Although long-term fire retardants (LTR) gain increasingly acceptance as effective tools for wildfire management, recent studies indicate their potential harmfulness in the environment, especially in aquatic ecosystems. This study comprises a first laboratory impact assessment of LTR action in temporal Mediterranean wetlands, using seed germination of Typha domingensis as indicator of impact. Our aim was to identify application rates upon which seed germination could be significantly affected. We tested for low (11 m(-2)) and high (31 m(-2)) application rates of Fire Trol 934((R)) which are recommended by the manufacturers as a function of fuel characteristics. In addition, we simulated the impact of a higher application rate of 51 m(-2) because inhomogeneous dispersal of the LTR during fire control and prevention operations can result in locally elevated applications. Results of a microcosm experiment indicate that application rates of I or 3 1 m(-2) can impact Typha germination rates in the short-term via indirect LTR-mediated effects on water quality, which suppressed necessary cues for germination. However, a subsequent experiment with Petri dishes, using seeds isolated from the LTR treated sediments did not show significantly different germination rates between the control and the treatments with application rates of I or 31 m(-2). This suggests that retardant pre-application germination success could be recovered in nature once the retardant is eliminated. By contrast, seeds almost completely failed to germinate in the microcosm experiment and the subsequent Petri dish essay when an application rate of 51 m(-2) was used. This suggests a critical level upon which Typha seed germination may be perpetually limited. Research should be extended to other plant species to provide fire managers with guidelines for environmentally safe use of LTR in the Mediterranean region. (C) 2003 Elsevier Ltd. All rights reserved.
