Floodplain forest loss in the Upper Mississippi Valley is concerning because of the habitat these forests provide, especially for birds. Native floodplain forest systems are increasingly susceptible to reed canarygrass (Phalaris arundinaceae; RCG) establishment and spread. Once invaded by RCG, ecological restoration of floodplain requires integrated treatments that reduce the existing RCG population, along with additional long-term control measures. A common secondary measure is tree or shrub plantings to provide shade and reduce the competitive ability of this shade-intolerant invader.
Experiments can provide insight on whether invasive plant dominance is caused by superior competitive ability (driver) or by environmental changes that facilitate plant invasion (passenger). Reed canarygrass (Phalaris arundinacea, hereafter RCG) displaces native plants and forms near-monocultures in North American wetlands. In the Upper Mississippi River (UMR) system, floodplain forests are negatively impacted by RCG invasion. We tested two RCG control techniques on a reforestation project at four sites in SE Minnesota. Treatments consisted of (1) applying glyphosate (Rodeo) herbicide and (2) mulching followed by applying sulfometuron methyl (Oust XP) herbicide. Treatments were applied in Fall 2016. We monitored herbaceous plant response and RCG performance over the 2017 and 2018 growing seasons. We also calculated the number of days flooded in each plot for the 2017 growing season using linear interpolation of river gauge data. Both treatment methods significantly reduced RCG performance relative to controls during the 2017 growing season; however, RCG performance in treatment plots was similar to control plots during the 2018 growing season. Herbaceous plants increased in species richness and cover relative to control plots, although volunteer plant diversity varied among sites. These results indicate that follow-up herbicide applications are necessary to control RCG to facilitate tree establishment. Further, flooding and lack of native propagules may be factors in RCG invasions. Our results indicate that RCG may behave as a driver in some parts of the UMR and as a passenger in parts of the UMR that are more affected by hydrologic alteration.
Floodplain forests provide ecosystem services such as nutrient storage and rapid biogeochemical cycling which may reduce transport of nitrogen (N) downstream. Invasion by nuisance plant species, however, may modify a floodplain's ability to capture N by altering soil properties, litter decomposition rates, N availability, and rates of N cycling. I examined the effect of flooding on soil properties and N cycling at a floodplain site in Pool 8 of the Upper Mississippi River with two different plant communities: mature native forest (Acer saccarinum) and patches of an invasive grass (Phalaris arundinacea). Plots were established within each vegetation type along an elevation gradient and sampled throughout the summers of 2013 and 2014. Spatial trends in flooding resulted in better conditions for microbial activity in low elevations. Nutrient processes and NH4 and NO3− availability, however, were best explained by vegetation type and time after flooding. Phalaris plots maintained higher rates of nitrification and higher concentrations of available NH4+ and NO3−. These results suggest that invasion by Phalaris may make nitrogen more readily available and could help to reinforce this species' persistence in floodplain wetlands. They also raise the possibility that Phalaris may decrease floodplain N storage capacity.
Reed canarygrass (Phalaris arundinacea L.) is a desirable pasture grass on wet areas in the Northern United States and Southern Canada, but it is a serious and troublesome ditchbank weed in the Pacific Northwest and Rocky Mountain States. The purpose of this study was to learn more about the development, growth habits, and control of reed canarygrass on irrigation ditchbanks. Ninety-seven percent or more of the seed of this species germinated ediately after harvest under favorable conditions. Seeds stored in damp sand at constant temperatures of 1 and 23C for periods of time up to one year did not germinate until they were subjected to alternating temperatures of 20 and 30C. The first rhizome development on reed canarygrass seedlings grown in the greenhouse was observed 26 days after emergence. Within 16 weeks after emergence, the plants were in bloom and had 48 short rhizomes (6.5 cm maximum) per plant. In the field, 88 percent or more of the emergent: shoots on established plants originated from rhizome or tiller buds located in the upper 5 cm of soil. Some shoots developed from buds located at depths up to 20 cm, but none arose from a greater depth. Several vegetative characteristics of reed canarygrass plants collected from six irrigation projects in four states differed widely when grown in a garden at Prosser, Washington. The plant height, seed weights, panicle length, leaf length, leaf width, number of stems per plant, stem diameter, and the rate of spread by rhizomes were statistically different at the 5% level of probability. Large differences in the color and posture of the leaves were also observed. Plants collected near Huntley, Montana, were the most vigorous. Total available carbohydrates in the roots and rhizomes of established reed canarygrass were not affected by single applications of 2,2,dichloropropionic acid (dalapon) at 22 kg/ha, 3-amino-s-triazoleammonium thiocyanate (amitrole-T) at 4.5 kg/ha, or 1,1i-dimethy1-4,4'- bipyridinium ion (paraquat) at 1.1 kg/ha until 2 months after treatments were applied in May. By October, single applications of dalapon and amitrole-T and five repeated applications of paraquat reduced the carbohydrates 24, 28, and 50 percent, respectively. Two additional treatments of dalapon or amitrole-T or five additional treatments of paraquat in the second year of the study did not reduce the carbohydrate levels below those present during the first year. Dalapon and trichloroacetic acid (TCA), applied to the soil or to the senescent foliage in November at rates from 22 to 88 kg/ha, provided good to excellent temporary control of reed canarygrass without denuding ditchbanks. Redtop (Avostis alba L.) and reed canarygrass seedlings developed on the treated areas the following summer and reed canarygrass retained the dominant position unless the seedlings and plants that escaped the fall treatment were controlled with post-emergence applications of dalapon or amitrole-T. When the latter plants were controlled, redtop developed from natural or artificial seeding of the ditchbank and became the dominant species. Maximum residue levels of TCA in irrigation water ranged from 104 to 225 ppb following fall applications of TCA at 82 kg/ha to both banks of three irrigation laterals that ranged from 4 to 14.5 kilometers long. Average residue levels at the downstream ends of the laterals during the first four hours that water flowed through them in the spring ranged from 34 to 47 ppb. Eight hours after the initial flow of water through laterals 4, 8.4, and 14.5 kilometers long, residue levels were less than 1 ppb in the two shortest laterals and only 2.7 ppb in the longest lateral. No residues were detectable in the water from any of the laterals after 48 hours.
