Use this URL to cite or link to this record in EThOS:
Title: The ecology and control of Japanese knotweed (Reynoutria japonica Houtt.) and Himalayan balsam (Impatiens glandulifera Royle.) on river banks in South Wales
Author: Beerling, David John
ISNI:       0000 0001 2436 6777
Awarding Body: University of Wales College of Cardiff
Current Institution: Cardiff University
Date of Award: 1990
Availability of Full Text:
Access from EThOS:
The distribution of R. japonica and I. glandulifera on river banks within the Welsh Region of the National Rivers Authority is described. R. japonica and I. glandulifora were found to be widespread along most of the rivers in the region, occurring on 84% and 71% respectively of the rivers surveyed. The distribution of R. japonica on selected rivers was also investigated in relation to land use. The results showed that on riparian wasteground and land drainage works, R. japonica occurred more frequently than expected and at a higher abundance: on grazed land or in natural/semi-natural communities of river banks it occurred less frequently and in low abundance. The results of a general soil survey to describe the nutrient status of soils on which both species typically grew in South Wales are presented. The survey results suggested that both species are capable of growing on fertile and relatively infertile soils, although from the sites sampled, I. glandulifera tended to exist in nature on soils with a higher nitrogen and phosphorus content than R. japonica. Aspects of the ecology of R. japonica were investigated, these included (i) above-ground growth analysis at open and shaded sites; (ii) below-ground growth measurements (expansion of clumps and growth of rhizomes); and (iii) rhizome fragment viability. R. japonica responded to shade with reduced biomass and stem height, and an increased Leaf Area Ratio compared with plants grown at open sites. These responses were consistent during the two years when observations were made (1988 and 1990). Clump expansion (expressed as radius of the clump) was independent of clump size, but dependent on environmental conditions. Cutting clumps down increased the lateral spread of the plant relative to the uncut clumps. Rhizome fragments of ≥ 7.8g (wet weight) were viable and produced shoots above-ground within 50 days. Aspects of the ecology of I. glandulifera were also investigated, these included (i) growth analysis in open and shaded sites; (ii) germination of seedlings; (iii) seedling mortality; (iv) effect of plant density on seed production; and (v) seed dispersal. I. glandulifera responded to shade with reduced biomass (but similar Relative Growth Rates and stem heights) compared with plants growing at the open sites. Seeds collected from plants of I. glandulifera had high ( > 79%) germination success, and were mostly dispersed within 3.5m of the parent stand. Seedling mortality was density dependent within the range measured in the study (80-300 plants m⁻²). Increased plant density also resulted in reduced seed production per plant. A simple model to describe temporal population changes using the results from investigations (ii) to (v) above was developed. The model simulated population changes in numbers of adult plants over a 10 year period. Field trials were conducted to assess quantitatively the effectiveness of foliar applications of non-persistent herbicides (approved for use in or near water courses) at controlling riparian stands of both species. The results suggested that one application of 2,4-D amine at 2790g active ingredient ha⁻¹ early in the season (May-June) would control I. glandulifera and prevent the development of a viable seed bank. Two applications per season at the same dose using 2,4-D amine or glyphosate with 2154 g a.i. ha⁻¹ would control R. japonica, but 2,4-D amine was preferable because it allowed grass swards to persist after treatment. The extent to which treated stands might recover in succeeding years has not been established and needs further study. On the banks of the River Cynon, Aberdare, South Wales, seven designs of flood revetment blocks were tested for resistance to penetration and displacement by R. japonica. With the subsequent failure of all designs tested, the development and testing of a new block is described. Its success is attributed to a microporous structure and interlocking/overlapping edges. The invertebrates associated with the foliage and litter of two introduced (R. japonica and I. glandulifera) and two indigenous (Urtica dioica and Epilobium hirsutum) species were surveyed in May, July and September 1988 to provide comparative estimates of family diversity, and attempt to assess the likely conservation impact of these introduced plants where they are widespread and abundant. The invertebrate faunal assemblages associated with the foliage of the two introduced species were impoverished relative to the foliage of the two native species sampled, with fewer taxa and fewer individuals. In the leaf litter the effect of the introduced plant species was less marked, with numbers of animals and taxa being reduced in the September samples only. An initial assessment of the flooding hazard, which R. japonica represents on river banks, was made. The results suggested that flooding might occur September-October when living plants had a high biomass and river flows are relatively high. Local authority estimates are given for the cost of repairing damage caused by the plant and for its control. It was concluded that when control measures are undertaken costs are high, even when the affected areas are small, due in part to the continued management required in subsequent seasons to ensure eradication of the plant. Aspects of the study relevant to local authorities are discussed in relation to improved management practices which may prevent or restrict the spread of the plant more effectively. Opportunities for further research are outlined.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Ecology ; Botany