Treat wetlands with herbicide
Overall effectiveness category Likely to be ineffective or harmful
Number of studies: 4
Background information and definitions
Wetlands can become overgrown with vegetation, particularly if they have unusually high nutrient levels, or if invasive plants (with no natural predators) are present. Very dense vegetation can exclude open-water species from a wetland, meaning that treating the wetland with a herbicide may be beneficial. However, conservationists should be aware that some herbicides, such as glyphosate, can be extremely toxic to amphibians and fish (Relyea 2005).
Relyea, R.A. (2005) The lethal impact of Roundup on aquatic and terrestrial amphibians. Ecological Applications, 15, 1118–1124.
Supporting evidence from individual studies
A controlled, replicated before-and-after study in 1990-1993 in 23 inland wetlands in North Dakota, USA (Linz et al. 1996), found that the densities of three songbird species were all significantly lower on sites sprayed with glyphosate, than on unsprayed sites (red-winged blackbirds Agelaius phoeniceus: 0.6 birds/ha on sites where 90% of the site was sprayed vs. 1.6 on controls; yellow-headed blackbirds Xanthocephalus xanthocephalus: average of 2.1 birds/ha on all treated sites vs. 3.1 on controls; marsh wrens Cistothorus palustris: 0.7 birds/ha on treated sites vs. 2.2). Experimental wetlands had significantly lower percentage covers of emergent vegetation. Sites were sprayed from the air with either 90%, 70% or 50% of the site treated. There were no differences between sites before herbicide application.Study and other actions tested
A randomised replicated study in 24 inland wetlands in North Dakota, USA (Linz & Blixt 1997), found that the number of black terns Chlidonias niger using sites in June 1991-1993 was positively correlated with the areas of open water and dead cattail Typha spp. present, following the aerial application of glyphosate during July 1990 and 1991. The numbers of mallard Anas platyrhynchos, blue-winged teal A. discors, northern shoveler A. clypeata, gadwall A. strepera, northern pintail A. acuta, redhead Aythya americana and ruddy duck Oxyura jamaicensis were all correlated with the amount of open water and the amount of cover present. Glyphosate was sprayed over 90%, 70% or 50% of the sites.
A controlled, replicated before and after study in 1990-1993 in 20 inland wetlands in North Dakota, USA (Linz et al. 1997), found that, two years after treatment, densities of American coot Fulica americana were significantly higher at wetlands sprayed with herbicide (0.8-1.0 birds/ha) than at untreated sites (0.2 birds/ha), but that sora Porzana carolina densities were significantly lower (0.1-0.3 birds/ha vs. 0.5). Coot densities were positively correlated with extent of open water and negatively with live emergent vegetation; sora densities were positively correlated with live emergent vegetation. Four sites had 90% coverage with glyphosate (aerially applied), eight had 70% coverage, four 50% and four were controls, with no coverage. No coots were found on any wetlands before treatment.Study and other actions tested
A controlled study in 2003-2004 on mudflats and areas of Spartina alterniflora meadows in Willapa National Wildlife Refuge, Washington, USA Patten & O\'Casey 2007), found that average densities of unidentified Calidris spp. sandpipers (62 birds/ha vs. 7), western sandpiper Calidris mauri (50 vs. 5) and waterfowl (16 vs. 0.8) were significantly higher on areas of Spartina meadow spayed with glyphosate (9 kg/ha) in 2002-2003, and imazapyr (1.7 kg/ha) in 2004, compared to control areas. However, densities of dunlin Calidris alpina, grey plovers Pluvialis squatarola and dowitchers Limnodromus spp. did not differ, all birds were less common than on ploughed areas of Spartina and all but dowitchers and waterfowl were less common than on adjacent mudflats.Study and other actions tested