Linking Vegetation Dynamics with Physical Processes to ...Linking Vegetation Dynamics with Physical...

Preview:

Citation preview

Artemisia tridentataSalix exigua

Lepidospartum sq.

Populus fremontii

Salix laevigataArundo do.

Floodplain scrub

Floodplain herb.Salix lucida

Salix ex.-Arundo

Baccharis pilularis

Populus balsamifera

Salix lasiolepis

Mixed willowforest

Artemisia californica

Mixed willow scrub

Salix exigua-Baccharis salicifolia

Introduced herbaceous

Baccharis salicifolia

Mixed riparian scrubHIGH RELATIVE ELEVATION

INLAND

NOTRECENTLYFLOODED

GAINING

COASTAL

LOSING

RECENTLYFLOODED

LOW RELATIVEELEVATION

The lower Santa Clara River and its floodplain have been significantly altered by flood control, land use practices, and invasion by Arundo donax that have disrupted natural processes and resulted in riparian habitat loss. Despite these alterations, the lower river presents a unique opportunity to conserve and restore remaining riparian functions and ecosystems. To this end, the California State Coastal Conservancy is establishing the Santa Clara River Parkway by acquiring and enhancing floodplain habitat and flood-prone property.

This poster presents a subset of the analytical tools, understanding of vegetation dynamics and physical process linkages, and Arundo donax (giant

reed) treatment and other strategies developed and used to assist with acquiring, managing, and

restoring Parkway lands.

BACKGROUND

For more information, please visit:www.santaclarariverparkway.org

0

10

20

30

40

50

1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

Water Year

Prec

ipita

tion

(inch

es) Dry Years

Wet Years

a)

b)

0

10

20

30

40

50

1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

Water Year

Prec

ipita

tion

(inch

es) Dry Years

Wet Years

0

10

20

30

40

50

1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

Water Year

Prec

ipita

tion

(inch

es) Dry Years

Wet Years

a)

b)

a)

b)

0

25,000

50,000

75,000

100,000

125,000

150,000

175,000

200,000

Salix lasiolepis

Baccharis pilularis

Artemisia califo

rnica

Artemisia ca- Erio

gonum fasc

Mixed willo

w forest

Mixed willo

w scrub

Salix exigua - B

accharis salic

Mixed rip

arian scrub

Populus balsamifera

Mixed scrub

Non-native grasses a

nd forbs

Salix lucida

Eucalyptus s

pp.

Baccharis salicifolia

Salix exigua - A

rundo donax

Riverwash scrub

Riverwash herbaceous

Arundo donax

Mixed rip

arian forest

Salix laevigata

Floodplain wetland

Lepidospartum squamatum

Salix exigua

Artemisia tri

dentata

Populus fremontii

Tamarix spp.

Dist

ance

from

rive

r mou

th (f

eet)

Dis

tanc

e fr

om r

iver

mou

th (ft

)

� Treat arundo in the spring/summer following major floods to take advantage of the removed biomass. � Control arundo in an upstream to downstream direction and from tributaries. � Control arundo in and adjacent to fire-prone shrub lands. � In monospecific stands of arundo, or where it occurs with immature native vegetation, use more economical mowing techniques to remove biomass prior to treatment. �Use hand-removal techniques around mature native trees and patches of native vegetation to enhance ability of native vegetation to recruit naturally after treatment.

The vegetation dynamics-physical

process linkages described above, as well as recent research by Coffman et al. (2010) on

the ecology and impact of arundo on the lower Santa Clara River, provides guidelines for prioritizing arundo treatment projects:

Promote and implement sensible revegetation

The nature of flooding and vegetation reset along the lower river should determine the area where active planting is implemented, to avoid the loss of expensive active revegetation efforts. Where active planting is reasonable, the riparian dynamics analysis results can be used to inform select plan species that are most likely to be successful at a given site.

�Allow natural recruitment in the active floodplain.

�Focus active revegetation in areas not subject to frequent, resetting floods.

�Actively revegetate sites following arundo removal.

Differences between the coastal fog belt and the arid inland portions of the watershed are at least partly responsible for the correlation between distance from river mouth and plant species distributions, although arundo occurs throughout the watershed.

Riparian vegetation along the lower Santa Clara River is sub-ject to infrequent but dramatic resets during large flood events, particularly during El Niño-Southern Oscillation

(ENSO) events, which are characterized by relatively high rainfall and, as a result, high flood magnitudes.

(LEFT) Riparian vegetation conditions before (June 2002), immediately after (February 2005), and nine months following the 2005 high-flow event (September 2005). (RIGHT) Average time since last flood (± 1 SE) of vegetation alliances in the analysis area.

Time since last flood and relative elevation above the channel were two of the strongest correlates with vegetation alliance distribution. This sug-gests that the extent and distribution of riparian vegetation is closely linked to the dramatic scouring of vegetation during ENSO cycles, and the progression of seral stages thereafter. Arundo is more common on surfaces that have been flooded within the past 40 years.

Vegetation Distribution CorrelationsCanonical correspondence analyrsis (CCA) was used to identify the environ-mental variables most affecting the distribution of vegetation alliances (list-ed below). Univariate analysis was then used to examine differences in re-lationships based on the most important environmental variables identified through CCA. This analysis identified linkages between physical processes and the establishment and maintenance of riparian vegetation and assisted in predicting how riparian vegetation may establish and persist on restored floodplains within the study area.

Linking Vegetation Dynamics with Physical Processes to Develop Arundo Control and Riparian Restoration Strategies for a Semi-arid River and its FloodplainZooey Diggory1*, Bruce Orr1, Amy Merrill1, Gretchen Coffman2, William Sears1,3, and Peter Brand4

INFORMED ARUNDO

TREATMENT and

RESTORATION STRATEGIES

ANALYTIC AL TOOLS

Accurate and current mapping of riparian vegetation, including invasive species, was a crucial first step in river corridor manage-ment and restoration planning. This project followed the State of California standard vegetation classification system, mapping to the vegetation alliance level, and included focused percent-cover mapping for invasive arundo and tamarisk.

Changes in the extent of riparian veg-etation along the lower Santa Clara River were quantified by mapping highly vegetated, partially vegetated and scoured areas on overlays of aer-ial photographs taken soon after six major floods.

01,0002,0003,0004,0005,0006,0007,0008,0009,000

10,00011,00012,00013,00014,00015,000

1938 1969 1978 1992 1995 2005

Scoured Channel BedPartially Vegetated FloodplainHighly Vegetated Floodplain

0

10

20

30

40

50

60

70

80

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

160,000

180,000

Percent VegetatedFlood Magnitude (cfs)

Acre

sPe

rcen

t Veg

etat

ed

Flood Magnitude (cfs)

Year

01,0002,0003,0004,0005,0006,0007,0008,0009,000

10,00011,00012,00013,00014,00015,000

1938 1969 1978 1992 1995 2005

Scoured Channel BedPartially Vegetated FloodplainHighly Vegetated Floodplain

0

10

20

30

40

50

60

70

80

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

160,000

180,000

Percent VegetatedFlood Magnitude (cfs)

Acre

sPe

rcen

t Veg

etat

ed

Flood Magnitude (cfs)

Year

Year Flow (cfs) Flood Recurrence Interval

1938 120,000 14 year

1969 165,000 24 year

1978 102,200 11 year

1992 104,000 12 year

1995 110,000 13 year

2005 136,000 16 year

VegetationMapping

Climate Flood Dynamics

Arundo has invaded, and in many areas replaced, native riparian vegetation in the watershed. In addition to reducing riparian habitat quality and quantity, accumulated arundo biomass increases the susceptibil-ity of the lower Santa Clara River riparian corridor to fire, and post-fire conditions further promote arundo growth rates by elevating nutrient levels and eliminating competition from native species (Coffman et al. 2010).

Floodplain DevelopmentOur historical flood mapping demonstrates the dramatic effect of le-vees and development in constraining the floodplain and limiting ri-parian vegetation. In the lowest reaches of the river where levees are most extensive, the riparian corridor has been reduced by nearly 70% compared with conditions in 1938.

Plant communities differ between gaining and losing groundwater reaches, with species such as Salix (willow) and Populus (cottonwood) occurring more frequently in gaining reaches, and Artemisia tridentata (sagebrush) and Lepidospartum squamatum

(scalebroom) occurring more frequently in drier, losing reaches. Arundo is established and survives in both gaining and losing reaches.

VEGE TATION-PROCESS LINK AGES

Preventing development in the floodplain and increasing floodplain width and connectivity with the river will ameliorate some of the most significant impacts on the lower Santa Clara River and promote a greater extent of native riparian vegetation.

Increase and improve floodplain connectivity

Invasion by Arundo

Implement strategic actionsto control arundo

0

10

20

30

40

50

60

Floodplain wetland

Riverwash scrub

Mixed willo

w scrub

Salix exig

ua - Arundo donax

Salix laevig

ata

Mixed willo

w forest

Arundo donax

Salix lucida

Baccharis salicifolia

Populus fremontii

Salix lasiolepis

Populus balsamifera

Salix exig

ua

Artemisia

californica - E

riogonum fasc

Mixed rip

arian forest

Lepidospartum squamatum

Tamarix spp.

Mixed rip

arian scrub

Salix exig

ua - Baccharis salic

Non-native grasses a

nd forbs

Artemisia

californica

Artemisia

tridentata

Mixed scru

b

Baccharis pilularis

Eucalyptus

Alliance

Time s

ince

last

floo

d (yr

s with

SE

bars

)

(BELOW) Flood events with corresponding aerial photographs used to examine changes in the extent of riparian vegetation in the lower Santa Clara River.

Historical Changes inRiparian Vegetation Extent

Riparian corri-dor conditions at river mile 11 in 1938 and 2005.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Mixed s

crub

Populus

balsa

mifera

Salix l

asiol

epis

Populus

fremon

tii

Salix e

xigua -

Arundo

dona

x

Mixed r

iparian

scrub

Mixed w

illow sc

rub

Euca

lyptus

Salix l

aevig

ata

Bacchari

s pilu

laris

Riverw

ash s

crub

Salix l

ucida

Riverw

ash h

erbace

ous

Flood

plain

wetlan

d sup

erallia

nce

Mixed w

illow fo

rest

Salix e

xigua -

Bacch

aris s

alicif

olia

Arundo

dona

x

Bacchari

s salic

ifolia

Non-nati

ve gr

asse

s and f

orbs

Artemisi

a cali

fornic

a

Mixed r

iparian

fores

t

Salix e

xigua

Artemisi

a trid

entat

a

Lepido

spart

um sq

uamatum

% o

f plo

ts w

ithi

n ga

inin

g or

losi

ng re

ache

s

% occurring in losing reaches

% occurring in gaining reaches

Percentage of plots within each vegetation alliance that occurs in gaining (blue) versus losing (orange) reaches. Overall, 54% of the reaches (red line) in the 1,490 alliance vegetation plots were in gaining reaches.

Extent of the 1938 and 2005 floods.

Reach 6 enlarged

The extent and density (percent cover) of arundo infestations in the lower river.

Groundwater Availability

(BELOW) a) Distribution of mean annual precipitation based on data from the period 1900–1960; b) Precipitation records indicate peri-ods of cumulatively wetter (blue) and drier (orange) periods in the watershed.

Distribution of gaining and losing reaches in the lower Santa Clara River watershed.

Land cover types within the lower Santa Clara River watershed.

Example of the detailed vegeta-tion mapping used for the lower Santa Clara River.

The Santa Clara River watershed. (BELOW) Area of vegetated

and unvegetated floodplain affected by selected floods.

Average longitudinal position (± 1 SE) of vegetation alliances in the analysis area.

�Acquire land for conservation through purchase or easements.

�Remove or stop repairing unnecessary berms and levees.

�Replace bank-edge levees with setback levees.

1Stillwater Sciences, 2855 Telegraph Avenue, Suite 400, Berkeley, California 94705; 2Riparian Invasive Research Laboratory, Marine Science Institute, University of California, Santa Barbara, CA 93106-6150; 3San Francisco Public Utility Commission, 1145 Market Street, 4th Floor, San Francisco, California 94103; 4California Coastal Conservancy, 1330 Broadway, 11th Floor, Oakland, California 94612; *Send correspondance to zooey@stillwatersci.com

Species distribution and vegetation reset Species distribution Species distribution and vegetation structure

Vegetation extent Native vegetation extent

The four physical variables most significantly correlated with vegetation types from the CCA.

Recommended