 0.58

% grass 0.47% grass, sedge, rush 0.41

% woody 0.44% herbaceous 0.56

y = -0.04x + 1.63

r2 = 0.94

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0 5 10 15 20 25

Sh

an

no

n-W

eav

er

Div

ers

ity

Results – Emerging Properties

03A

0

0.10.2

0.30.4

0.5

0.60.7

0.80.9

1

A B C D E F G H I J K L MF

requ

ency

05C

00.10.2

0.30.40.50.60.7

0.80.9

1

A B C D E F G H I J K L M

Fre

quen

cy

10A

00.10.20.30.40.50.60.70.80.9

1

A B C D E F G H I J K L M

dbh Size Class

Fre

quen

cy

15C

00.10.20.30.40.50.60.70.80.9

1

A B C D E F G H I J K L M

dbh Size ClassF

requ

ency

Size Class Lower Bound Upper BoundA 0 1.0B 1.01 2.5C 2.51 5.0D 5.01 10.0E 10.01 15.0F 15.01 20.0G 20.01 25.0H 25.01 30.0I 30.01 35.0J 35.01 40.0K 40.01 45.0L 45.01 50.0M 50.01 55.0

dbh Size Class (cm)

Natural Wetland

Results – Emerging Properties

y = 0.33x - 0.04

r2 = 0.46p 0.01

0%

20%

40%

60%

80%

100%

0% 20% 40% 60% 80% 100%

Water Content (%)

Org

anic

Mat

ter

(%)

y = 0.02e5.62x

r2 = 0.89p << 0.01

0%

20%

40%

60%

80%

100%

0% 20% 40% 60% 80% 100%

Water Content (%)

Org

anic

Mat

ter (

%)

y = 1.36e-3.35x

r2 = 0.31p < 0.01

0.0

0.5

1.0

1.5

2.0

0% 20% 40% 60% 80% 100%

Organic Matter Content (%)

Bu

lk D

ensi

ty

y = 1.17e-2.97x

r2 = 0.98p << 0.01

0.0

0.5

1.0

1.5

2.0

0% 20% 40% 60% 80% 100%

Organic Matter Content (%)

Bul

k D

ensi

ty

r2 = 0.81

0

0.5

1

1.5

2

2.5

3

0% 20% 40% 60% 80% 100%

Organic Matter Content (%)

Bul

k D

ensi

ty

r2 = 0.95

0%

20%

40%

60%

80%

100%

0% 20% 40% 60% 80% 100%

Water Content (%)

Org

anic

Mat

ter (

%)

18 Years Old Natural Wetland5 Years Old

Confidence Intervals Prediction Intervals

Discussion

• Challenges• Selecting data collection methods

to support comparisons• across developmental stages• with literature values

Discussion

• Canopy trees• height

• approach values of natural wetland• fall short of literature values

• dbh• sites > 12 years old met or exceeded literature values

• stand basal area• only the oldest sites approached literature values

• Subcanopy trees• Myrica cerifera andSalix caroliniana similar to

literature values • Ilex cassine and Persea palustris fall below

literature values

Discussion

• Shrubs• lower species richness• similar frequency, dbh, and density

• Understory• similar to richness of cypress domes and

bayheads, but less than hardwood swamps and marshes.

• Soil• weak, but significant trends in organic

matter

Conclusions

• Canopy tree development alone may not indicate restoration success; it suggests that the site is developing conditions indicative of a mature forest.

• Subcanopy, shrub, and understory• community structure is approaching those

of wetlands described in the literature. • richness and diversity is on the low end of

the range of reported in the literature.

Conclusions

• The trajectory approach shows great promise. • In spite of the added variation inherent

in the space-for-time substitution, highly significant trends were detected.

• Uniform data collection methods should be established.

• How much time should be allocated for a wetland to meet a particular assessment criterion?