No abstract

Although originally designed to measure surface denudation, the micro-erosion meter (MEM) can be adapted easily to measure deposition rates of chemical sediments such as travertines and speleothems. At Louie Creek, northwest Queensland, Australia, travertine deposition rates measured using the MEM average 4·15 mm a −1 . However, this figure masks considerable rate variability. Both purely hydraulic and hydraulically related variables appear to be the major mechanisms controlling deposition rates. The most rapid rates occur within relatively high-energy hydraulic regimes (impact and flow zones), whilst deposition rates in pools separating individual travertine barrages (standing water zones) are relatively slow. Deposition rate variations within spray and impact zones are related directly to discharge. The highest rates in flow zones correlate with the incorporation into the travertine of in situ and allochthonous biogenic material, such as caddis fly larvae, green algal mats and phytoclasts, which proliferate or are entrapped easily under such hydraulic conditions. Considerable spatial variability in deposition rates also prevails. The highest rates for a given set of hydraulic conditions occur at two sites, the Upper Everglades and the Lower Everglades. The MEM also measures net erosion of travertines. At Louie Creek, Most of the travertine erosion occurs in the wet season and is confined primarily to standing water zones. changes in palaeoenvironmental parameters such as regional climate, discharge, sediment loads and stream chemistry. Finally, the detailed examination of deposition rates may expose significant spatial and temporal variations which equip researchers with a better understanding of the dynamics of travertine-depositing systems.Compared with research on physico-chemical and biological aspects of travertine deposition, there is relatively little published work which deals specifically with rates of travertine deposition. Viles and Goudie (1990) summarized a selection of travertine deposition rate data and noted rates ranging from <1mm a −1 to approximately 0·5 m a −1 . The highest rates appear to occur in hot-spring (or thermogene) travertines (sensu Pentecost and Viles, 1994), such as those from geothermal regions in Italy and the USA (Allen, 1934;Folk et al., 1985;Pentecost, 1994;Folk, 1994); Kitano (1963, cited in Folk et al., 1985, for example, noted rates of up to 1m a −1 from hot-spring deposits in Japan. Rapid deposition of cool water (or meteogene) travertines (sensu Pentecost and Viles, 1994) seems to be more the exception than the rule (e.g. Pentecost, 1978Pentecost, , 1981, although Emig (1917) noted rates of up to 1mm per month during summer months in an Oklahoma stream.Most published estimates of deposition rates have been either inferred using mass balance models (e.g. Jacobson and Usdowski, 1975;Kempe and Emeis, 1985;Lorah and Herman, 1990) or derived from one of several possible direct methods of measurement. Mass balance models are input/output-based and use chemical and hydrologic...

Fire is a significant determinant of vegetation structure in Australia’s savannas and has been implicated in the decline of many species. Identifying the patterns of fire in the landscape is fundamental to understanding vegetation dynamics but variation over time and space makes generalization difficult and specific management recommendations elusive. In order to improve the knowledge base for fire management in tropical savannas, we investigated interregional variation in fire patterns in two Queensland bioregions, the Mount Isa Inlier (MII) and Cape York Peninsula (CYP), over a 5‐year period (1999–2003). Remotely sensed satellite data were used to identify burnt areas on a monthly basis for the western half of the CYP bioregion and about two‐thirds of the smaller MII. Fire scars were mapped from JPEG‐compressed, low‐resolution Landsat images using geographical information system technology and data were investigated to determine annual burning patterns. Patterns were interpreted with regard to meteorological information and recent fire history. The area burnt per annum on western CYP was generally an order of magnitude greater than the area burnt on the MII. In the biggest fire year, nearly 74% (5 295 098 ha) of the CYP landscape burnt, compared with 35% (1 770 771 ha) of the MII landscape. The minimum percentage of the CYP study area burnt in 1 year between 1999 and 2003 was 43.1%, compared with 1.6% for the MII. The reliability and amount of seasonal rainfall was a strong determinant of differences in time of fire occurrence and area burnt between regions. Widespread wildfires were significantly related to above average rainfall in the preceding 12 months in the Mt. Isa area but not in CYP. Rainfall also affected fire frequency. Predictable wet season rainfall on CYP allowed for a biennial fire return interval, while on the semiarid MII, the average fire return interval was 5 years or longer. We conclude that the fire patterns in the semiarid MII are similar to those reported for arid Australia, while fire patterns in western CYP are comparable with other mesic savanna areas.

Our ability to model global carbon fluxes depends on understanding how terrestrial carbon stocks respond to varying environmental conditions. Tropical forests contain the bulk of the biosphere's carbon. However, there is a lack of consensus as to how gradients in environmental conditions affect tropical forest carbon. Papua New Guinea (PNG) lies within one of the largest areas of contiguous tropical forest and is characterized by environmental gradients driven by altitude; yet, the region has been grossly understudied. Here, we present the first field assessment of aboveground biomass (AGB) across three main forest types of PNG using 193 plots stratified across 3,100-m elevation gradient. Unexpectedly, AGB had no direct relationship to rainfall, temperature, soil, or topography. Instead, natural disturbances explained most variation in AGB. While large trees (diameter at breast height > 50 cm) drove altitudinal patterns of AGB, resulting in a major peak in AGB (2,200-3,100 m) and some of the most carbon-rich forests at these altitudes anywhere. Large trees were correlated to a set of climatic variables following a hump-shaped curve. The set of "optimal" climatic conditions found in montane cloud forests is similar to that of maritime temperate areas that harbor the largest trees in the world: high ratio of precipitation to evapotranspiration (2.8), moderate mean annual temperature (13.7°C), and low intra-annual temperature range (7.5°C). At extreme altitudes (2,800-3,100 m), where tree diversity elsewhere is usually low and large trees are generally rare or absent, specimens from 18 families had girths>70 cm diameter and maximum heights 20-41 m. These findings indicate that simple AGB-climate-edaphic models may not be suitable for estimating carbon storage in forests where optimal climate niches exist. Our study, conducted in a very remote area, suggests that tropical montane forests may contain greater AGB than previously thought and the importance of securing their future under a changing climate is therefore enhanced.

Relationships with climate and local resources are developed for soils, vegetation and tree foliage as well as levels of herbivory for the dominant eucalypts at sites representing a regional gradient in climate and local contrasts in landscape position. Indicators of site productivity such as soil nitrogen and phosphorus, canopy height and cover, foliar nitrogen and water, and average leaf area tended to increase as climate became more favourable. Many were also higher in locally richer parts of the landscape. In contrast, specific leaf weight, an indication of sclerophylly, decreased as climate and local resources became more favourable. Rates of herbivory tended to increase with increasing site productivity and the associated changes in soil, vegetation and foliar properties, in broad agreement with models relating herbivory to resource availability and plant vigour. We found no evidence to support models relating high herbivory to low-resource environments and plant stress. The apparent level of herbivore damage on mature leaves was highest at intermediate levels of resources; this could reflect interactions between resource availability, rates of herbivory and rates of leaf replacement. Implications of these findings are discussed with respect to ways of measuring herbivory, regional patterns in rates and levels of herbivory, and the regional distribution of rural dieback associated with high herbivory.