Metal tolerant and non-tolerant plants from two grass species (<i>Festuca rubra</i> and <i>Agrostis capillaries)</i> were grown in standardised conditions with control, copper, zinc and salt treatments. Leaf reflectance properties (400 – 1100 nm) of the different treatments were measured using an integrating sphere. Pigment concentration analysis was also undertaken. Differences between treatments were examined using the raw reflectances, as well as a number of published and novel vegetation indices and red edge position. Non-parametric statistical tests were used. Tolerant plants showed different spectral responses to non-tolerant plants irrespective of treatment. No spectral analysis technique consistently showed a stress response in all non-tolerant treatments, although different techniques did show different stress responses. In the second part of the study canopies of monocultures and mixtures of tolerant and non-tolerant ecotypes of <i>Festuca rubra</i> were grown in plots with control and zinc treatments. Reflectance measurements (400 – 2500 nm) were made under artificial lighting using a spectroradiometer mounted inside a light proof tent. Pigment analysis was also undertaken. Spectral analysis techniques were the same as those used in the first part of the study. No vegetation index gathered from the literature distinguished the control from the metal treatments for the monoculture plots, but many were successful for the mixture plots. Indices developed during this study based on the green-red region were successful. Leaf and canopy reflectance models were used to stimulate reflectance changes following community responses to metal stress. These showed that the detection of contaminated ground via vegetation stress requires that the area being surveyed is either comprised of a non-tolerant ecotype which is stressed (i.e. before being outcompeted by tolerant plants) or that the contamination level is so high the tolerant ecotype shows stress. The results indicate that the use of remote sensing to detect contaminated ground via vegetation stress is far from straightforward. To be successful the measurements have to be made during a community transition from non-tolerant stressed plants to tolerant plants (although this transition could vary from weeks to years). The measurements should also be made over the senescence time period, to locate areas senescing early. As many methods of analysis (indices, red edge position etc.) as possible should be used. The users also have to accept that false negative results will occur where tolerant populations are on contaminated ground. The cause of stress will not be identifiable.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:652103 |
| Date | January 2005 |
| Creators | Hardy, Nicholas J. |
| Publisher | University of Edinburgh |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/1842/24679 |
Page generated in 0.1637 seconds