2009 Dry Creek Sesbania Monitoring Report

Red Sesbania has a bright red flower and a medium, oval-shaped leaf.

Executive Summary

In 2009, the Dry Creek Red Sesbania Control Program continued and two control sweeps of the watershed were conducted; one in July and one in October. A November monitoring visit conducted by SAFCA indicated that less than 1% of the Sesbania population remained. Therefore, the 2009 program met its success criterion. A comparison of 2008 and 2009 monitoring results and control data is presented, and a paper was prepared estimating the time to eradication and control costs for Red Sesbania in Dry Creek.


The Sacramento Area Flood Control Agency (SAFCA) conducted an invasive plant control program for red sesbania (Sesbania punicea)(Sesbania) in the Dry Creek watershed from 2004 to 2006. The project was known as the Dry Creek Watershed Red Sesbania Control Project, Phase I (Phase 1 Project). The Phase 1 Project was funded by a California Department of Water Resources Proposition 13 Flood Protection Corridor Program grant administered by SAFCA. Since 2007, the maintenance phase of the removal program has been funded through a partnership between SAFCA, Sacramento County, Placer County, and the cities of Sacramento and Roseville. The project is administered by Placer County Resource Conservation District. The goal of the project is to reduce the potential for flooding in Dry Creek and to improve wildlife habitat in the watershed. The objective is to continue to remove and control seedlings of Sesbania, anticipating a diminishing level of effort over time.

This report presents the results of the 2009 monitoring effort. Monitoring was conducted by Ramona Robison with the assistance of SAFCA staff in June, and by SAFCA staff in November. The 2009 Dry Creek Sesbania removal contractor was Agri-Chemical and Supply, Inc., of Salinas, California. Gary Omori, Agri-Chemical’s Salinas office Area Manager, directed the removal effort. Agri-Chemical conducted two control “sweeps” of the watershed and monitoring was conducted before sweep 1 and after sweep 2.

The contract performance criterion was:

Table 1. Annual Success Criterion
Year Anticipated Condition at Start of Work Year Success Criteria at End of Work Year Overall Goal
Year 1 (2008)
  • 10-30% cover in lower watershed
  • > 10% cover in upper watershed
  • Removal of 99% of red sesbania by October 15
  • Eradication of Red Sesbania
  • < 1% remaining in treatment area at end of program


In 2008, the Dry Creek Sesbania monitoring program changed to meet requirements of grants and to provide better information on the condition of the infestation. In previous years a series of over 50 monitoring photographs were taken throughout the watershed. These photographs were useful during the early phases of the project when large shrubs were being removed, creating obvious “before and after” photographs. However, for the past few years the emphasis of control work has been on seedling removal which is not as easily monitored using photographs alone. The performance criterion for the project is removal of 99% of the Sesbania cover, and a quantitative means of assessing cover before and after treatments was needed. The methods developed in 2008 were repeated again in 2009, and a monitoring protocol was developed for future use (Appendix A [PDF*, 1.9 MB]).

Monitoring locations are also listed in Table 2.
Figure 1. 2009 Dry Creek Red Sesbania Monitoring Locations

In 2009 we monitored the Sesbania infestation in six locations, collecting data on four 1 meter square quadrats at each location. Sampling locations with sand bars exposed during low flows were chosen to represent the upper, middle, and lower watershed (See Figure 1, above, and photographs in Appendix B [PDF*, 3 MB]). Several of the locations in the upper watershed (Royer Park in Roseville and downstream of Walerga Road) had few Sesbania plants, but were chosen to represent the condition in that portion of the Dry Creek watershed. Our choice of quadrat locations within the sandbar was also biased — we tried to choose quadrats with at least one live or dead Sesbania plant in them, even if the majority of the sandbar was free of Sesbania plants. This resulted in a higher cover classification than if randomly placed quadrats were used. However, the sampling unit was really the entire sandbar and the Sesbania occurring there was the focus of the monitoring effort. Permanent quadrats were also not chosen because of the dynamic nature of the river system.

During monitoring visits we took photographs of each quadrat and photographed a six foot tall pole marked off in one foot intervals to record overall vegetation height. We collected percentage cover data in each quadrat for live Sesbania alone and for all other live plant species present. The cover classes used are listed in Table 2. We counted all live Sesbania plants in the quadrat and measured the height of 10 plants to determine mean height per quadrat.

The first monitoring visit was completed on June 24, 2009 by Ramona Robison and Caitlin Talkington of SAFCA. Agri Chemical and Supply, Inc. conducted their first sweep July 20 to 28 and their second sweep October 9 to 12. Only two sweeps of the watershed and two monitoring visits were conducted in 2009 due to budget constraints. The final monitoring visit was completed on November 12, 2009 by KC Sorgen and Sarah Somers of SAFCA.

Results and Discussion

Ground is lush with Sesbania.
Before Sesbania Treatment 6-24-09
Ground is bare where treatment was completed.
After Sesbania Treatment 11-12-09

Photographs of the “before and after” condition of the Sesbania infestation in Dry Creek and its tributaries are presented in Appendix B [PDF*, 3 MB]. The table below summarizes the cover classes observed in June (before) and November (after). Data sheets with all quadrat information are included in Appendix C [PDF*, 54 KB].

Once again, in June, the largest number of live Sesbania plants per square meter was at the SAFCA property north of Hansen Ranch where we counted 82 plants in one of our quadrats. During the November visit the largest number of live plants observed in any of the quadrats was 13 which were treated plants that had started re-sprouting. In many cases there were no live Sesbania plants in the quadrats (Walerga and Hansen locations).

The cover class information presented in Table 2 presents the highest and lowest cover classes observed in the four quadrats at each monitoring location. It should be noted that sample locations within the monitored sandbar were not randomly chosen. An attempt was made to place each quadrat in a location with either a live or dead Sesbania plant. All locations had 5% cover or less in November. When live Sesbania plants were observed during monitoring visits we measured them and then pulled them by hand. There were so few plants remaining in the November visit that we were able to remove all live plants in the six monitoring locations. Most of the quads observed in November had 1% or less Sesbania cover with only three quads overall containing 5% or less Sesbania cover.

Table 2. 2009 Dry Creek Sesbania Monitoring Data Summary
Monitoring Location June Cover Classes November Cover Classes June Mean Number of Plants per Meter November Mean Number of Plants per Meter Notes
Hidden Valley, Granite Bay 4, 3, 1, 2 1, 2, 1, 1, 1, 1, 1 16.75 2.3
Royer Park, Roseville 2, 3, 2, 5 1, 1, 2, 2 14.75 4.5
Walerga Road, Placer County 2, 3, 2, 3 1, 0, 0, 0 8.75 NONE
Gibson Ranch, Elverta 2, 3, 4, 4 1, 1, 1, 1 15.5 NONE
Above Dry Creek Road 2, 6, 5, 4 Not monitored 53 Not monitored
Roy Hayer Park, Rio Linda Not monitored 1, 1, 1, 1 Not monitored 1.75
Upstream of Hansen Ranch, Rio Linda 3, 4, 4, 4 1, 0, 0, 1 56.3 2.5
Cover Classes: 0-1% = 1; 1-5% = 2; 5-25% = 3; 25-50% = 4; 50-75% = 5; 75-95% = 6; 95-100% = 7

Outside the six monitoring locations, a few Sesbania plants were observed by monitors downstream of Royer Park. They had not flowered and were approximately 4 feet tall. Sesbania plants were also observed re-sprouting in some locations during the November monitoring visit (see November Appendix C [PDF*, 54 KB] Monitoring Forms).

In 2009 we again collected information on the plant species associated with Sesbania and results were similar to 2008. Many quadrats contained grasses, sedges (Cyperus spp.), rushes (Juncus spp.), willows (Salix spp.), knotweed (Polygonum spp.), fireweed (Epilobium spp.), mugwort (Artemesia douglasiana), beggar’s ticks (Bidens frondosa), and valley oak seedlings (Quercus lobata). Other non-native and invasive plants were also observed including Himalaya berry (Rubus discolor), plantains (Plantago lancelata, P. major), cut-leaf geranium (Geranium dissectum), cocklebur (Xanthium strumarium), sweet clover (Melilotus sp.), cudweed (Gnaphaliumsp.), and Mexican tea (Chenopodium ambrosioides).

In 2009 we observed a few other invasive plants of note while monitoring for Sesbania. Waterprimrose (Ludwigia sp.) was observed upstream of Dry Creek Road and Acacia seedlings were observed upstream of Hansen Ranch during 2009 monitoring.

Comparison of 2008 and 2009 Results

The following information compares results of 2008 and 2009 monitoring and control efforts on Dry Creek.

Table 3. 2009 Dry Creek Sesbania Monitoring Data Comparison for June 2008 and 2009 Visits
Monitoring Location 2008 June Mean Number of Plants per Meter 2009 June Mean Number of Plants per Meter Notes
Hidden Valley, Granite Bay 12 16.75
Royer Park, Roseville 2 14.75
Walerga Road, Placer County 3.5 8.75
Gibson Ranch, Elverta 12.5 15.5
Roy Hayer Park, Rio Linda (2008)
Dry Creek Road Upstream (2009)
16.8 53
SAFCA Property North of Hansen Ranch, Rio Linda 72.5 56.3

Other Invasive Species Observations

In 2008, Chinese tallow (Sapium sebiferum) seedlings were observed in one quadrat in June and two quadrats in October in Roseville at Royer Park. This species was not observed in monitoring quads in 2009. Similarly, Eucalyptus seedlings were observed in June and October on the SAFCA property north of Hansen Ranch in 2008 and were not observed in 2009.

Summary of 2009 Weed Removal Contractor Work

Gary Omori and Rich Gresham.
Gary Omori and Rich Gresham

Appendix D [PDF*, 22 KB] contains a summary prepared by Gary Omori, manager of weed removal with Agri-Chemical. The following table is summarized from his results. Note that the number of tanks used and number of bags of seeds collected decreased in 2009 compared to 2008.

Table 4. Dry Creek Sesbania Weed Removal Contractor Summary 2008 and 2009
Removal Dates Number of Tanks1 Herbicide Used Number of Bags2 Collected Number of Operator Hours
July 15 to 23, 2008 434 70
September 8 to 12, 2008 32.5 14
July 20 to 28, 2009 230 25 430
October 9 to 12, 2009 80 4 200
  1. 1 tank load is approximately 4.5 gallons of mixed herbicide solution.
  2. 1 seed bag is 33 gallons.


The 2009 treatment year was successful in removing 99% of the germinated Sesbania population. The contractor also removed all Sesbania pods from the floodway, preventing future germination of those seeds. Since the project started in 2004 the focus has been on removal of upstream seed producing Sesbania plants, and given the reduction in seed production each year (i.e., all seed pods removed), the seedbank should continue to decline over time. While results this year are promising, treatments must continue until the population is reduced to low levels. The size of the Dry Creek Sesbania population increased with the heavy rainfall in 2006 due to high water flows bringing seeds from low-flow channel sediments up onto the upper floodplain where they germinated far from the channel. Therefore the level of treatment effort needed in high rainfall years is expected to increase. This year, 2009, was the first year it was possible to compare the Sesbania percentage cover and seedlings per square meter in order to determine if the amount of seed germinating from the seedbank is decreasing over time. As Table 3 shows, only one monitored location showed a decrease in the number of seedlings per meter.

In order to estimate the time to eradication and costs involved in management, Gina Darin and I prepared a paper for the California Invasive Plant Council’s annual symposium in October 2009. The paper is included in this report as Appendix E [PDF*, 40 KB]. We found the model we chose to use, WeedSearch, under-predicted the amount spent on Sesbania removal and management, and predicts that the time to eradication may be as long as 22 years. However, since Sesbania did not begin to rapidly expand its range in California until after 2000, we do not have any direct evidence to indicate the actual time to eradication.

Report prepared by: Ramona Robison, Ph.D., California Botanical Surveys and Tours, 1925 Meer Way, Sacramento, CA 95822.