Seed bank dynamics of French broom in coastal California grasslands: effects of stand age and prescribed burning on control and restoration

  • Published source details Alexander J.M. & D'Antonio C.M. (2003) Seed bank dynamics of French broom in coastal California grasslands: effects of stand age and prescribed burning on control and restoration. Restoration Ecology, 11, 185-197.


In coastal grasslands of California, USA, exotic French broom Genista monspessulana (native to Europe) may be invasive, forming dense stands and accumulating an abundant seed bank belowground. Prescribed burning has been used in attempts to reduce broom cover and the seed bank. This study examined the effect of repeated burning on French broom and associated grassland species. The relationship between broom stand age and seed bank size was also examined.

Study sites: The effects of prescribed burning on invasive French broom Genista monspessulana was undertaken on coastal grassland sites within protected areas in Marin County, California, southwest USA. The climate is Mediterranean, with cool wet winters and warm dry summers. Common species in the study areas include many naturalized annual grasses of European origin, e.g. oat Avena spp., annual fescue Vulpia spp. and brome Bromus spp., with a mix of native and exotic forbs.

Sites were chosen with dense French broom cover. In a few cases, Scotch broom Cytisus scoparius (another introduced European species) was also present but where it occurred made up less than 5% of plants with a correspondingly small seed bank. French and Scotch broom have similar life histories and for the purposes of this study, no distinction between their seeds was made.

Stand age and seed bank sampling: In autumn 1998 and 1999, 13 broom stands that had established within grasslands were sampled. Stand age was assessed by counting growth rings of plants with the largest stem diameter. These ages are an estimate of the minimum stand age, as in California, Scotch broom plants live up to 17 years, and French broom around 15 years, so the oldest plants may have died by the time of sampling. However, land managers verified the age estimates.

The seed bank of each broom stand was sampled with soil cores (5.5 cm diameter x 10 cm depth) at regular intervals along transect lines. For small stands (<6 m diameter), one transect was placed through the centre and pairs of cores taken every metre. In larger stands, two transects were placed to best cover the centre area, and up to 10 cores taken. A soil core depth of 10 cm was used as Scotch broom seeds do not germinate below 8 cm depth and very few broom seeds occur below 10 cm. The volume of soil collected was within the range recommended for characterization of grassland seed banks.

Autumn 1998 samples were spread individually into small trays with potting soil and placed in a lathehouse to germinate from November 1998 to August 1999. Samples were regularly watered. Broom germination in California occurs from December through July. Seedlings of all species were identified, counted and removed as they appeared.
Seedlings were categorized as follows:

Native perennial grasses - typically the focus of grassland restoration and management activities

native forbs - important components of diversity and indicators of grassland health
exotic annual grass and exotic forb s - primarily of Mediterranean origin and making up a large percentage of California grassland species and cover

Exotic perennial grasses - of European origin, relatively recent introductions and potentially very aggressive, invaders

"California Exotic Pest Plant Council (CEPPC) group" - species highlighted as causing serious problems in California wildlands

After germination appeared to have ceased, the soil was sifted to remove any ungerminated broom seeds. Dormancy was checked by scarring the seed coat with a scalpel and placing the seed on damp filter paper overnight. If the seed turned green and imbibed water it was classified as shallowly dormant. Viability for those scarified but not germinating, was assessed via tetrazolium red tests. Non-germinating but viable seeds were classified as deeply dormant. All non-germinating and non-viable seeds were classified as dead.
In autumn 1999, broom seeds were sifted directly from the soil and counted and classified as germinable, deeply dormant or dead.

Seed data were used to calculate mean number of seeds for each stand and seeds/m² to a 10 cm depth. Sites were grouped together in age classes: 5–6, 8–9, 10–11 and 13–15 years old, with two to three sites per age class. Relationships between seed number and stand age were assessed.

Seed bank response to burning: In autumn 1998 and 1999, the seed bank was sampled across stands that had either never burned or burned from one to four times, and also in two sites immediately before and after controlled burns to look at direct effects of fire on the number of viable seeds in the soil. There is generally a 1- to 2-year period between burns, which allows broom seedlings to germinate and grow but not to set seed. Standing broom is cut by hand in the summer and left it to dry until burning in autumn. Fires generally do not burn evenly and are affected by weather and topographic conditions. At each site, soil cores were taken and seed germinated following the same methods in 1998 and 1999 as described above.

Stand age and seed banks: The study sites ranged in age from 5 to 15 years old, with broom seed bank ranging from 900 to 10,582 seeds/m². One site had three to four times as many broom seeds as any other. It was located on a flatter slope than most of the other sites, thus perhaps allowing more seeds to collect and remain in the soil, or it may have been a better site for seed production or the results due to sample bias. This site was not included in further analyses.

Variation among sites was large, and there was no clear relationship between the age of a broom stand and the abundance of its soil seed bank. Younger and older stands had fewer seeds than those stands in the middle age range. There was a slight negative correlation between number of broom seeds and stand age.

There was no statistical differences between the different age classes for any of the other categories of species, but there was a trend toward decreasing seed numbers as broom stands aged. Proportions of dormant and dead broom seeds in the soil varied between sites and stand ages. There were no statistically significant relationships between age and the number of dormant or dead seeds in the seed bank. However, there was a trend toward an increasing percentage of deeply dormant seeds with increasing stand age.

Burning and the seed bank: Across the four sites sampled, burn history ranged from never burned to burned four times. The broom seed bank at these sites ranged from 52 to 10,582 seeds/m², with far more broom seeds in the unburned site. Seed banks were similar between sites burned once and more than once. Pre- and post-burn seed bank data show that repeated burning does not significantly decrease the seed bank or result in more dead broom seeds. Reductions in the broom seed bank are therefore likely due to germination rather than fire-induced mortality.

Though the broom seed bank did not change with repeated burning compared to a single burn, other species were affected. At one site, there was a trend toward increasing exotic seeds with the first two burns. The exotic seed bank in twice-burned plots was equal to (exotic perennial grasses, exotic forbs) or greater than (exotic annual grasses, CEPPC species) the seed bank of the never-burned stand. A third burn decreased exotic perennial grasses, exotic annual grasses and CEPPC species, whilst native forb seed increased.

Pre- and post-burn samples generally concur with these trends, specifically a decrease in exotic grass species and an increase in native forbs were found after the third burn. CEPPC species however, appeared to decrease in the seed bank after two or three burns. Unfortunately, due the low replication of treatments, it is difficult to say if these observed patterns are really attributable to repeated burning, or the specific nature of the individual fires.

The primary exotic forb at one site were scarlet pimpernel Anagallis arvensis and the main CEPPC species was Italian ryegrass Lolium multiflorum, but neither are considered a great management threat.

Exotic annual grass seeds were significantly higher at the four-time burned plots, with a similar trend in exotic forbs and CEPPC species. At the same time, native forbs decreased by about 40% compared with unburned areas. The native seed bank at a four-times burmed site was 150–300 seeds/m², whereas exotics were 2,000–9,300 seeds/m².

Across treatments and sites, there was no effect of burning on the proportions of shallowly dormant (50%, n = 91), deeply dormant (40%, n = 112) and dead broom seeds (7%, n = 115) in the soil.

Conclusions: This study (concurring with others) indicates that fire reduces the size of the broom seed bank. The soil seed bank of burned stands had fewer broom seeds than unburned sites but repeated burning did not reduce the seed bank beyond that of one burn. The prescribed fires had no consistent effect on the seed banks of any other species. There was no strong relationship between broom stand age and broom seed bank size, with data merely suggesting that the broom seed bank stays constant or declines slightly with stand age. However, non-broom seed numbers did decrease as stands aged.

Thus, whilst burning is an effective means of reducing broom cover and initially reducing the seed bank, the authors state (as others) that follow-up treatments are an essential part of broom control, but it appears that repeated burns are not the most effective means of continuing to reduce the broom seed bank or promoting native species.

Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at:

Output references
What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.

Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust