Biological Control of Giant Salvinia
(Salvinia molesta)
in Texas and Louisiana

Cooperators:

 Philip W. Tipping, Ph.D. and Ted D. Center, Ph.D.
USDA-ARS Invasive Plant Research Laboratory, Ft. Lauderdale, FL.

Howard Elder, Rhandy Helton and John Findeisen
Texas Parks and Wildlife Department, Jasper and Mathis, TX.
 

Research Reports:

2004

• Nov - After establishing the salvinia weevil and proving its ability to control Salvinia molesta in Louisiana and Texas, Phil Tipping, USDA/ARS now draws the five year study to completion
• June - Integrated approach adopted at Louisiana golf course pond and weevils prevail in Texas
• May - April trip finds weevils arresting Salvinia molesta in Texas; Louisiana receiving an integrated approach
• Mar - Early spring finds low weevil numbers and winter damaged giant salvinia poised for a rebound, all as expected for the time of the year

2003

• Oct -  Giant salvinia mat is dramatically decreased in east Texas while weevils demonstrate exceptional dispersal abilities
• Sept - Weevils maintain biomass reduction of Salvinia molesta
• Aug - Fourty-four ton per acre difference measured between weevil release and control sites on Salvinia molesta
• Aug - Salvinia weevil suppresses giant salvinia to an undetectable level by July at release site in Louisiana – encouraging results concurrently recorded in Texas
• Jul - Excitement brews over rotting Salvinia molesta and striking differences between weevil and control plots
• Mar - Cyrtobagous weevil on Salvinia molesta in Texas and Louisiana

2002

• Dec - Final research report for the 2002 season
• Sept - Damage found severe at all release sites
• Jul - Brazilian weevil Cyrtobagous salviniae continues to grow and spread on Salvinia molesta at release sites
• Jun -  Salvinia Weevil Grows, Spreads and Damages Salvinia molesta at Release Sites
• May - Evidence Salvinia Weevil Reproducing and Increasing at Release Sites
• Apr - Salvinia Weevil Successfully Overwinters at Trial Release Sites

2001

• Dec - Weevils Recovered and Damage Observed after Initial Releases
• Oct - Biological Control Advances!
   

October 2001

• Biological Control Advances!
  
  The first field releases of the Australian imported weevil, Cyrtobagous salviniae, were made early October on Salvinia molesta in Texas and Louisiana. The U.S. Dept. of Agriculture, Invasive Plant Research Laboratory, collaborating with Texas Parks & Wildlife Dept. released a total of 1100 weevils at four sites. The insects are expected to produce at least one generation of offspring before cooler temperatures close in. For that reason not much damage may be seen this year, however the release will at the very least, be a good test of the weevil's ability to survive the winter. The Louisiana release site is adjacent to Toledo Bend Reservoir in the western part of the state. Private ponds in Liberty and Chambers County and Lake Texana, all eastern Texas, also received a dose of the plant-eating insect.
  
  The team will continue to monitor these release sites as well as five check sites (no weevils released) throughout the winter. Data will be collected on the survival, fecundity, and establishment of the weevil. Plant damage, plant nutrition, and the relationship between plant nutrition and insect population will also be assessed. Currently the Texas side of Toledo Bend Reservoir is at an all-time low making potentially attractive sites inaccessible. However, additional weevil releases may be conducted in the spring or summer of 2002. (18 Oct.)

December 2001

• Weevils Recovered and Damage Observed after Initial Releases
  
  Nearly two months after their release, weevils were recovered at 3 of the 4 experimental sites in Texas and Louisiana. Significant weevil damage was noted at 2 of the 4 release sites. About 25 to 90 adult weevils are suspected to remain in the experimental squares. Phil Tipping, USDA, ARS and principal investigator for the project relates:
   

"We'll just have to wait and see if the damage spreads outside the release square. We have seen this type of damage and impending collapse of salvinia in release squares before, most notably in the old Harris Pond site, followed by resurgence of the weed."   (10 Dec.)

April 2002

• Salvinia Weevil Successfully Overwinters at Trial Release Sites
  
  Returning from their first spring visit to research plots in Texas and Louisiana, USDA/ARS entomologists and colleagues were pleased to find salvinia weevils, Cyrtobagous salviniae, at 2 of the 4 October 2001 release sites. Brazilian by origin, C. salviniae was recently imported from Australia (where it is reared and used for biocontrol) and experimentally released to sites in Texas and Louisiana. While it cannot be determined if the persisting weevils are either the original individuals released or their offspring, it is clear that C. salviniae was able to survive the winter in the Toledo Bend and Lake Texana areas. The Toledo Bend region experienced several hard freezes this winter, where temperatures dropped as low as 12 ºF (-11 ºC). Finding the weevils capable of persisting through these temperatures is a good sign for future use.
  
  Phil Tipping (USDA/ARS), Rhandy Helton, John Findiesen and their crew at TPWD sampled 100 plants by hand at plots in Louisiana, near Toledo Bend Reservoir, and at Lake Texana, Texas. Although searching plants by hand is not very efficient, in this case it was necessary because it does not disturb the weevil. Processing similar amounts of plants with destructive Berlese funnel techniques usually yields 5-10 times as many adults, but results in removal of insect agents from the test site.
  
  Adult weevils were not detected at the remaining two sites, ponds in Liberty Co. and Chambers Co., Texas. One pond had suffered heavy salvinia damage from weevil feeding and had yielded several adults during sampling in Dec. 2001. Since then heavy flooding has occurred there. Floodwaters are believed to have flushed the weevils out of the site, along with much of its host plant, a likely explanation as to why no weevils and no plant damage were found during this visit. The second pond had significantly less Salvinia molesta occurring than was found in Dec. 2001, yet reasons for the plant’s decline are unknown. Weevil damage was clearly observed at this site, but, unfortunately, no adults were detected.
  
  Overall the team is very encouraged by the successful overwintering, and plan another release of weevils later this spring. (28 Mar.)

May 2002

• Evidence Salvinia Weevil Reproducing and Increasing at Release Sites
  
  Additional releases of Brazilian weevils were conducted in May at the four research sites in Texas and Louisiana. A total of 1200 adults were placed in release squares to augment previous releases made in Oct. 2001. During the May trip, adults were found at every site except one. This particular site once again experienced high water levels, which may have flushed out plants and weevils. Adult weevils were found up to one meter from the original release area in 3 of 4 sites. At the Lake Texana site, a station that has had no additional releases of adults for 8 months, hand-searching plants within the release square and up to one meter away found 13 adults. Given the difficulty in detecting weevils using this method, we estimate that 60-120 adults were actually present in the meter-square release area. One-third of those adults were a light brown color, indicating that they had emerged within the last week or so. Older weevils turn a darker brown color. This is strong evidence that at least one generation has been completed and that the resident weevil population is increasing. Significant weevil feeding and damage were evident at this site.

In summary, there is good evidence that the weevil Cyrtobagous salviniae is establishing at selected sites and beginning to spread out of the release areas. Warmer temperatures and additional releases should speed up the process as long as the release sites remain stable.  Monthly releases of weevils are planned up until November of this year.  (17 May)

June 2002

• Salvinia Weevil Grows, Spreads and Damages Salvinia molesta at Release Sites
  
  Additional releases of Brazilian weevils were conducted in June at four sites in Texas and Louisiana. A total of 500 adults was placed in release squares to augment previous releases in May 2002. Adults were found at every site and approximately 30% were a brown color indicating they were progeny of earlier releases.  Significant plant damage was evident at 3 of the 4 release sites and giant salvinia biomass was reduced an average of 61% compared to the previous month. The surviving salvinia plants were extremely brittle and broke apart even with gentle handling. Biomass in the control sites increased an average of 44% during the same period.
  
  In summary, there is evidence that populations of the weevil Cyrtobagous salviniae continue to grow and spread in release sites. Although it is early in the project, there has been significant damage inflicted on the salvinia in and around the release squares. Time will tell if this level of damage persists and grows or if the plants will be able to compensate and recover. Warmer temperatures and additional weevil releases should speed up the process as long as the research sites remain stable, without interference from flooding or drought. Monthly releases of weevils are planned up until November of this year.  (25 Jun.)

July 2002

• Brazilian weevil Cyrtobagous salviniae continues to grow and spread on Salvinia molesta at release sites
  
  Additional releases of Brazilian weevils were conducted in June at four sites in Texas and Louisiana. A total of 275 adults were placed in release squares to augment previous releases made in April, May, and June 2002. Adults were found at every site. Approximately 25% of the adult weevils were brown in color, indicating they were progeny of earlier releases. On average, the amount of biomass (fresh weight) in the study squares was 82% less in release sites as compared with control sites.  Also, at release sites, biomass in the vicinity of the squares was 47% less than control sites.
  
  Significant weevil damage occurred at one of the release sites.  All the giant salvinia in and around the square had been destroyed, forcing the new releases to be made more than 5 meters away from the original square. Flooding impacted a second release site, which flushed out the weevil-damaged salvinia and replaced it with new plants from ‘upstream’. Intermediate levels of damage were observed at the remaining two study sites.  A new control site was set up at Lake Texana, replacing one destroyed by vandalism.
  
  In summary, populations of the weevil Cyrtobagous salviniae continue to grow and spread in release sites that remain undisturbed. Damage has increased in severity and continues to spread outward from the release square at one site, which has remained undisturbed by flooding. Other release sites show intermediate levels, probably as a result of disturbances that make evaluation difficult. Monthly releases of weevils are planned until November of this year.

September 2002

• Damage found severe at all release sites
  
  Another release of Brazilian weevils was conducted in September at one of the four release sites in Texas and Louisiana.  A total of 125 adults was released 5 meters from the study square to augment previous releases in April, May, June, and July 2002. More adults were found at every site than ever before although a lower percentage were light brown, indicating they were progeny of earlier releases. At two sites, weevils were detected on 20% and 18% of all plants by hand-searching. The site in Louisiana showed large areas of salvinia with a dark brown color and waterlogged appearance. This type of damage is usually followed by wholesale sinking of the mats. Open water was present at another site in the vicinity of the study square for the first time in 3 years. Another release site in a marsh showed much higher levels of damage but the property owner will not permit future use and evaluation of the site. The site in Lake Texana suffered some vandalism which made assessment difficult. In general, there was about a 7-fold difference in biomass found in the study squares between control and release sites with the average fresh weight of salvinia equivalent to 39 tons per acre in control sites and 5.5 tons in release sites. In addition, the trend in the release sites continues to be negative, i. e. continues to decline, while the trend in the control sites has been largely positive as biomass continues to increase.
  
  In summary, populations of the weevil Cyrtobagous salviniae continue to grow and spread in undisturbed release sites. Unfortunately, another long-term release site has been lost because of the property changing hands. Damage is severe at all release sites and another month or two of warm weather may permit the weevils time to inflict enough damage to actually sink large areas of salvinia. Monthly releases of weevils are planned up until October of this year.

December 2002

• Final research report for the 2002 season
  
  The research plots in Texas and Louisiana were visited and data collected in late October 2002.  At Lake Texana a small number of Brazilian weevils (163) were released to restore populations that had earlier been flushed out of the study site. Cooler temperatures probably reduced the overall rate of weevil population growth and subsequent damage to the salvinia mats. Fewer adults were found at study sites and all were the darker brown color indicative of older individuals. At one site, weevils were detected on 27% of all plants by hand searching.  The site in Louisiana continued to show large areas of damaged salvinia as characterized by a dark brown color and waterlogged appearance, however, there was no wholesale sinking of the mats as anticipated. The open water around the study square at another site increased in size, allowing us to see the bottom there for the first time in 3 years. Overall, a direct comparison between control and release sites showed significantly less water surface covered by the weed, less biomass in the square, and less biomass adjacent to the square at the release sites. This was the last sample date in 2002. Sampling will begin again in March 2003.

March 2003

• Cyrtobagous weevil on Salvinia molesta in Texas and Louisiana
  
  The year’s first visit to the giant salvinia (Salvinia molesta) research plots in Texas and Louisiana occurred March 17-19, 2003.  No new weevils (Cyrtobagous salviniae) were released as the trip was intended to assess the populations that had over wintered.  Adult weevils were found at all release sites and in numbers larger than any date in 2002. This marks the second winter that the weevils have survived. The recent winter was much colder than the first. More than 30 adults were found at one site by hand-searching 100 plants. Many of the adults were already mating. The adults recovered were all dark brown in color and probably were adults last fall. This is not surprising because the temperatures since the fall have been below the threshold for egg laying.

Giant salvinia biomass was reduced at control sites as well as research plots, most likely due to winter killing of some plant mats. However, few damaged buds were present and rapid regrowth should be expected as spring temperatures rise. Open water was still available around one release site and weevils there were actively feeding and mating. Overall, a direct comparison between control and release plots continues to show significantly less water surface covered by the weed, less biomass in the study square, and less biomass adjacent to the square at the release sites. In addition, the numbers of weevils per biomass of giant salvinia is much higher than at this time last year, suggesting that the overall populations of weevils and the damage they cause will be proportionally greater in 2003. Our next sampling date will be in early June 2003.

July 2003

• Excitement brews over rotting Salvinia molesta and striking differences between weevil and control plots

Research plots of giant salvinia (Salvinia molesta) in Texas and Louisiana were visited and data collected during June 23-25, 2003. No salvinia weevils (Cyrtobagous salviniae) were released on this trip. The number of adult weevils found at all release sites was less than the number found earlier this year in March. The reason for this is simple and very exciting: much of the giant salvinia is no longer suitable for feeding.

The floating mats are rotting and sinking, particularly at one site. At this site, virtually all the salvinia examined was severely damaged. The absence of healthy, undamaged buds indicated that plant compensation and recovery will be minimal. The mat was dark brown and covered about 60% of the water’s surface, as compared with 100% coverage this past March.

Over all sites, the average wet weight of giant salvinia in the study squares was equal to 1.5 tons per acre in the weevil release sites and 24.5 tons per acre in the control sites. Comparing adjacent biomass between sites, we found an average of about 3 tons per acre at weevil release sites and 48 tons per acre at control sites. The difference in the percentage of surface water covered with salvinia was also very striking this trip: 13% and 100% coverage in release and control sites, respectively.

Unfortunately, the Lake Texana sites were compromised by extremely low water and vandalism.  The release site was completely dry and the square from the control site was stolen, again. We will not reestablish that site at Lake Texana State Park.

Now that significant reductions in salvinia biomass are occurring, the next important question will be how much of a decline we will see. At some point, a natural balance or equilibrium between weevil and weed should occur. This equilibrium level, whatever it turns out to be, will almost certainly be dynamic with periodic flushes of the weed followed by declines as the insects catch up and sink it.  Hopefully, this will continue at a level where any further control measures will not be necessary.

In Florida, common salvinia (Salvinia minima) typically covers the surfaces of water bodies while the weevil populations lag behind. However, these mats are neither thick nor persistent, and are eventually sunk by the activities of the weevils, especially in full sun sites. Based on the Australian experience with giant salvinia, once the equilibrium levels are reached, chemical or mechanical attempts to further reduce salvinia are unwise, generally backfiring and resulted in a resurgence of the weed.

August 2003

• Salvinia weevil suppresses giant salvinia to an undetectable level by July at release site in Louisiana – encouraging results concurrently recorded in Texas

The research plots in Texas and Louisiana were visited and data collected during July 21-23, 2003. We were pleasantly surprised to find not a single plant of Salvinia molesta at our Louisiana golf course pond. Nothing! (PowerPoint -1.7mb or Low Resolution Images 1 | 2 | 3). Any vegetation visible around the periphery of the pond or floating was either algae, duckweed, or alligatorweed. The pond is divided by a boom, which we installed for aesthetic reasons, so that the more visible end could be kept clear of salvinia using regular applications of herbicide. At our experimental end, the salvinia had been completely covering the surface for years. Despite a great deal of looking, we could not come up with a single plant, even a damaged one, of giant salvinia. Interestingly, the only giant salvinia we found was at the end where the spraying takes place. These plants were very young and growing rapidly. A small sample taken from this fringe did not yield any weevils. Back at our experimental end of the pond, we estimated that thousands of weevils had been attacking the rapidly weakening salvinia last month so that now complete collapse of the salvinia had occurred. It is likely that the weevils dispersed to nearby Toledo Bend reservoir where giant salvinia is growing in a cove less than 100 meters away.

The big question now is what kind of equilibrium will establish here between the weevil and the weed. There can no longer be any doubt about the ability of the weevil to suppress giant salvinia. Several scenarios are possible including 1) the salvinia disappears and does not return, and 2) the salvinia flushes back to cover the site until the weevils catch up and sink it again.  Scenario number two is more likely, but the salvinia will probably grow back faster than the weevil numbers can build up. It took about 21 months for weevils to build up to the present point of destroying the mat at the Louisiana site.  In Florida, common salvinia goes through a boom and bust cycle that lasts about a year as plant populations rise then crash as controlled by the weevil. However, south Florida is not subject to freezing temperatures experienced in Louisiana, which may act to further weaken plant mats, especially if they are already thinner than usual because of weevil attack. Of course, the colder temperatures in Louisiana may also negatively affect the weevils. Given the extremely warm temperatures and high nutrient conditions in the pond, it is likely that the rapidly growing fringe of giant salvinia at the sprayed end will recolonize that end. We will continue to monitor events in both ends of the golf course pond.

On to Texas, where at one of the release sites the extremely large mat of salvinia continues to be reduced. Despite prevailing winds that force the salvinia directly into the study square area, the weevils are keeping the surrounding water open. A few meters outside of the square, a thick peat mat is forming as salvinia dies and is compressed. As it happens, we flew over this site on our way to Louisiana and could clearly see that the salvinia only covers about 25% of the 200-acre pond. Normally this time of year, coverage would exceed 75%. The scale of the infestation in this pond will probably require another year of attack by the weevils to attain the level of control seen in Louisiana.

At Lake Texana, Texas, last month in June, our study square was high and dry on the bank as the reservoir water level had fallen more than 3 feet. Then came Hurricane Claudette in mid-July and back up went the water. Assuming the site had been flushed clear by hurricane flooding, we brought 100 weevils in late-July to augment the release site. We were surprised to find the salvinia that had survived both drought and flood with weevils still feeding on it. We added our weevils to those already there.

Giant salvinia biomass and coverage remained high at all of our control sites with an average of 41 tons of wet plants per acre and 100% coverage. In contrast, biomass at the release sites averaged less than 2 tons per acre, with zero for both biomass and coverage at the Louisiana site.

In summary, it appears that we have finally turned a corner. The sustainable suppression of giant salvinia is just a matter of time and a question of numbers. Our plans are to continue evaluating until winter while redistributing weevils to other infested areas. This fall we plant to release weevils at our control sites.

We will visit the sites again for our next update in August.

August 2003

• Fourty-four ton per acre difference measured between weevil release and control sites on Salvinia molesta
  
  The research plots in Texas and Louisiana were visited and data collected during August 25-27, 2003.  The site at the Louisiana golf course pond saw giant salvinia completely re-colonize the sprayed side of the boom.  A few plants of giant salvinia were present in the biological control side of the boom.  No weevils were found on either side.   It’s likely that they have dispersed to the nearby Toledo Bend reservoir.  Our last report predicted a rebound in the giant salvinia in the biological control side but to date it has been minor.
  
  As in July, another release site with a large mat of salvinia continues to be reduced. Despite prevailing winds which force the salvinia directly into the study square area, the weevils are keeping the surrounding water open.  There is a thick peat mat forming a few meters outside the square as salvinia dies and is compressed. Giant salvinia plants are actually growing on the peat mat and under heavy attack by the weevils. Here, however, the damaged plants don’t sink.  We are finding many weevils in fairly small samples of plants indicating that the weevil population at this site is probably very high.
  
  The Lake Texana site yielded four weevils from the release square and giant salvinia coverage increased dramatically in the general area.
  
  We also shipped several hundred weevils to cooperators at Lake Conroe for release and establishment.
  
  Giant salvinia biomass and coverage remained high at our control sites with an average of 46 tons of wet plants per acre and 100% coverage. In contrast, biomass at the release sites averaged less than 2 tons per acre with zero for both biomass and coverage at the Louisiana site.
  
  Our plans are to continue evaluating until winter while redistributing weevils to other infested areas. We will visit the sites again in September.

September 2003

• Weevils maintain biomass reduction of Salvinia molesta
  
  The research plots in Texas and Louisiana were visited and data collected during September 22-24, 2003. The Louisiana golf course pond saw giant salvinia increase slightly on the biological control side of the boom while tertiary growth stage salvinia completely covered the surface of the sprayed side of the boom.  Unlike August, we did find one weevil in the salvinia on the sprayed side indicating that weevil populations, while low, are still present. Unfortunately, the land owner has decided to use Sonar to destroy all the salvinia so we will not be able to learn what equilibrium level the salvinia will attain. Current evidence from the biological control side of the boom would indicate that any equilibrium level would probably be low. The treatment will be applied next spring or summer, thereby ending the research value of the site.  Again, this is unfortunate because it is clear that, given enough time, the weevils will eliminate most of the salvinia at this site at no cost to the land manager.
  
  Giant salvinia continues to decline at the 200 acre site in east Texas and weevil numbers remain high.  If weevil numbers continue to increase, it is likely that the mat will collapse sometime in 2004, perhaps as early as mid-summer.
  
  Weevils are persisting at the Lake Texana site which has remained relatively stable, allowing giant salvinia densities to increase.  (Salvinia molesta densities are still persisting, while the race between weevil and Salvinia densities continues until a critical point is reached where Salvinia mats begin to collapse.  This is a long process, usually taking 24-36 months after the first release of the weevils.  Before the weevils can build up to a critical density (~ 200 adults/square meter), the site must remain under stable conditions, which is not happening at the Texana site. The site is experiencing flooding which thins out weevils and provides new space and nutrients for the plants to recolonize.) Conditions are perfect for weevil spread and increase at this sheltered site as long as droughts or especially floods do not intervene.
  
  Giant salvinia biomass and coverage remained high at our control sites with an average of 44 tons of wet plants per acre and 100% coverage. In contrast, biomass at the release sites averaged less than 2 tons per acre.
  
  Our plans are to continue evaluating until winter while redistributing weevils to other infested areas. This fall we plan to release weevils at our control sites as well. We will visit the sites again in October.

October 2003

• Giant salvinia mat is dramatically decreased in east Texas while weevils demonstrate exceptional dispersal abilities
  
  The research plots in Texas and Louisiana were visited and data collected during October 20-22, 2003. As in the previous month, the Louisiana golf course pond saw giant salvinia increase slightly on the biological control side of the boom while tertiary salvinia continues to completely cover the surface of the sprayed side of the boom. Unlike August, we did find one weevil in the salvinia on the sprayed side indicating that weevil populations, while low, are still present. The good news here is that the land manager has apparently been persuaded not to use sonar^tm on the pond, but instead allow the weevils to control the infestation.  If this holds up, we should gain a great deal of insight on the equilibrium dynamics once the main mat is destroyed, i.e. how low will it go?

There has been a dramatic decline in the giant salvinia mat at the 200 acre site in east Texas and weevil numbers remain high.  When this project started, we estimated the total fresh weight biomass in the pond to exceed 20,000 tons. Most of the site is now open water and 7 weevils were found feeding from a small sample in a narrow fringe of salvinia on the far side of the lake opposite the study square. This may be demonstrative of the likely equilibrium that will emerge after the collapse of the larger mat.  The plants that were sampled were growing amongst the emergent vegetation around the periphery of the lake and were almost unnoticeable. It also demonstrates the dispersal ability of the weevil which has spread completely across the long axis of the pond.  Another 53 adults were found around the study square from a very small sample of salvinia.  What is left of the mat will probably collapse by next summer which will make this site an excellent location for examining the equilibrium dynamics that will follow.  We appreciate the access to the site given by the landowner, Gary Nelson.  

There was also a dramatic event at the Lake Texana site, but this one was wholely negative: the site was mistakenly treated with herbicide by the river authority. This unfortunate event destroyed the continuity of the experiment but will not impact giant salvinia in the long run. We did find one weevil in the study square but the majority of the population has probably dispersed from the site because of a lack of suitable plant material, much of which is brown and rotting from the herbicide application.  The lack of a stable, non-disturbed site makes evaluation very difficult and slows the impact of the biocontrol agent. On the bright side, it’s likely that the weevils are now established in the area. We are currently re-evaluating our future commitment to working on this lake.

Giant salvinia biomass and coverage remained high at our control sites with an average of 43 tons of fresh weight biomass per acre and 100% coverage. In contrast, biomass at the release sites averaged less than 2 tons per acre.

This was our last trip this year.  We will be shipping weevils for release at our control sites this week after performing pathogen screening.  Our next trip will be in March 2004.

March 2004

• Early spring finds low weevil numbers and winter damaged giant salvinia poised for a rebound, all as expected for the time of the year

The research plots in Texas and Louisiana were visited and data collected during March 22-24, 2004, our first visit since Oct. 2003. No weevils were released on this trip. The Louisiana golf course pond site had a reversal of salvinia coverage, namely, less than 1% coverage over the previously herbicide treated portion and 70% coverage at the biocontrol end. We removed the boom dividing the site and will now manage the infestation with new releases of weevils, coupled with targeted spraying if needed for aesthetics. Because no weevils were found and most of the salvinia buds remained undamaged, we anticipate rapid regrowth.  Recolonization of the weevils is expected to lag behind salvinia regrowth, hence our plans to augment the population.  This pond has also served as a source for weevil-infested salvinia, many samples of which have been removed by interested parties for placement in other salvinia infested areas.

On our way into Houston airport we flew over the 200 acre pond in east Texas and estimate this second site is now only 10% covered by giant salvinia.  Weevils were found along the fringe of salvinia extending several hundred meters from the original release location. The biomass of salvinia was less than 50 pounds per acre fresh weight.  It should be remembered that in August 2001, one month before the release of weevils, fresh weight biomass stood at more than 95 tons per acre. This site has remained undisturbed and will be studied extensively this year to gain insight on the equilibrium dynamics of salvinia under attack by the weevil.

The Lake Texana site was recovering from the accidental herbicide application from last fall.  Despite this, we did find a weevil on the salvinia inside the previous.

May 2004

• April trip finds weevils arresting Salvinia molesta in Texas; Louisiana receiving an integrated approach
  
  The research plots in Texas and Louisiana were visited and data collected during April 19-21, 2003.  We released a total of 400 adult weevils. The Louisiana golf course pond site showed increased coverage by giant salvinia and samples from four locations around the pond yielded no weevils.  Over the last two years this site has been through a succession of complete coverage and large salvinia biomass, to complete control by the weevil on one half of the divided pond, to likely dispersal and/or manual collection of adults for releases elsewhere, to flushing out by flooding, to recolonization by giant salvinia. An integrated approach is now underway that will re-establish weevils and still allow limited herbicide treatments for aesthetic purposes that will leave a fringe or refuge of salvinia for the weevils, thereby discouraging them from dispersing. Accordingly, we released 100 adults in the area of the pond most heavily infested.  More releases are planned.  Adjacent biomass measurements yielded an average of 20.2 pounds of fresh weight salvinia per acre although with the patchiness of the existing mat, it is probably more than that.
  
  We expected that the 200 acre pond in east Texas would show an increase in coverage (up from 10% in March) but were pleasantly surprised that coverage had actually decreased to about 5%.  As in March 2004, we found many weevils along the now even more narrow fringe of salvinia at every sample point, including those hundreds of meters from the original release location.  In addition, surrounding canals once choked with salvinia are now filled with dark brown, waterlogged plants which will probably sink before our next trip. The biomass of salvinia is still less than 50 pounds per acre fresh weight but, unlike the Louisiana golf course pond, it is probably less than that on a per acre basis. It should be remembered that in August 2001, one month before the release of weevils, fresh weight biomass stood at more than 95 tons per acre. This site has remained undisturbed and will be studied extensively this year to gain insight on the equilibrium dynamics of salvinia under attack by the weevil.
  
  A new site (private pond) near Lake Texana was visited and 100 weevils were released. This site will continue to receive weevils but no extensive evaluations will be conducted. The old Lake Texana site was evaluated but received no weevils and none were found. However, it is likely that the weevils are established, particularly in this backwater site and time is all that is required for the eventual suppression of salvinia.
  
  Our two previous check sites received 100 weevils each. Fresh weight biomass in these sites averaged 15.7 and 47.2 tons of salvinia per acre, respectively.  We will be releasing weevils every month to provide our long-suffering and patient land owners relief from this weed. As always, we appreciate their patience and cooperation throughout the course of this project.

June 2004

• Integrated approach adopted at Louisiana golf course pond and weevils prevail in Texas

The research plots in Texas and Louisiana were visited and data collected during June 7-9, 2004.  We had shipped a total of 275 adult weevils to the sites two weeks earlier. The Louisiana golf course pond site is now completely covered by giant salvinia and samples from four locations around the pond yielded no weevils. However, in the study square we found significant damage to salvinia, indicating that the newly released weevils are settling in and starting over again.

Our experimental portion of the pond was initially separated from the rest by a boom, permitting the golf course manager to maintain for aesthetic reasons the larger portion with herbicide spray. By July 2003 the weevils had completely destroyed the salvinia in our area and, coupled with spraying in the front portion of the pond, resulted in a sudden and drastic reduction in their food source.  At that time the salvinia weevil population numbered in the thousands.  Adult weevils either starved or dispersed, driving their populations to near zero and allowing the salvinia to resurge. This site should serve as an example of how not to manage salvinia after weevils are released.  It would have been better to not spray and to remove the boom, allowing the weevils to move to the front of the pond and attack the remaining salvinia.  The key lesson to learn here is to avoid massive changes in salvinia populations. Striving for a more controlled reduction in plants will permit the weevils to adjust their numbers accordingly, a strategy that should promote a sustainable equilibrium between weevil and weed. One problem in this approach is that it requires patience, a virtue that some lands managers find difficult to practice if they can get a quick (but temporary) kill of salvinia with herbicides.  An integrated approach is now underway that will re-establish weevils and still allow limited herbicide treatments.   The new approach will leave a fringe or refuge of salvinia for the weevils, thereby discouraging them from dispersing.

In stark contrast to the upset at the golf course pond, the 200 acre pond in east Texas is an example of what can happen if there is no interference and the weevils are allowed to work.  The coverage of this pond is now conservatively estimated it at less than 1%, even less than was found in April.  The width of the fringe of salvinia around the pond has shrunk once again from an average of 13 meters wide in April, to less than 2 meters.  This fringe is almost unnoticeable because of other floating and emergent vegetation. It should be remembered that in August 2001, one month before the release of weevils, fresh weight biomass stood at more than 95 tons per acre. It now stands at much less than one ton per acre and is probably overestimated at 1500 pounds per acre.

A new site (private pond) near Lake Texana received additional weevils but was not visited on this trip. The old Lake Texana site was evaluated but received no weevils and none were found.  However, it is likely that the weevils are established, particularly in this backwater site and time is all that is required for the eventual suppression of salvinia.

Our two previous check (control) sites each received 75 weevils before our trip and bud damage from weevils was evident. At one site we found 5 adults, including two light brown weevils, indicating that a new generation has developed. None were found at the second check site.  Fresh weight biomass in these sites averaged about 48 tons per acre.  More weevils will be released to provide our long-suffering and patient landowners relief from this weed.  As always, we appreciate their patience and cooperation throughout the course of this project.

Our next visit to these sites will not be until September or October 2004.

November 2004

• After establishing the salvinia weevil and proving its ability to control Salvinia molesta in Louisiana and Texas, Phil Tipping, USDA/ARS now draws the five year study to completion.

The research plots in Texas and Louisiana were visited and data collected during Oct. 18-22, 2004.  The Louisiana golf course pond site is virtually clear of giant salvinia. This same site was completely covered in June following a classic herbicide upset which illustrated the problem of dramatically reducing the plant density with herbicides, thereby driving the weevils away and resulting in a rapid rebound of the weed.  Examinations of the few damaged plants along the fringe of the pond yielded multiple weevils on each plant. The land manager originally planned to spray a portion of the pond but held off as he witnessed the plants turning brown and sinking.  He asked us to transfer weevils to another pond on the course which he had been spraying for years. This smaller site was completely covered with salvinia.  We collected salvinia from a nearby cove in Toledo Bend and placed it in the pond.  We estimate conservatively that we transferred about 200-300 adults.

The 200 acre pond in east Texas continues to exhibit open water with an estimated coverage of less than 1%.  The waters of this pond were noticeably more clear on this trip.  It should be remembered that in August 2001, one month before the release of weevils, fresh weight biomass stood at more than 95 tons per acre. Just over three years later it now stands at less than 200 pounds per acre and that is probably an overestimate. We consider our work here completed and pulled out the study square.

The old Lake Texana site was examined but no weevils were found. This site has become dominated with water hyacinth to the point where we had to hike in because the boat could not get through. Despite this, it is likely that the weevils are established, particularly in this backwater site and time is all that is required for the eventual suppression of salvinia.

Adult weevils were found at both of our two previous check sites.  At one site we found 8 adults.  This site had significantly lower biomass when compared to historic data although the pond remains covered.  Although the weevils are established and beginning to have an impact, we plan to turn over these sites to APHIS for augmentative releases in order to provide our long-suffering and patient land owners relief from this weed in a shorter period.  As always, we appreciate their patience and cooperation throughout the course of this project.

This concludes our research effort to establish and evaluate Cyrtobagous salviniae on giant salvinia in the eastern Texas and western Louisiana region.  We have conducted the most thoroughly documented study of suppression of this plant by C. salviniae and have proven the value of this biological control agent in the United States. The weevils will continue to spread naturally throughout this area and provide sustainable suppression of this weed in perpetuity. We would like to thank our local cooperators, especially the Texas Parks and Wildlife Department, as well as all the land owners and managers who provided access to their properties.

FAQs on Biological Control Program

Biological Control Program by Philip Tipping, Ph.D., Research Entomologist,
USDA-ARS, Invasive Plant Research Laboratory
Updated: 02 November 2004

Home || News & Notes || US Distribution || Biological FAQs
Identification || Contributor Acknowledgements || Bibliography