Photo by the USGS.

BAY-DELTA SCIENCE CONFERENCE: Ecosystem Engineering Impacts of Invasive Submerged Aquatic Vegetation (SAV) in the Delta

Plants that grow under the water surface are known as submerged aquatic vegetation (SAV).  Some species of SAV have been introduced from other regions of the world and outcompete many native species. These invasive species of SAV often disrupt waterways by reducing flows, obstructing navigation,  and clogging water intakes.  In California, such invasive SAV include Brazilian waterweed (Egeria densa), curlyleaf pondweed (Potamogeton crispus L.), Eurasian watermilfoil (Myriophyllum spicatum), and fanwort (Cabomba caroliniana).

Invasive SAV are of particular concern  throughout the Delta region. Despite ongoing efforts to control the problem, the total area invaded by aquatic weeds has doubled between 2004 and 2018.  Aquatic weeds are now so pervasive that they threaten restoration investments, endangered species,  recreation,  the local economy,  and water project operations.

Species of SAV can act as ecosystem engineers by altering water temperature and nutrient dynamics, providing habitat for invasive fish, reducing phytoplankton productivity, and transforming plant community composition.  Less is known, however, about how invasive SAV affects long-term ecosystem processes such as carbon storage and sediment dynamics.  For this reason, researchers studied Egeria densa, a species of SAV highly prevalent in the Delta, to determine its ecosystem engineering impacts.  The authors of the study were Dr. Judith Drexler (U.S. Geological Survey (USGS)-Sacramento); Dr. Jessica Lacy (USGS-Santa Cruz); Dr. Shruti Khanna (California Department of Fish and Wildlife (CDFW)-Stockton, CA); Dr. Maureen Downing-Kunz (ESA Associates); and Dr. Paul Work (USGS-Sacramento).  The research was supported by the USGS Priority Ecosystems Science Program.

At the 2021 Bay-Delta Science Conference, Dr. Judith Drexler discussed the results of three separate but related studies on how Egeria densa (hereafter abbreviated as E. densa) alters carbon storage and sediment dynamics.

E. densa is a species of freshwater submerged aquatic vegetation (SAV) native to the central Minas Gerais region of Brazil, coastal Argentina, and coastal Uruguay. It has spread to all continents except Antarctica, and as the map shows, it’s almost everywhere across the USA.  It was used as an ornamental in aquariums and ponds, and has been spread by boat traffic and when people dump their aquariums into waterways. 

The plant has very dense leaves and extensive branching that extends through the entire water column. It has an amazing ability to persist in the environment as it can reproduce from stem fragments, tolerate a wide range of temperatures, and thrive in disturbed areas.

The slide shows a conceptual model for how E. densa affects the Delta’s ecosystem.  In a typical channel environment, SAV lines the channel edge and the marsh grows behind the SAV.

What we hypothesize is that the SAV reduces flows, the suspended sediment settles out of the water column, and the sediment becomes trapped within the patch or below the patch,” said Dr. Drexler.  “What happens is you get a clear water column which is bad for sensitive fish, but also the sediment that’s really needed for the marsh to maintain its elevation on the sea level rise is not getting to the marsh. It’s not being deposited.”

The slide lists the several ecosystem engineering properties of E. densa, that can transform ecosystems to the detriment of native species. The researchers focused on the last two: whether  patches of SAV dominated by E. densa in the Delta are retarding flow velocities, reducing bed shear stress, and altering sediment transport properties.   E. densa-dominated patches were chosen because they are highly widespread in the Delta and the plants extend throughout the entire water column, which is important in regard to its ability to trap sediment.

The overall research question was, do E. densa-dominated SAV patches trap and store carbon and inorganic sediment and alter sediment dynamics?   The objectives of the study were:

  • To quantify inorganic sedimentation rates and carbon accumulation rates in SAV patches and tidal freshwater marshes under a range of hydrodynamic conditions.
  • To determine whether SAV patches are sites of carbon storage with rates similar to other blue carbon systems. (Blue carbon systems are coastal ecosystems such as mangroves, seagrasses, and tidal marshes, that accumulate and store high amount of organic carbon.)
  • To determine instantaneous sediment trapping by SAV at the patch scale.
  • To determine patch-scale measurements of current damping and reduction of suspended-sediment concentrations.

Background and study sites

E. densa arrived in the Delta in 1946 and was widespread by 1990.   In 2015, about 5000 hectares of the Delta were estimated to be covered by invasive SAV, which includes E.  densa as well as other invasive SAV.   In 2019, SAV covered about 35% of Delta waterways, with E. densa comprising 52%.  It is found along channel edges and flooded islands.

Dr. Drexler noted that she would not be talking about the impact of E. densa on flooded islands in this presentation because they weren’t able to study them for this project, but they hope to do so in the future.

The three study sites chosen were Lindsey Slough in the North Delta, a backwater area of very low hydrodynamic energy; the Mokelumne River at its confluence with the San Joaquin, which is a high energy site; and Middle River, which is, in terms of energy levels, in between the other two.

The results

For the study, they collected surficial push cores in E. densa patches and the unvegetated channel and collected deeper 50 cm cores within E. densa patches and adjacent tidal freshwater marshes.  The cores were analyzed for organic carbon and bulk density and were dated using isotopes of cesium (137Cs) and lead (210Pb).    

They first looked at whether the patches of E. densa differed from the rest of the bed environment in the channel, so they measured the bulk density and percent organic carbon in the push cores.  “Two very simple parameters, but they can tell you a lot about the different kinds of depositional environments,” said Dr. Drexler.

In the push cores, the unvegetated channel had a much greater bulk density and lower percent organic carbon value than the SAV patch dominated by E. densa.   

As for the marsh cores and the cores collected under the E. densa patches (slide below), the bulk density in the SAV cores was greater than in the marsh cores; the percent organic carbon was greater in the marsh cores than the SAV cores.

The slide below shows the mean vertical accretion rate in the marsh cores and the cores under E. densa.  Vertical accretion is measured in centimeters per year of material and represents the depth of material that deposits on a SAV patch or the marsh surface.  

The SAV cores have much greater mean vertical accretion than the marsh cores,” said Dr. Drexler.  “This is not surprising because the SAV cores are in a subtidal environment, and the marsh cores are in an intertidal environment; a lot more sediments are coming by the SAV cores than the marsh cores.”

That same logic applies for mean inorganic sedimentation rates; the rate in the SAV is much higher than in the marsh.” (shown below)

She noted that for the cores at the Mokelumne site, they could not determine inorganic sedimentation rates because they could not be dated with 210Pb due to disturbance in the sediment layers.  

The slide below shows mean carbon accumulation in marsh cores and cores under E. densa.  Carbon accumulation is measured in grams of carbon per meter squared per year.

There isn’t a significant difference between the marsh cores and the cores under E. densa,” said Dr. Drexler.  “The cores under E. densa have carbon accumulation rates well within the range of all major blue carbon systems: marshes, seagrasses, and mangroves.”

She noted that one of the cores from Lindsey Slough is very close to the global seagrass mean, and one of the cores from Middle River has a carbon accumulation rate even higher than the global salt marsh mean.

Instantaneous sediment trapping

Dr. Drexler then briefly reviewed the work that Dr. Paul Work and colleagues did for the instantaneous sediment trapping component of the study.

The goal of this part of the study was to conduct a closed-circuit course around the vegetation patches within 15 minutes in a kayak and log the velocity, acoustic backscatter, and depth data continuously.  Water samples were also collected to be able to relate the suspended sediment concentration to the acoustic backscatter.

Dr. Work took the measurements and computed net and gross fluxes of water and sediment, as well as the instantaneous trapping efficiency.

Dr. Drexler then gave the results.  “First of all, instantaneous trapping efficiency averaged about 5% or 3.7 kilograms of sediment per meter squared per year; this matched really well with the mean mass accumulation of sediments in the SAV cores, which was 3.8 kilograms per meter squared per year.  And these were done with completely different methods, so we were really pleased to see how closely they matched.”

Suspended sediment has declined by 1.8% per year over the past 60 years, according to research Paul did for this paper,” she continued.  “This decrease in suspended sediment concentration in rivers will likely reduce marsh resiliency in the Delta.”

Patch-scale measurements of current damping and reduction of suspended sediment concentrations by SAV

Dr. Jessica Lacey looked at patch-scale measurements of current damping and reduction of suspended sediment concentration by SAV patches, focusing on Lindsay Slough, the low energy site, and the Mokelumne River, the high energy site.   She looked at vegetation density and time series of current velocity and suspended sediment outside and within the patches of SAV.

The results for the effect of SAV on currents are shown on the upper right. 

The patches really have a strong impact on current speed with attenuation over 90%,” said Dr. Drexler.  “In the second plot, you can see a strong relationship between the vegetation drag coefficient and Red, which is the stem Reynolds number, a non-dimensional representation of current speed and vegetation that incorporates stem width.  This drag coefficient Cd can be used for E. densa of any density as long as the frontal area is known, where frontal area is the surface area if you’re looking at the vegetation from within the water column.”

She noted that the attenuation result in the first plot is specific to a particular patch and is representative of other patches with similar plant density.  “These kinds of parameters could be used for modeling, so these are really important to figure out for this particular species,” she said.

The results for the reduction of suspended sediment concentration in SAV patches are shown on the slide below.  

In the low energy site, Lindsay Slough, the suspended sediment concentration in SAV is 20% lower than in the channel,” said Dr. Drexler.  “In the Mokelumne River, the differences between suspended sediment concentration in the SAV and channel increases with suspended sediment concentration. There’s a 20% trapping at low suspended sediment concentrations and then 25% trapping at higher suspended sediment concentrations.  Overall the percent reduction is very similar at the two sites. Equal percent reduction means more trapping at sites with greater suspended sediment concentrations, and the cores confirmed that there was more sediment trapping at Mokelumne than Lindsey Slough.”

Implications for marshes

The results suggest that there will be less sediment for marshes along channels lined with SAV.  “The rising tide passes through the fringing SAV as it floods the marsh; there’s 20 to 25% lower suspended sediment concentration in the SAV than there is in the channel,” said Dr. Drexler. “This same reduction of sediment flux is going to be translated to the marshes, and this is occurring in addition to the system-wide decrease in Delta suspended sediment concentration that Paul showed us from his work.”


The major conclusions from this invasive aquatic vegetation project in the Delta are:

  1. densa-dominated SAV patches act as an inorganic sediment sink in the Delta and likely anywhere where they are found. In the Delta, less common SAV species such as watermilfoil and fanwort are probably storing sediment as well, but that has not been studied.
  2. densa dominated SAV patches store blue carbon at rates similar to seagrasses and are within the range of all major blue carbon systems. Dr. Drexler noted that other species might also be storing carbon, but that has not been studied, nor any species in flooded island environments.

If we scale up to the Delta level, E. densa stores about 3500 metric tons of carbon per year, which represents about 38% of the carbon stored annually in the 7800 hectares of restored marshes in the Delta region,” said Dr. Drexler.  “This represents just a tiny percent of the carbon lost since mass conversion to agriculture, but it shows that carbon stored in E. densa represents a meaningful component of recent blue carbon storage in the Delta.”

Despite providing a valuable ecosystem service, E. densa is highly disruptive and should be controlled as much as possible.  This is not a panacea for carbon pollution, but in places where it already exists, it is serving a function of storing carbon and sediment.”

Regarding inorganic sediment, E. densa stores approximately 100,000 metric tons of sediment per year in the Delta, representing about 6% of the 1.6 million metric tons per year of sediment supplied to the Delta from the Sacramento and San Joaquin rivers.

Flow penetrating SAV patches drops between 5 to 25% of its sediment load within a patch.

Sediment is at least partially being blocked by SAV from depositing on marshes, and this together with a 1.8% decline in suspended sediment concentration per year is reducing marsh resilience under sea level rise,” said Dr. Drexler.  “This obviously will become much more important as sea level rise accelerates in mid-century and beyond.”

Invasive SAV may also be altering the carbon balance of the Delta, particularly if patches emit methane during the hot summer season.  “This has been shown to occur in France in some lakes, and we have a proposal in to do some measurements to see if this is also occurring in the Delta.”


COMMENT: Paul Work, one of the colleagues on the study, noted, “The trend that Judy mentioned in terms of the decrease in suspended sediment concentrations that we’ve observed was measured at Freeport, which is upstream of the Delta, which is our longest record.  We have other stations that have, at this point, a couple of decades of data. And as the record gets longer, we’ll have to look at trends at those locations as well. But it’s definitely concerning when you see a consistent downward trend like that in most things.” 

QUESTION: Are there areas with a large amount of SAV filling in channels since they attract sediment?

Dr. Drexler:What’s happening is that … the SAV tends to channelize the flow to the deeper part of the channel that does not have vegetation because it’s too deep for the vegetation to grow. So usually it won’t clog up the whole channel. But if you had a very shallow dead-end slough, it could be in the entire thing.  And in the flooded islands, depending on how deep those are, it’s usually in the shallower parts. It’s not going to be where there are more than three meters of water.”

QUESTION: Could your work inform the estimation of a change in turbidity if SAV is removed?

Dr. Drexler:SAV has definitely decreased turbidity. And if you want to look at a really nice paper on that, you can look at Erin Hestir’s paper from 2016. So the SAV does clear the water column. And if the SAV wouldn’t be there, you would probably have higher turbidity, but of course, it really depends on site-specific characteristics on how much that effect would hold.”

QUESTION: Where do you see this going? So if you had your ideal situation over the next one to five years, where would you like to see this research go?

Dr. Drexler: I would love to look at more plant species in more environments, like flooded islands

Paul Work:  “What we did was we estimated how much trapping is going to happen in one particular location over a short timescale, basically. And for management purposes, we need to go much bigger than that, both in time and space.  We also think that those flooded islands have the potential to be a big sink for sediment, too, with the low velocities. So I think probably some kind of hybrid approach that combines modeling and observations of suspended sediment concentrations could be a good way to get there and look at the larger scale.”

Jessica Lacy:I think the inclusion of a flooded Island would be really valuable. One of the complications about the flooded islands is that, of course, whether there’s vegetation there or not, they’re going to retain sediment because the velocities are so much lower there.  So the approach of just measuring sediment in and sediment out is not necessarily as effective because we have to also account for the settling that would happen in the absence of vegetation.  The work we did on attenuation of currents and reduction of sediment concentration allowed us to kind of extract what we call scaling. So what ratios of factors might influence whether a certain environment would be more or less trapping? And we did think that flooded islands are a strong candidate for being even more effective at trapping than the channels. My big push was to try to develop parameters that can be used in models so I’d love to see that used in models. And I think one of the things that could be an exciting partnership is our increased ability to map where vegetation is. There’s the aerial mapping, but we also now have acoustic methods that allow us to see the edge of vegetation. And so by combining that, I think we could have some really strong modeling done.”

QUESTION: Do you think that the presence of SAV creates deeper, narrower channels?

Dr. Drexler: “I think that the vegetation is so dense that it is actually almost acting like a levee along the channels. And in some situations, it can be increasing the amount of sediment flux through the channel rather than decreasing it. That depends on whether it’s in a more energetic channel, where it could be reducing sediment flux, but in low-energy channels, I think it can actually increase the sediment flux by channelizing the flow. So I do think that’s an important aspect of what’s going on. And that’s one of the interesting ways in which there can be a tipping point as far as the overall effects of the vegetation depending on the environmental setting.“


Carbon storage and sediment trapping by Egeria densa Planch., a globally invasive, freshwater macrophyte

Judith Z. Drexler, Shruti Khanna, Jessica R. Lacy

Abstract: Invasive plants have long been recognized for altering ecosystem properties, but their long-term impacts on ecosystem processes remain largely unknown. In this study, we determined the impact of Egeria densa Planch, a globally invasive freshwater macrophyte, on sedimentation processes in a large tidal freshwaterregion. We measured carbon accumulation (CARs) and inorganic sedimentation rates in submerged aquatic vegetation (SAV) dominated by E. densa and compared these rates to those of adjacent tidal freshwater marshes. Study sites were chosen along a range of hydrodynamic conditions in the Sacramento-San Joaquin Delta of California, USA, where E. densa has been widespread since 1990. Cores were analyzed for bulk density, % inorganic matter, % organic carbon, 210Pb, and 137Cs. Our results show that E. densa patches constitute sinks for both “blue carbon” and inorganic sediment. Compared to marshes, E. densa patches have greater inorganic sedimentation rates (E. densa: 1103–5989 g m−2 yr−1, marsh: 393–1001 g m−2 yr−1, p < 0.01) and vertical accretion rates (E. densa: 0.4–1.3 cm yr−1, marsh: 0.3–0.5 cm yr−1, p < 0.05), but similar CARs (E. densa: 59–242 g C m−2 yr−1, marsh: 109–169 g C m−2 yr−1, p > 0.05). Sediment stored by E. densa likely reduces the resilience of adjacent marshes by depleting the sediment available for marshbuilding.  Because of its harmful traits, E. densa is not a suitable candidate for mitigating carbon pollution; however, currently invaded habitats may already contain a meaningful component of regional carbon budgets.  Our results strongly suggest that E. densa patches are sinks for carbon and inorganic sediment throughout its global range, raising questions about how invasive SAV is altering biogeochemical cycling and sediment dynamics across freshwater ecosystems.

Trapping of Suspended Sediment by Submerged Aquatic Vegetation in a Tidal Freshwater Region: Field Observations and Long-Term Trends

Paul A. Work & Maureen Downing-Kunz & Judith Z. Drexler

Abstract: Widespread invasion by non-native, submerged aquatic vegetation (SAV) may modify the sediment budget of an estuary, reducing the availability of inorganic sediment required by marshes to maintain their position in the tidal frame. The instantaneous trapping rate of suspended sediment in SAVpatches in an estuary has not previously been quantified via field observations. In this study, flows of water and suspended sediment through patches of invasive SAV were measured at three tidally forced, freshwater sites, all located within the Sacramento-San Joaquin Delta in California. An acoustic Doppler current profiler deployed from a roving vessel provided velocity and backscatter data used to quantify fluxes of bothwater and suspended sediment. Sediment trapping efficiency, defined as instantaneous net trapped flux divided by incident flux, was positive in 24 of 29 cases, averaging + 5%. Coupled with 3 years of measured sediment flux data at one site, this suggests that trapping averages 3.7 kg m−2 year−1. This estimate compares favorably with the mean mass accumulation rate of 3.8 kg m−2 year−1 estimated from dated sediment cores collected at the study sites. Long-term measurements made upstream reveal a strong negative trend (−1.8% year−1) in suspended sediment concentration, and intra-annual changes in both suspended sediment concentration and percent fines. The large footprint and high spatial density of invasive SAV coupled with declining sediment supply are diminishing downstream suspended sediment concentrations, potentially reducing the resiliency of marshes in the Delta and lower estuary to future sea-level rise.

Influence of Invasive Submerged Aquatic Vegetation (E. densa) on Currents and Sediment Transport in a Freshwater Tidal System

Jessica R. Lacy, Madeline R. Foster-Martinez, Rachel M. Allen, and Judith Z. Drexler

Abstract: We present a field study combining measurements of vegetation density, vegetative drag, and reduction of suspended-sediment concentration (SSC) within patches of the invasive submerged aquatic plant Egeria densa. Our study was motivated by concern that sediment trapping by E. densa, which has proliferated in the Sacramento–San Joaquin Delta, is impacting marsh accretion and reducing turbidity. In the freshwater tidal Delta, E. densa occupies shallow regions frequently along channel margins. We investigated two sites: Lindsey Slough, a muddy low-energy backwater, and the lower Mokelumne River, with stronger currents and sandy bed sediments. At the two sites, biomass density, frontal area, and areal density of the submerged aquatic vegetation (SAV) were similar. Current attenuation within E. densa exceeded 90% and the vegetative drag coefficient followed Cd  174Red1.46, where Red is stem Reynolds number. The SAV reduced SSC by an average of 18% in Lindsey Slough.  At the Mokelumne River the reduction ranged 0%–40%, with greatest trapping when discharge and SSC were elevated. This depletion of SSC decreases the transport of sediment to marshes by the same percentage, as the rising tide must pass through fringing SAV before reaching marshes. Sediment trapping in E. densa in the Delta is limited by low flux through the canopy and low settling velocity of suspended sediment (mostly flocculated mud). Sediment trapping by SAV has the potential to reduce channel SSC, but the magnitude and sign of the effect can vary with local factors including vegetative coverage and the depositional or erosional nature of the setting.

Print Friendly, PDF & Email