Can Brambles Be Eradicated From Prairies and Savannas? A Long Story

Brambles (genus Rubus) may be native but they are generally undesirable in a prairie or savanna. This post is based on over 20 years of restoration work at Pleasant Valley Conservancy.

The story starts when Kathie and I first started clearing invasive brush and trees from prairie and savanna remnants at PVC in 1997. There was lots of bad stuff: shrubs such as buckthorn, honeysuckle, and prickly ash and trees such as elm, cedar, red pine (planted), box elder. If there were brambles, we did not notice them. Probably they were shade-suppressed.

But once the site was opened up, brambles started to flourish. Originally they were just small patches, which in our ignorance we ignored. In 1999 some of these bramble patches were large enough that we got rid of them. There may have been some larger patches, but in a property as large as Pleasant Valley Conservancy, it was easy to overlook them.

What was the source of these brambles? Both buried root stock and buried seeds are likely. I doubt the root stock is long lived, but Rubus seeds are notoriously long-lived. (A U.S. Forest Service study in New England found viable Rubus seeds that had been buried for 95 years.) 

The Ecology of Brambles

Brambles are members of the genus Rubus. They have perennial roots systems but biennial tops. The roots continue to spread but new canes are formed each year. The first year cane just grows vegetatively and in the second year flowers and forms fruits. After fruit formation it dies. Thus, a bramble patch is a mixture of first year and second year canes, as well as dead canes which senesce.

We have three species of Rubus at PVC: blackberries, black raspberries, red raspberries. All are bad, and we do not distinguish between them, although red raspberries presents a special challenge.

The First Survey

It was 2005 before brambles started to be so obvious that we could not ignore them. That year I made a survey, marking each patch with a numbered stake. There were 17 patches, all in open savannas. My notes in the table here give some idea of how large these bramble patches were.

Data from June 5-6 2005

Stake #	Unit #	Location	Notes
1	12A	~75 ft below the upper road (below Toby’s); large patch in flower	White Oak Savanna
2	12A	~75 ft farther down the hill from #1; edge of Unit 12E; large patch in flower	White Oak Savanna
3	12A	Smaller patch but taller stems west of #1	White Oak Savanna
4	12C	Large patch among two large dead oaks; also outliers east and downhill	White Oak Savanna
5	12C	About 75 ft east of trail down; along a drainage; follow the drainage up and down	White Oak Savanna
6	11A	Edge of open area and edge of ridge; shorter plants but lots; forming berries	Bur Oak Savanna
7	11A	Near road; low but lots forming berries	Bur Oak Savanna
8	11A	50 fit West of #7 near fallen oak	Bur Oak Savanna
9	13B	Lots of 1st year plants mixed in with Geraniuum etc. Can cut it all. Whole top of knoll, also	Woods near east property boundary
10	17	N of Toby’s in former aspen zone; scattered but large patch; 1st year plants	Between Toby’s NE corner and woods
11	11B	Above Paul Ihm’s wood pile	Bur Oak Savanna
12	11A	SE corner of saddle; large patch with little good; mow up hill and to the east also	Bur Oak Savanna
13	23	Large patch of low brambles; uphill to the west from the saddle; do the whole patch	East-facing hill above Hickory Ravine
14	7	Very large patch uphill from Hickory Ravine; mow down hill and to the east
15	6	Top of hill near fire break; 75 ft west of side road; large patch of low brambles; mow down the hill and into the gully
16	10	Across the fire break from stake #15; big patch mixed in with brush; cut it all and then treat the brush	Should have a second person with you for treating the cut stumps
17	5C/5B	West of Unit 4 along and below the fire break on both sides of the gully; huge area	This is a two-three person job!

Getting Serious

At that time the only information I could find about how to eradicate brambles was: Mowing several times a year will reduce their vigor.

Reducing vigor did not sound like a promising method for eradication.

By the late fall of 2005 Kathie and I had decided that our goal should be to get rid of all brambles, as well as other invasive brush that had sprung up in our restored areas. After all, we had spent a lot of effort opening up these prairies and savanna. If we did not get rid of all these new invaders, we could be back where we started. An awful thought!

We experimented with several different methods of herbicide application: foliar spraying the bramble patches (low herbicide concentration); basal bark (high concentration, in oil); cut and treat the cut stems (high concentration, either aqueous of in oil). We rejected foliar spraying of the large patches because of the possible peripheral herbicide damage. We tried basal bark, but application from a backpack sprayer was also “wasterful” of herbicide, and because of the high concentration, there were bare zones around the treated patches. We also tried a sponge method for basal bark, but the thorns on the brambles tore the sponges to shreds.

Cut-and-treat method

We chose the “cut-and-treat” method because only small amounts of herbicide would be used, and it could be applied carefully by a spray bottle. Because of the extent of the infestations, this was usually a four- to five-person job. One person operated the Stihl brush cutter and the others followed along with spray bottles containing 50%  glyphosate. Since we were working primarily in the winter, when it was below freezing we diluted the glyphosate with windshield washer fluid. Where there was a lot of buckthorn, we used Garlon 4 in oil, which presented no freezing problem in the winter. The “treaters” fanned out behind the brush cutter and treated each cut stem.

A great advantage of this method is that it could be done any time of year, and especially at times when weed work, plant surveys, seed collecting, etc. could not be done. We worked with a combination of hired employees and volunteers.

I ran some preliminary versions of this method in late November 2005. We then set up a major effort on the White Oak Savanna starting in early January 2006. Todd Shumate from Michler & Brown ran the brush cutter and three or four “treaters” followed behind. Todd cut every woody shrub in the savanna. Brambles constituted at least half of the brush, but honeysuckle and prickly ash were substantial.

The brush cutter was equipped with a saw blade. Todd ran the brush cutter until he ran out of gas (about 45 minutes per tank). He then sharpened the saw blade with a special Stihl file. This kept the cut stems clean and flat so they could be treated properly. Cut stems that were dead were not treated. (Living cut stems are green even in the winter.)

The First Major Effort

The table gives an example of this work for a five-acre white oak savanna. Finishing the whole unit took 9 days, working about 40 person hours per acre. Fortunately, January of 2006 was surprisingly mild and snow-free. Often we were in shirt sleeves at noon.

Table shows brush cutting work on Unit 12A (the White Oak Savanna). Total area 5.3 acres

	2-Feb-06	31-Jan-06	26-Jan-06	24-Jan-06	19-Jan-06	17-Jan-06	12-Jan-06	5-Jan-06	3-Jan-06	Totals
Todd	5.5	5.25	7	6.75	6.75	6.75	5.5	6.75		50.25
Susan	5.5				5.75	5.25	5.5	5.75		27.75
Elke	5.5		6		5.75					17.25
Tom	5.5	4.9	6	5.5	5.75	5.25	5.5	4	2	44.4
Kathie	5.5	4.9	6	5.5	5.75	5.25	5.5	5.75	2	46.15
Brian	2.25				4.25			2		8.5
Final total										194.3
Averages>>	36.7	hrs/acre
	If assume $25/hr, acreage cost is
	$917.50
	Roughly $1000/acre

Extending the Effort (2006-2007)

Once the White Oak Savanna work was completed, we extended the work to the rest of the areas where brush and brambles were rampant. The table shows the 2006-2007 work (81 work days; 30 days in 2006 and 51 days in 2007.)

In 2007 I established winter intern positions, and we had 3 (or 4) UW-Madison students working two days a week starting in January and continuing until mid May.

Date	Unit	Notes
7 Feb 2006	18	Lots of honeysuckle; brambles; prickly ash; walnut resprouts
9 Feb 2006	18	Lots of honeysuckle; brambles; prickly ash; walnut resprouts
14 February 2006	18	Lots of honeysuckle; brambles; prickly ash; walnut resprouts
21 Feb 2006	6	Mostly honeys; small buckth; few brmbles; some dogwood
23 Feb 2006	2 below Diag Trail	Lots honeys few buckth; fee brmbles
28 Feb 2006	12B	Mostly brambles
2 Mar 2006	12B
7 Mar 2006	3 (around big bur oak)	Mostly brambles
7 Mar 2006	19C	Buckth mostly
10 Mar 2006	6/23	Lots little buckth
11 Mar 2006	6/23
14 Mar 2006	11	Big bramble; buckth; prickly ash
21 Mar 2006	11
23 Mar 2006	10	Lots tiny buckth; brambles
28 Mar 2006	6 (east)	Honeys; brambles; s umac
30 Mar 2006	19
17 Nov 2006	7	Lower part and adjacent to 18; shrubs/brambles small honeys; cut & treat glypho
19 Nov 2006	7	small honeys buckth brambles; few walnuts
21 Nov 2006	7
27 Nov 2006		Lots of notes about cut & treat and then plant
30 Nov 2006	3/6	Gully above big bur; lots brambles; some honeys; few buckth; grape
5 Dec 2006	6/9	Gully between 6 & 9; see record for details of how we did it
12 Dec 2006	6/9	top of gully; gray dogw
15 Dec 2006	6/9	Continue from 12 Dec
17 Dec 2006	12A	Check for brush/bramble problems; 9 areas listed; mostly brambles (all 3 spp)
19 Dec 2006	6/9	gully/brambles
26 Dec 2006	6	gully below S fire break;
28 Dec 2006	6	Finish unit 6
28 Dec 2006	3/9	above gully
2 Jan 2007	5	Up to fire break
2 Jan 2007	3
3 Jan 2007	4/9	Winter interns; paint stick!
4 Jan 2007	3/4/9	lots of honeys on Unit 9; walnuts; Autumn olive!
8 Jan 2007	19A	Top, near Toby’s; Chris instead of Todd
9 Jan 2007	3/4/9
10 Jan 2007	3/5	gully above diagonal trail
11 Jan 2007	5B/C
15 Jan 2007	20	near Pocket Prairie; snowed 5 inches the day before; worked anyway
17 Jan 2007	20	Also bur oak grove near chain
18 Jan 2007	20	Along woods road; lots blackberry lots hazel elderberry
22 Jan 2007	20
24 Jan 2007	20	Finish Unit 20
25 Jan 2007	21	snow too deep move to top of hill
26 Jan 2007	20
29 Jan 2007	21	West of Ridge Prairie
30 Jan 2007		too cold, did not work
31 Jan 2007	21	snow mostly gone
1 Feb 2007	21	Just T & K
5 Feb 2007		Below zero in morning; work cancelled all week
12 Feb 2007	12A	Lots of dead buckth but some alive
13 Feb 2007	19E/11E	19E too windy; moved to 11D
14 Feb 2007	11D
15 Feb 2007	11D	large blackberry patch; Mid Savanna Trail
16 Feb 2007	11D	finished this unit
19 Feb 2007	2	gully above 2 bur oaks
20 Feb 2007	2/3	finish gully up to fire break; lots sumac; some honeys; some buckth; rare bramble
21 Feb 2007	2	above diagonal trail
22 Feb 2007	2	above diagonal trail
26 Feb 2007	7	Lower & middle on west side
27 Feb 007	7	Middle & upper; around East Overlook rocks
28 Feb 2007	6/7	Up from Hickory Ravine into Unit 6
2 Mar 2007	Ravine	Ravine between 12A and 20B; lots of snow drifts!
5 Mar 2007	14	Corner PV Rd &  Cty F; snow deep!
5 Mar 2007	1	Less snow; road cut for Unit 1
6 Mar 2007	5A
6 Mar 2007	20B	after lunch finish 20B
7 Mar 2007	14	below quarry
8 Mar 2007	5A	Up to Ridge Trail
8 Mar 2007	6	Afer lunch, 5A too sloppy, moved to 6 lower east around rocks, lots of honeys but small
12 Mar 2007	5A
13 Mar 2007	5A	Above Unit 1 and along Ridge Trail; FINISH WHOLE SOUTH SLOPE!!
13 Mar 2007	Quarry	After lunch
14 Mar 2007	2/19	Finish Unit 2; start Unit 19
15 Mar 2007	6	along dolomite shelf
15 Mar 2007	19
20 Mar 2007	19D/E
22 March2007	19C/D
28 Mar 2007	20C/19D	Finish 20C
4 Apr 2007	19C
5 Apr 2007	19C

Fire and Brambles: Spraying Resprouts

Brambles are fairly fire sensitive and are killed by a controlled burn.

Fire moving through a bramble patch in a savanna. All  bramble stems are killed.

Starting in 2008 we added another method of bramble control: foliar spraying of resprouts after the bramble canes had been killed by early spring burns. The details of this method are covered in several Tom’s Blog posts in 2008, 2009, and 2010.

Although this method is very effective, it is limited to May or early June, when the resprouts are small enough to spray without too much peripheral damage. To set up the spraying routine, I created another bramble database. By then I had learned how to  use ArcGIS. On 14 and 16 April 2009 I walked the whole south side of PVC, recording the coordinates of each patch

After a burn, bramble patches are fairly easy to spot, as the following photo shows.

Large bramble patch easily seen after a spring burn. Unit 11B; red raspberry patch

Typical resprouts after the brambles have been top-killed by fire; Mid-May

Spraying bramble resprouts about mid-May. Any other woody invasives such as
buckthorn or honeysuckle are also sprayed. Often a crew of 4-5  would work on the same unit

Table used to create an ArcGIS map of brambles. The date burned was used to determine which units to spray first.

Patch	Lat	Lon	Unit	Species	Date burned	Priority	Notes
1	43.10559567	-89.80980858	Cabin	Blackberry	4/3/2009	1	Small
2	43.10625558	-89.81188151	Crane	Blackberry	4/3/2009	2	Moderate
3	43.10738856	-89.81244938	Crane	Blackberry	4/3/2009	6	Huge!
4	43.10493216	-89.80763666	Valley	Blackberry	4/3/2009	4	Large in line
5	43.10473401	-89.80720634	Valley	Blackberry	4/3/2009	1	Small
6	43.10489771	-89.80679244	Unit 7	Red raspberry	3/19/2009	4	Large-blackberry + red raspberry
7	43.106673	-89.80296275	Unit 12B	Black raspberry	4/8/2009	2	Moderate
8	43.10670745	-89.80279796	Unit 12B	Black raspberry	4/8/2009	2	Moderate
9	43.10649136	-89.8028195	Unit 12A	Red raspberry	4/8/2009	3	Both red and black raspberry
10	43.10642933	-89.80255279	Unit 12A	Red raspberry	4/8/2009	6	Huge!--also hazel
11	43.10747255	-89.80242924	Unit 19E	Black raspberry	4/3/2009	3	Loose patches scattered
12	43.10759794	-89.80236797	Unit 19E	Blackberry	4/3/2009	1	Small plus burdock
13	43.10737649	-89.80267533	Unit 19E	Black raspberry	4/3/2009	1	Small, near road
14	43.10721648	-89.80272344	Unit 11B	Red raspberry	4/8/2009	2	By road, across from 19E, linear along road
15	43.10718882	-89.80436982	Unit 11B	Red raspberry	4/8/2009	6	Very large! Also clone S and W of AP1
16	43.10696955	-89.8050278	Unit 11B	Red raspberry	4/8/2009	6	Very large clone around stake AP2
17	43.10694985	-89.80539626	Unit 11A	Red raspberry	4/8/2009	6	Very large clone around stake AP3
18	43.10729401	-89.80591837	Unit 19C	Red raspberry	4/8/2009	5	Large clone near road
19	43.10664257	-89.80375467	Unit 11D	Blackberry	4/8/2009	2	100 feet W of Junction of the two trails
20	43.1063953	-89.80440527	Unit 11D	Red raspberry	4/8/2009	5	In middle of unit, large clone
21	43.10640126	-89.80471448	Unit 11D	Red raspberry	4/8/2009	3	In middle, farther toward gully than BR20
22	43.10649178	-89.80462773	Unit 11D	Red raspberry	4/8/2009	1	25 ft S of Mid Sav Trail; Small clone
23	43.10664115	-89.80473007	Unit 11C	Red raspberry	4/8/2009	1	Above Mid Sav, near trail, near large hickory; small clone
24	43.10663964	-89.80479612	Unit 11C	Red raspberry	4/8/2009	1	Small clone, short stems, above Mid Sav
25	43.10649354	-89.80546198	Unit 11C	Blackberry	4/8/2009	3	Near Mid Sav and just E of large white oak
26	43.10438147	-89.80366163	Unit 20	Black raspberry	4/18/2009	6	By woods road, very large clone
27	43.10416186	-89.80353146	Unit 21	Black raspberry	4/18/2009	1	Fairly small, near woods road
28	43.10581922	-89.79962222	Unit 13	Black raspberry	4/8/2009	6	Top of knoll, very large clone, also down all sides
29	43.10412599	-89.80182314	Unit 21	Black raspberry	4/18/2009	1	At S end of Ridge Prairie, smallclone
30	43.10433671	-89.801943	Unit 21	Black raspberry	4/18/2009	3	West of BR29, larger, fairly loose clone
31	43.10670761	-89.79977419	Unit 13	Black raspberry	4/8/2009	5	N side, down hill from knoll and near fire break separating Toby's N; very large but loose
32	43.10691012	-89.79992732	Toby's N aspen	Blackberry	4/18/2009	1	Near fire break to Unit 13, small
33	43.10757849	-89.80411207	Unit 19D	Red raspberry	4/8/2009	3	Middle betw woods road and N fire break, med-sized clone
34	43.10703333	-89.80819951	Unit 10	Black raspberry	4/8/2009	1	Near road and bur oak #704,fairly small clone
35	43.10691431	-89.8089574	Unit 8	Black raspberry	4/8/2009	1	Near Parking, small clone near downed logs
36	43.1066963	-89.8089258	Unit 8	Black raspberry	4/8/2009	6	Very large loose clone; lower SE side of unit
37	43.10709964	-89.8086979	Unit 19B	Black raspberry	4/8/2009	5	fairly large, near end road and AP-8
38	43.10714708	-89.80853898	Unit 19B	Black raspberry	4/8/2009	6	very large, up from N fire break and E of AP-8
39	43.10716728	-89.80809876	Unit 19B	Red raspberry	4/8/2009	1	small clone by road
40	43.10708078	-89.80763918	Unit 19B	Red raspberry	4/8/2009	6	very large clones of all three spp., between woods road and N fire break--MAJOR JOB
41	43.10713316	-89.80734958	Unit 19B	Black raspberry	4/8/2009	5	Large patch, about 1/2 did not burn
42	43.10599859	-89.80708379	Unit 23	Black raspberry	3/19/2009	1	Near saddle road; may have been sprayed?
43	43.10604746	-89.80672413	Unit 11A	Black raspberry	4/8/2009	6	Up from saddle road, E of AP stake, very large clones almost merging; runs NE/SW along the edge of the ridge, about 15 ft E of AP stake
46	43.10588418	-89.8065357	Unit 11A	Black raspberry	4/8/2009	2	Downhill from BR43, right next to saddle road; smallish clone
47	43.10568452	-89.8063311	Unit 11A	Black raspberry	4/8/2009	6	Very large clone between saddle road and Mid Sav Trail; all day job for 1 person!!
48	43.1054303	-89.80646202	Unit 7	Red raspberry	3/19/2009	5	Below saddle road;large scattered loose; from Hickory Ravine to rock outcrop
49	43.10424141	-89.80488572	Valley	Black raspberry	4/3/2009	3	Far E end by fire break; narrow patch along break; also some East of the fire break
50	43.10448431	-89.80189255	Unit 21	Black raspberry	4/18/2009	5	Near top, W of Ridge Prairie; large; good area to plant
51	43.10472043	-89.80143623	Unit 21	Blackberry	4/18/2009	2	Moderate
52	43.10612004	-89.80752091	Unit 6	Black raspberry	3/19/2009	2	Top of gully near S fire break; moderate
53	43.10624929	-89.80765586	Unit 6	Black raspberry	3/19/2009	5	Up from gully near bur oak grove; large loose clone
54	43.10630059	-89.80788318	Unit 6	Black raspberry	3/19/2009	2	Nearer gully than BR53, mixed in with sumac

By the beginning of 2010 we had fairly good control of brambles except for red raspberry, which forms really dense patches of brambles with stems growing very close together. The cut-and -treat method is not recommended because the cut stems are so close together and it is difficult to keep track of which ones had been treated. Fortunately, most of the areas where we were cutting and treating did not have red raspberries.

After almost ten years working on bramble control, I found that the best way to control red raspberry is to spray (with 3.5% aqueous Garlon 3A) all the resprouts that arise after burns. However, not all shoots come up at the same time, so it is necessary to spray again two weeks after the first time, and again after another two weeks. Monitoring again in mid-summer for any “escapes”, and again the next year is also advisable.

The importance of annual burns

One of the most important reasons why annual burns should be done in oak savannas is because of brambles. Putting savannas on a three-year burn cycle, such as is often done for prairies, is exceedingly ill advised. Savannas almost always have a “legacy” of woody invasives, of which brambles are one of the most important. Giving woody vegetation the chance to grow unimpeded for three years does serious damage to the herbaceous understory. In addition, the savanna will not carry a fire well with all those woody shrubs hogging the savanna floor.

We have been burning our savannas annually for 15 years (since 2002).

The best time to search for undesirable woody plants is after a burn. This photo, of the Unit 10 savanna, was taken on March 28, 2017 (the spring burn had been done on Feb 22, 2017). Note the absence of woody vegetation (except for the blow-downs). You could walk from one end of this savanna to the other without encountering a single obstacle. Compare with the 2008 photo shown above.

Bur oak savanna after 15 annual burns
Note the absence of woody invasives