Applying evidence-based practice in conservation management: Lessons from the first systematic review and dissemination projects

Lack of systematic evaluation of the effectiveness of conservation practices has hindered advances in scientific management of biodiversity. Development of an evidence-based framework of the kind used in the health services has been advocated as an approach to address this problem. Here we report on and evaluate the first two ecological systematic reviews undertaken using this framework. The subjects were the effectiveness of burning as a conservation intervention for UK sub-montane, dry dwarf shrub heaths (heath review) and blanket bog (bog review). Systematic search yielded 13 data sets from seven ‘heath’ articles and 11 data sets from eight ‘bog’ articles. Data from the heath review were of sufficient quality to enable meta-analysis whereas data from the bog review was synthesised by “vote counting”. Meta-analysis of the heath data indicates that floristic diversity is variable, particularly in early post-fire successional sequences, but older stands experienced significantly greater loss of diversity than younger stands. The bog review indicates that there are no consistent changes in floristic composition in response to burning. The evidence-base for upland management by burning is insufficient to generate robust management recommendations highlighting the necessity for further work. Feedback from the dissemination of the reviews provided valuable lessons for future reviews in terms of stakeholder involvement in question formulation and the balance between a reductionist and holistic approach. Systematic reviews are appropriate tools for conservation management and formalise the information available based on the weight of evidence. Evidence will commonly be lacking in many areas of conservation biology and the evidence-based framework effectively exposes this situation in a way that can determine requirements for needs-led research.     

Silvicultural and reserve impacts on potential fire behavior and forest conservation: Twenty-five years of experience from Sierra Nevada mixed conifer forests

Most administrators and ecologists agree that reducing the levels of hazardous fuels on forests is essential to restore healthy watersheds and protect adjacent human communities. The current debate over the appropriateness, technique, and timing of treatments utilized to restore vegetation structure and composition is currently on-going at local, state, and national levels. To provide information for these forums, the efficacy of seven traditional silvicultural systems and two types of reserves used in the Sierra Nevada mixed conifer forests is evaluated in terms of vegetation structure, fuel bed characteristics, modeled fire behavior, and potential wildfire related mortality. The systems include old-growth reserve, young-growth reserve, thinning from below, individual tree selection, overstory removal, and four types of plantations. These are the most commonly used silvicultural systems and reserves on federal, state, and private lands in the western United States. Each silvicultural system or reserve had three replicates and varied in size from 15 to 25 ha; a systematic design of plots was used to collect tree and fuel information. The majority of the traditional silvicultural systems examined in this work (all plantation treatments, overstory removal, individual tree selection) did not effectively reduce potential fire behavior and effects, especially wildfire induced tree mortality at high and extreme fire weather conditions. Overall, thinning from below, and old-growth and young-growth reserves were more effective at reducing predicted tree mortality.     

Release, movement and recovery of 3,4-dimethylpyrazole phosphate (DMPP), ammonium, and nitrate from stabilized nitrogen fertilizer granules in a silty clay soil under laboratory conditions

 The composition of soil microbiota in four heated (350  °C, 1 h) soils (one Ortic Podsol over sandstone and three Humic Cambisol over granite, schist or limestone) inoculated (1.5 μg chlorophyll a g^–1 soil or 3.0 μg chlorophyll a g^–1 soil) with cyanobacteria (Oscillatoria PCC9014, Nostoc PCC9025, Nostoc PCC9104, Scytonema CCC9801, and a mixture of the four) was studied by cultural methods. The aims of the work were to investigate the potential value of cyanobacteria as biofertilizers for accelerating soil recolonization after fire as well as promoting microbiotic crust formation and to determine the microbial composition of such a crust. The inoculated cyanobacteria proliferated by 5 logarithmic units in the heated soils which were colonized very quickly and, after 2 months of incubation, the cyanobacterial filaments and associated fungal hyphae made up a matrix in which surface soil particles were gathered into crusts of up to 1.0 cm in thickness. These crusts were composed, on average, of 2.5×10¹⁰ cyanobacteria, 2.8×10⁶ algae, 6.1×10¹⁰ heterotrophic bacteria (of which 1.2×10⁸ were acidophilic, 1.3×10⁶ were Bacillus spp. and 1.5×10⁸ were actinomycetes) and 77.8 m fungal mycelium (1.4×10⁶ were fungal propagules) g^–1 crust. Counts of most microbial groups were positively correlated to cyanobacterial numbers. The efficacy of treatment depended on both the class of inoculum and the type of soil. The best inoculum was the mixture of the four strains and, whatever the inoculum used, the soil over lime showed the most developed crust followed by the soils over schist, granite and sandstone; however, the latter was comparatively the most favoured by the amendment. In the medium term there were no significant differences between the two inocula rates used. Biofertilization increased counts of cyanobacteria by 8 logarithmic units while heterotrophic bacteria, actinomycetes, algae and fungal propagules rose by >4 logarithmic units, acidophilic bacteria and Bacillus spp. by around 3 logarithmic units and fungal mycelia showed an 80-fold increase. The results showed that inoculation of burned soils with particle-binding diazotrophic cyanobacteria may be a means of both improving crust formation and restoring microbial populations. Received: 8 March 2000     

Effect of season, location and fire on Collembola communities in buttongrass moorlands, Tasmania

Collembola of buttongrass moorlands in Tasmania were surveyed monthly over a 12 month period, using sweep nets and pitfall traps, at two locations, one at 760 m and the other at 325 m altitude. At each location, there were two pairs of sites with each pair straddling a fire boundary. The younger regrowth sites had been burnt less than 12 years previously and the older regrowth sites burnt more than 24 years previously. Nearly 30,000 individuals were collected belonging to 42 taxa. Most Collembola were identified to morphospecies with virtually all mature specimens identified to genus level and immature specimens recognised to family level. Three named species were recorded. Genera with Gondwanan (41%) and cosmopolitan (33%) distributions dominated the Collembola of buttongrass moorlands. The Collembola communities differed significantly between locations and this, in part, reflected differences in vegetation. Brachystomellidae, Odontellidae, Paronellidae, Katiannidae n. gen. and Sminthurididae were more common at the high altitude sites where the vegetation contained more grasses and sedges and occurred on relatively fertile soils. Parakatianna spp., cf. Tomocerura spp., cf. Drepanura sp., Sminthurinus spp. and Rastriopes sp. were more common at the low altitude sites where the vegetation contained more shrubs and occurred on relatively infertile soils. The Collembola community differed significantly between seasons with one group of taxa being more common in autumn and/or winter (Acanthomurus spp., cf. Tomocerura spp., Paronellidae, Lasofinius spp., Polykatianna cf. aurea, Parakatianna spp., Sminthurinus spp. and Katianna spp.) and another group of taxa more common in summer or summer and autumn (Odontellidae, cf. Drepanura sp., Rastriopes sp., Corynephoria sp. and Dicyrtomidae). We found some evidence of a difference between Collembola communities in the two ages of regrowth we compared but only for the low altitude sites. Polykatianna cf. aurea, Parakatianna spp., Sminthurinus spp., cf. Tomocerura spp. and Rastriopes sp. were caught in lower numbers on young regrowth sites compared with old regrowth sites. We also present data from another low altitude site where we used a space-for-time design to further investigate the impact of fire on Collembola. We found a significant positive relationship between age of buttongrass regrowth and the number of Rastriopes sp. caught and this was correlated with the ground cover of plant species within the family Myrtaceae.     

Carbon stocks, soil respiration and microbial biomass in fire-prone tropical grassland, woodland and forest ecosystems

A thorough understanding of the role of microbes in C cycling in relation to fire is important for estimation of C emissions and for development of guidelines for sustainable management of dry ecosystems. We investigated the seasonal changes and spatial distribution of soil total, dissolved organic C (DOC) and microbial biomass C during 18 months, quantified the soil CO₂ emission in the beginning of the rainy season, and related these variables to the fire frequency in important dry vegetation types grassland, woodland and dry forest in Ethiopia. The soil C isotope ratios (δ¹³C) reflected the 15-fold decrease in the grass biomass along the vegetation gradient and the 12-fold increase in woody biomass in the opposite direction. Changes in δ¹³C down the soil profiles also suggested that in two of the grass-dominated sites woody plants were more frequent in the past. The soil C stock ranged from being 2.5 (dry forest) to 48 times (grassland) higher than the C stock in the aboveground plant biomass. The influence of fire in frequently burnt wooded grassland was evident as an unchanged or increasing total C content down the soil profile. DOC and microbial biomass measured with the fumigation-extraction method (C_mic) reflected the vertical distribution of soil organic matter (SOM). However, although SOM was stable throughout the year, seasonal fluctuations in C_mic and substrate-induced respiration (SIR) were large. In woodland and woodland-wooded grassland C_mic and SIR increased in the dry season, and gradually decreased during the following rainy season, confirming previous suggestions that microbes may play an important role in nutrient retention in the dry season. However, in dry forest and two wooded grasslands C_mic and SIR was stable throughout the rainy season, or even increased in this period, which could lead to enhanced competition with plants for nutrients. Both the range and the seasonal changes in soil microbial biomass C in dry tropical ecosystems may be wider than previously assumed. Neither SIR nor C_mic were good predictors of in situ soil respiration. The soil respiration was relatively high in infrequently burnt forest and woodland, while frequently burnt grasslands had lower rates, presumably because most C is released through dry season burning and not through decomposition in fire-prone systems. Shifts in the relative importance of the two pathways for C release from organic matter may have strong implications for C and nutrient cycling in seasonally dry tropical ecosystems.     

Activated carbon amendments to soil alters nitrification rates in Scots pine forests

The influence of charcoal on biotic processes in soils remains poorly understood. Charcoal is a natural product of wildfires that burned on a historic return interval of ∼100 years in Scots pine (Pinus sylvestris L.) forests of northern Sweden. Fire suppression and changes in forest stand management have resulted in a lack of charcoal production in these ecosystems. It is thought that charcoal may alter N mineralization and nitrification rates, however, previous studies have not been conclusive. Replicated field studies were conducted at three late-succession field sites in northern Sweden and supporting laboratory incubations were conducted using soil humus collected from these sites. We used activated carbon (AC), as a surrogate for natural-occurring fire-produced charcoal. Two rates of AC (0 and 2000 kg ha⁻¹), and glycine (0 and 100 kg N as glycine ha⁻¹) were applied in factorial combination to field microplots in a randomized complete block pattern. Net nitrification, N mineralization, and free phenol concentrations were measured using ionic and non-ionic resin capsules, respectively. These same treatments and also two rates of birch leaf litter (0 and 1000 kg ha⁻¹) were applied in a laboratory incubation and soils from this incubation were extracted with KCl and analyzed for NH₄⁺ and NO₃⁻. Nitrification rates increased with AC amendments in laboratory incubations, but this was not supported by field studies. Ammonification rates, as measured by NH₄⁺ accumulation on ionic resins, were increased considerably by glycine applications, but some NH₄⁺ was apparently lost to surface sorption to the AC. Phenolic accumulation on non-ionic resin capsules was significantly reduced by AC amendments. We conclude that charcoal exhibits important characteristics that affect regulating steps in the transformation and cycling of N.     

The effect of fire spatial scale on <Emphasis Type="Italic">Bison</Emphasis> grazing intensity

To determine whether fire spatial and temporal scales affect foraging behavior and grazing intensity by Bison (Bison bison), we burned three different patch sizes (225, 900, and 3600 m²) across an otherwise homogeneous grassland landscape. We then monitored grazing intensity for the succeeding 14 months. During the first 5 months after the burn (August–January), the Bison grazing intensity pattern was affected by whether a plot was burned and only marginally affected by plot size. During the next 5 months (January–June), grazing intensity was unaffected by plot size, but was greatest in the unburned 225 and 3600-m² plots. The final 4 months (June–October), grazing intensity was unaffected by treatments other than being higher in the unburned 3600-m² plots. By the final sampling date, biomass was significantly greater in the burned plots and grazing intensity appeared to be responding to the amount of biomass present and the total amount of N present. The pattern displayed within the first 5 months after the burn is congruent with the expectations of optimal foraging theory with overmatching in the smallest plot size of 225 m² (BioScience 37 (1987) 789–799). The next two sampling periods displayed a matching aggregate response relative to biomass availability (Oecologia 100 (1999) 107–117) and total nitrogen mass (g m⁻²). The temporal shift that we found in Bison response to burn patch size is, to our knowledge, the first such examination of both spatial and temporal responses by Bison to landscape heterogeneity. We now have quantitative evidence of how native herbivores can alter their foraging responses to changes in landscape structure over time.     

Spatial patterns of fire occurrence in Catalonia,NE, Spain

In this paper, we analyse spatial patterns of fire occurrence in Catalonia (NE Spain) during 1975–98. Fire scar maps, discriminated by means of 30–60 m resolution remote sensing imagery, have been used as a source of fire occurrence. We employ several visual or analytical approaches to interpret fire occurrence in this region, such as those of Minnich and Chou (1997), Ricotta et al. (2001) or Krummel et al. (1987). Crucial spatial patterns such as fire size distribution, fire frequency distribution, spots and residual vegetation islands are documented. In addition, several geographical layers were overlaid with burned area maps in order to determine interactions between fire occurrence and environmental parameters such as altitude, slope, solar radiation, and burned land cover. Assuming that fire occurrence is well determined by such a posteriori empirical factors we detect areas most prone to fire in this region and aim to enhance the local forest management and conservation plans.     

Factors Responsible for the Co-occurrence of Forested and Unforested Rock Outcrops in the Boreal Forest

Rock outcrops in the boreal forest of Québec can show either of two different states: a forested state with >25% tree cover, and an unforested state (<25% tree cover). We tested three different hypotheses that might explain the co-occurrence of forested and unforested rock outcrops: (1) differences in bedrock geology, with unforested outcrops associated to bedrock types inimical to tree growth; (2) unforested outcrops as recently disturbed sites undergoing secondary succession towards a forested state; (3) unforested outcrops as an alternative stable state to forested outcrops, induced by post-fire regeneration failure. Digitized forest inventory maps were used along with bedrock geology maps and time-since-fire maps to compare forested and unforested outcrops for bedrock geology type and date of the last fire. Field surveys were conducted on 28 outcrops (14 forested, 14 unforested) to gather information regarding tree species composition and site characteristics (thickness of the organic matter layer, percent cover of lichens, mosses and ericaceous shrubs). None of the three hypotheses explain the co-occurrence of forested and unforested rock outcrops in the boreal forest of Québec. Both outcrop types occur on the same bedrock geology types. Unforested outcrops are not recently disturbed sites in early-successional states, as no clear distinction could be made in tree species composition and date of the last fire between the two outcrop types. Forested and unforested outcrops are not alternative stable states, as unforested outcrops are unstable and cannot maintain themselves through time in the prolonged absence of fire. Hence, unforested rock outcrops could be viewed as degraded, diverging post-fire types maintained by the late Holocene disturbance regime, characterized by high fire frequencies.     

Assessment of the impact of resistant and susceptible canola on Plasmodiophora brassicae inoculum potential

The impact on clubroot severity of growing susceptible canola or mixtures of resistant and susceptible canola genotypes was examined. Bioassays revealed greater clubroot severity and incidence, and reduced plant height, where 100% of a susceptible cultivar had been grown. A higher proportion of susceptible plants within a resistant canola crop increased root hair and secondary infections. Regression analysis of root hair infection and the amount of Plasmodiophora brassicae DNA (as determined by quantitative PCR) revealed strong linear relationships between the two parameters. The linear relationships between root hair infection and P. brassicae DNA were stronger for the resistant cultivar than for the susceptible cultivar when regression analysis was conducted by cultivar over the sampling dates. In conclusion, the cropping of a resistant cultivar reduced clubroot severity, while the presence of susceptible volunteer canola increased inoculum potential. Quantitative PCR was a reliable tool for the quantification of root hair infection.