Identifying Zebra Mussel Distribution in the Hudson River Estuary Using High-Resolution Geophysical Data
Angela L. Slagle*, W.B.F. Ryan, F.O. Nitsche, S.M. Carbotte, and R.E. Bell
Lamont-Doherty Earth Observatory, 61 Route 9W, Palisades, NY 10964, USA
ABSTRACT
Zebra mussels (Dreissena polymorpha), an invasive species, have fluorished
and continue to spread in lakes and rivers of the northeast U.S. since their 1980s
introduction. Documented effects, including declining phyto- and zooplankton and
dissolved oxygen, increasing benthic primary production and water clarity, and
damage to water intakes, reveal that this species plays an important role in
ecosystems. Studies show that zebra mussel distributions change through time,
influenced by sediment properties as well as biological and chemical parameters.
In the Hudson River Estuary, a dataset including multibeam bathymetry, sidescan
sonar, subbottom profiles, and over 1,000 sediment cores and grabs, acquired
as part of the Hudson River Benthic Mapping Project funded by the New York
State Dept. of Environmental Conservation, allows us to characterize existing zebra
mussel habitats. We identify zebra mussels from the northern extent of our study at
the Troy Dam south to Croton Point, where summer salinity range is 10-13 ppt.
Zebra mussels primarily colonize coarse-grained substrates, but are rarely found in
depositional or dynamic areas where we find mobile sediment waves. We use our
process-based analysis of the riverbed that incorporates geophysical data with
sediment samples and GIS software to map substrates threatened by future
colonization, providing a useful tool for estuary managers.
KINGSTON-SAUGERTIES AREA
In the Kingston-Saugerties stretch of the estuary, we use ArcGIS software
to create a Geographic Information System (GIS) that incorporates sidescan
sonar data and multibeam bathymetry with sediment samples, including grain
size information and descriptive characteristics for each sample. GIS systems
offer many possibilities to display and compare data and provide a valuable
environment for detailed interpretation of our high-resolution data. We produce
maps of sediment type by incorporating point samples and associated grain
size information with sidescan sonar data (Fig. 1, 2). Using all available data,
we develop maps of process-based sedimentary environment, classifying the
estuary bottom as depositional, erosional/nondepositional, or dynamic (Fig. 3).
We use a similar method and ArcGIS analysis tools to investigate the
Sample ID
distribution of zebra mussels in the Kingston-Saugerties area.
LW3-G31
LW3-G32
LW3-G33
LW3-G34
LW3-G35
LW3-G36
LW3-G37
* contact: [email protected]
Using ArcGIS Spatial Analyst, we can analyze all data types to determine
which parameters are most often associated with the presence of zebra
mussels in this region of the estuary. We combine sediment type and
environment maps to generate polygon coverage of the area and overlay
point sample data, with zebra mussel presence/absence information for
each point (Fig. 4, 5). We join this with sidescan sonar mosaics and
multibeam bathymetry maps. For each sample site, we now have information
from all data types: presence/absence of zebra mussels, distance from shore,
water depth, mean sidescan value for 20m radius around each point,
dominant sediment type and sedimentary environment (Table 1).
Zebra mussels
0
0
1
1
1
0
1
Dist. to shore (m)
Water depth (m)
Mean sidescan
343.13580515300
-6.30654001236 215.57099914551
437.85240295200
-10.68900012970 181.80099487305
-18.10519981384 171.66600036621
171.38350073100
84.64399342630
-18.82419967651 109.48600006104
401.55286750200
-15.47239971161 100.92099761963
-12.63389968872 135.91000366211
432.31422187100
242.33025107400
-10.74740028381 150.84800720215
Sed. Type
muddy sand
sand
sand
muddy sand
gravelly sand
gravelly sand
gravelly sand
Sed. Environment
Dynamic - waves
Dynamic - waves
Dynamic - scour
Dynamic - scour
Dynamic - debris
Dynamic - waves
Dynamic - debris
OTHER AREAS OF HUDSON RIVER ESTUARY
The Hudson River Benthic Mapping Project includes geophysical and sediment
sampling data for the entire Hudson River Estuary, from the New York Harbor to the
Troy Dam. We find zebra mussels in most of the freshwater areas of the system.
Although these areas have not yet been completely analyzed with respect to zebra
mussels, we have developed sediment type and environment maps and compared
visually with point samples. In the New Baltimore and Stockport Flats areas to the
north, zebra mussel presence often coincides with areas of high sidescan backscatter
(Fig. 6, 7). Further south in Newburgh Bay, we find live zebra mussels on coarse
deposits that may be associated with tributary input (Fig. 8) and in dynamic areas
where we believe the estuary floor is influenced by the Newburgh-Beacon bridges (Fig. 9).
73°48'0"W
73°46'0"W
73°50'0"W
New Baltimore
73°46'0"W
42°34'0"N
42°34'0"N
Table 1. Example showing the results of combining the point sample data and shoreline polygon (Fig. 4) with
sidescan sonar (Fig. 1), multibeam bathymetry, and sediment type and environment polygon data (Fig. 5). Each
sample site is now associated with a specific set of attributes from all data types.
73°48'0"W
Stockport Flats
Sidescan Backscatter
42°22'0"N
42°22'0"N
High
Low
0
1
2
Kilometers
Sidescan Backscatter
Backscatter
Sidescan
Highbackscatter: 255
Low
HIGH BACKSCATTER
73°57'0"W
73°56'0"W
73°55'0"W
Low backscatter: 0
High
42°32'0"N
42°32'0"N
73°58'0"W
73°54'0"W
0
42°20'0"N
42°4'0"N
42°30'0"N
42°30'0"N
+
live zebra mussels
=
+
42°2'0"N
Tivoli Bay
42°2'0"N
Flat
Sample LW3-G33
Zebra mussels
171 m from shore
18 m water depth
Mean sidescan 172
Sand
Dynamic scour
und
42°3'0"N
no zebra mussels
42°16'0"N
73°48'0"W
Deposition - thin
Deposition - over bedrock
Deposition - unres. thickness
Dynamic - waves
no zebra mussels
Dynamic - lineation
42°0'0"N
42°0'0"N
live zebra mussels
74°W
sandy mud
Dynamic - debris
gravelly mud
Dynamic - scour
muddy sand
Dynamic - slump
sand
Dynamic - streaks
gravelly sand
Erosion - bedrock
muddy gravel
Erosion - truncated
sandy gravel
gravel
41°38'0"N
Figure 2. Sediment type interpretation for the Kingston
to Saugerties, NY stretch, distinguished by combining
grain size analyses from >150 grab samples and core
tops and sidescan sonar data (Fig. 1).
Kingston, NY
Combination of data types
live zebra mussels
Figure 4. Grab sample sites in the Kingston-Saugerties
area. Red circles indicate sites where live zebra
mussels were found; gray circles, sites where live
zebra mussels were absent.
Figure 5. Polygons resulting from the combination of
sediment type and sedimentary environment data
overlain with samples sites using ArcGIS. Each
polygon has a unique ID and a set of attributes
contributed by both the sediment type map (Fig. 2)
and the sedimentary environment map (Fig. 3).
Deposition - over bedrock
Deposition - unres. thickness
Dynamic - waves
Dynamic - lineation
Dynamic - debris
Dynamic - scour
Dynamic - slump
Dynamic - streaks
41°56'0"N
41°56'0"N
t
u
ndo
Ro ek
Cre
New York
City
0
1
Kilometers
73°58'0"W
73°57'0"W
73°56'0"W
2
ATLANTIC OCEAN
73°55'0"W
Erosion - bedrock
41°36'0"N
Erosion - truncated
0
1
2
Erosion - non-deposition
Marlboro, NY
Kilometers
41°55'0"N
73°54'0"W
Figure 1. Basemap of the Kingston-Saugerties area of the Hudson River Estuary, showing
100kHz sidescan sonar data. Grab sample sites are shown as circles; red circles indicate
live zebra mussels, gray circles indicate no living zebra mussels. Note the presence of live
zebra mussels in areas of high sidescan backscatter.
ACKNOWLEDGEMENTS
The Hudson River Benthic Mapping Project is funded by the New York State Dept. of Env. Conservation.
This research is also supported by the Estuarine Reserves Division, Office of Ocean and Coastal Resource
Management, National Ocean Service, National Oceanic and Atmospheric Administration. Special thanks
go to Dave Strayer from the Institute of Ecosystem Studies in Millbrook, NY for sharing data and insight.
Unsurveyed
41°30'0"N
0
1
41°30'0"N
2
Kilometers
GRAVEL DEPOSITS
Newburgh, NY
Danskammer Pt.
41°57'0"N
Newburgh-Beacon
Bridges
Dynamic - drift
DYNAMIC AREAS
41°34'0"N
41°34'0"N
41°57'0"N
41°32'0"N
Deposition - thick
Deposition - thin
41°36'0"N
NEW YORK
FUTURE DIRECTIONS
• Explore correlations between live zebra mussels and estuary morphology, distance
from nearest tributary, mean salinity, and hydrographic parameters.
• Incorporate zebra mussel sampling data from 6-7 additional areas in the estuary to
better constrain conditions favorable to zebra mussels.
• Extrapolate critical parameters to whole estuary to map substrates suitable for
zebra mussel colonization.
• Consider extent of suitable substrate that falls in saline areas, which may be
vulnerable to colonization during wet years or even periods of high river flow
(e.g. floods, spring freshet).
74°W
Sedimentary Environment
sand
gravelly sand
muddy gravel
sandy gravel
gravel
no zebra mussels
Unsurveyed
SUMMARY OF KEY RESULTS
• Zebra mussels are most often found in areas of coarser grain size, where the
dominant sediment type is sand or gravel.
• More than 80% of live zebra mussels in the Kingston-Saugerties area are found
in dynamic sedimentary environments, particularly areas of debris (potential dump
sites) and scour (moderate to high backscatter, rough surface in bathymetry).
• Live zebra mussels are rarely found in areas where sediment waves are imaged
(dynamic waves class), although fragments of zebra mussels shells are found.
• Where live zebra mussels are found in muddy sediments, they are always
classified as regions of dynamic scour in the Kingston-Saugerties area.
• Although live zebra mussels are often imaged in areas of high sidescan backscatter,
there is no statistical correlation between backscatter value and presence of zebra
mussels in our data for the Kingston-Saugerties area.
• In the Kingston-Saugerties area, there is no correlation between zebra mussel
presence and water depth or distance to shoreline.
73°46'0"W
Newburgh Bay, south
41°32'0"N
mud
sandy mud
gravelly mud
muddy sand
Sample Sites
41°58'0"N
41°58'0"N
73°48'0"W
74°2'0"W
41°38'0"N
Sediment Type
Erosion - non-deposition
Figure 3. Process-based sediment environment
interpretation of the Kingston-Saugerties area. We sorted
the different processes into three major classes:
depositional, erosional, and dynamic. Within each class,
sub-classes define the details of the dominant process.
73°58'0"W
Newburgh Bay, north
41°59'0"N
41°59'0"N
73°50'0"W
Figure 7. Sidescan sonar mosaic of the Stockport Flats
area in the northern estuary. Red circles indicate the
presence of live zebra mussels; gray circles indicate
their absence. Zebra mussels are often found in areas
of high backscatter, specifically along the main channel
in this region.
Sample Sites
Dynamic - drift
mud
Sediment Type
73°46'0"W
Figure 6. Sidescan sonar mosaic of the New Baltimore
area in the northern estuary. Red circles indicate the
presence of live zebra mussels; gray circles indicate
their absence. Zebra mussels are commonly found in
high backscatter areas, which may correspond to harder
estuary floor, coarser grained deposits, or highly
dynamic locations.
Deposition - thick
42°1'0"N
42°16'0"N
Hudson, NY
New Baltimore, NY
Sedimentary Environment
42°1'0"N
42°28'0"N
42°28'0"N
Mid
Low
Sample Sites
42°18'0"N
42°18'0"N
High
42°3'0"N
42°20'0"N
Stockport
Creek
s
reek
us C
Esop
Sidescan Backscatter
2
Castleton-on-Hudson
Bridge
Gro
42°4'0"N
1
Kilometers
HIGH BACKSCATTER
dle
Saugerties, NY
42°5'0"N
Schodack
Islands
±
42°5'0"N
Wappinger
Creek
41°28'0"N
41°28'0"N
41°32'0"N
41°32'0"N
Storm King
Mountain
41°26'0"N
41°26'0"N
Beacon, NY
74°W
73°58'0"W
Figure 8. Sediment map of north Newburgh Bay, based
on the combination of grab and core samples and
sidescan sonar data. Red circles indicate the presence
of live zebra mussels; gray circles indicate their absence.
In this mainly muddy area, live zebra mussels are found
on coarse deposits (gravel lags, tributary deposits).
74°2'0"W
74°W
Figure 9. Sedimentary environment map of south
Newburgh Bay, based on all available data. Red circles
indicate the presence of live zebra mussels; gray circles
indicate their absence. Live zebra mussels are found
near the Newburgh-Beacon bridges, where the estuary
bottom is dominated by dynamic drifts and scour.