Relationship between leaf traits and fire-response strategies in shrub species of a mountainous region of south-eastern Australia

Abstract

Background and aims: Resprouting and seed recruitment are important ways in which plants respond to fire. However, the investments a plant makes into ensuring the success of post-fire resprouting or seedling recruitment can result in trade-offs that are manifested in a range of co-occurring morphological, life history and physiological traits. Relationships between fire-response strategies and other traits have been widely examined in fire-prone Mediterranean-type climates. In this paper, we aim to determine whether shrubs growing in a non-Mediterranean climate region exhibit relationships between their fire-response strategy and leaf traits.

Methods: Field surveys were used to classify species into fire-response types. We then compared specific leaf area, leaf dry-matter content, leaf width, leaf nitrogen and carbon to nitrogen ratios between (a) obligate seeders and all other resprouters, and (b) obligate seeders, facultative resprouters and obligate resprouters.

Key results: Leaf traits only varied between fire-response types when we considered facultative resprouters as a separate group to obligate resprouters, as observed after a large landscape-scale fire. We found no differences between obligate seeders and obligate resprouters, nor between obligate seeders and resprouters considered as one group.

Conclusions: The results suggest that facultative resprouters may require a strategy of rapid resource acquisition and fast growth in order to compete with species that either resprout, or recruit from seed. However, the overall lack of difference between obligate seeders and obligate resprouters suggests that environmental factors are exerting similar effects on species' ecological strategies, irrespective of the constraints and trade-offs that may be associated with obligate seeding and obligate resprouting. These results highlight the limits to trait co-occurrences across different ecosystems and the difficulty in identifying global-scale relationships amongst traits.

Fig. 1.

Location of the study area. (A) The Australian continent, indicating the regional location of the study area as a shaded rectangle in (B). (C) Enlargement of the study area, including the extent of the 2003 fires. ACT, Australian Capital Territory; NSW, New South Wales.

Fig. 2.

Study species (81 in total) represented in a family-level phylogenetic tree. Fire-responses for each species are shown. Twelve species were sampled from more than one location, giving 94 populations in total: *, species sampled from two locations; **, species sampled from three locations.

Fig. 3.

Position of study populations (94 in total) on the first two axes of a principal components analysis (PCA) based on species leaf traits. PC1 = principal component 1, PC2 = principal component 2. Populations are coded by their fire-response. LDMC, leaf dry-matter content; SLA, specific leaf area; LNC, leaf nitrogen content; C:N = carbon to nitrogen ratio; LW = leaf width.

Fig. 4.

Comparison of PCA loading values between fire-response types: obligate seeders vs. all resprouters along (A) principal component 1 (PC1) and (B) principal component 2 (PC2); and obligate seeders. vs. obligate resprouters vs. facultative resprouters on (C) principal component 1 and (D) principal component 2. Error bars indicate s.e. Groups with different letters within each panel are significantly different.