火灾预测的模糊马尔柯夫模型

火灾发生并不是一个严格马尔柯夫随机过程，而是一个“近似具有马尔柯夫性”的模糊马尔柯夫过程。本文在对我国55年火灾统计资料“相对化”处理的基础上，根据模糊马尔柯夫理论和方法，建立我国火灾相对变动模糊马尔可夫预测模型。根据隶属度最大原则，确定所属状态，进行火灾预测。2005年我国火灾预测值与实际值的相对误差为0．0148，表明模型具有相当高的预测精度。对我国2006年进行预测，得到火灾相对变动状态等级为微降，我国2006年的火灾发生率与2005年相比降低了0～0．1，火灾频数约为212347～235941起，预测结果可为有关部门的决策提供一定的依据。     

Livestock excreta management through vermicomposting using an epigeic earthworm Eisenia foetida

In India, millions of tones of livestock excreta are produced. Our study explores the potential of an epigeic earthworm Eisenia foetida to compost different livestock excreta (cow, buffalo, horse, donkey, sheep, goat and camel) into value added product (vermicompost) at the laboratory scale. Vermicomposting resulted in lowering of pH, electrical conductivity, potassium and C:N ratio and increase in nitrogen and phosphorus contents. Total K was lower in the final cast than in the initial feed. C:N ratios of the vermicomposts ranged from 16.2 ± 2.17 to 75.4 ± 6.84. Microbial activity measured as dehydrogenase activity in buffalo, donkey and camel wastes increased with time up to day 90. But in sheep and goat wastes, maximum dehydrogenase activity was recorded on day 60 and decreased thereafter. The cocoons and hatchlings production by Eisenia foetida in different excreta were also investigated. The greatest number and biomass of hatchlings was recorded in horse excreta followed by cow, goat and sheep excreta. Thus, cow, horse, sheep and goat excreta show potential as good substrates in vermicomposting using Eisenia foetida, although further research is required to explore the feasibility of use of buffalo, donkey and camel excreta in combination with cow/sheep/goat excreta.     

REAL-TIME SOIL WATER MONITORING FOR OPTIMUM WATER MANAGEMENT1

Abstract: Efficient water resource management is one of the most important policy issues facing agriculture in Hawaii in the years ahead. Soil water sensors, multisensor capacitance probes (MCP), have been successfully used for different water management projects. These MCPs monitor water content at multiple depths and at various locations in real-time. The objectives of this study were to determine the effect of water content on field soil bulk density of Wahiawa silty clay tropical soil; measure field saturated hydraulic conductivity of this tropical soil: calibrate MCP system for this soil: and monitor and evaluate real-time soil water content variations under a tomato crop using the calibrated MCP system. Sensor calibration was conducted under laboratory conditions. Soil bulk density at different water contents and saturated hydraulic conductivity were measured on the field. Bulk density increased with increasing water content: there was a 30 percent bulk density increase as a result of 0.25 cm³ cm^-3 water content variation. Compared with the manufacturer's calibration, site specific laboratory calibration of MCP gave a more accurate determination of soil water. Field determined saturated hydraulic conductivity was higher than laboratory determined values reported in the literature for the same soil type. Real-time soil water content monitoring within the root zone showed substantial variations due to water input (irrigation and rainfall) and water output (evapotranspiration and deep percolations). However, water content variations were much further reduced in the soil layer below the root zone.     

DOCUMENTATION OF HIGH SUMMER FLOWS ON THE POTOMAC RIVER FROM THE WOOD ANATOMY OF ASH TREES1

ABSTRACT: Ash trees (Fraxinus americana L. and F. Pennsylvanica Marsh.) collected from the flood plain of the Potomac River near Washington, D.C., were studied for evidence of associations between known periods of above-average summer flows and changes in wood-growth anatomy. Concentric bands of latewood fibers with atypically large lumens and thin walls commonly developed in trees growing near the low-water channel. Discharge records indicate that roots of most trees with these “white rings” were flooded temporarily during the latewood-growth interval. Trees apparently were not damaged and a concomitant reduction of internal water stresses seems to have accelerated the rate of radial growth. The intra-ring position of anomalous fibers generally corresponded to the time of increased discharge within the estimated interval of latewood growth. Anomalous fibers occasionally formed in unflooded trees, but their position also coincided with episodes of increased discharge. The results of these studies may have applications for streamflow-reconstruction techniques where hydrologic data are incomplete or lacking.     

A model of radiocesium cycling in a sand hills—Turkey Oak (Quercus laevis) ecosystem

Digital simulation models of radiocesium cycling in Turkey Oaks were developed from in situ ¹³⁴Cs tagging studies. Predictions of ¹³⁴Cs steady-state distribution for 3-, 4- and 5-compartment, donor-controlled models were compared with the estimated fallout ¹³⁷Cs distribution as a measure of model validation; output from the 5-compartment model compared best. Sensitivity analysis demonstrated that Turkey Oak burden of ¹³⁴Cs was equally sensitive to the output rate from the tree compartment and the availability of ¹³⁴Cs for uptake (i.e., presence in the root zone) but not the rate of uptake by Turkey Oaks. Observed distribution and model predictions indicate that radiocesium is readily bioaccumulated by Turkey Oaks (～13% of the ecosystem burden) from the soil and is cycled within the sand hills—Turkey Oak ecosystem.     

Environmental Vulnerability Indicators for Environmental Planning and Decision-Making: Guidelines and Applications

Environmental decision-making and policy-making at all levels refers necessarily to synthetic, approximate quantification of environmental properties such as vulnerability, conservation status, and ability to recover after perturbation. Knowledge of such properties is essential to informed decision-making, but their definition is controversial and their precise characterization requires investments in research, modeling, and data collection that are only possible in the most developed countries. Environmental agencies and governments worldwide have increasingly requested numerical quantification or semiquantitative ranking of such attributes at the ecosystem, landscape, and country level. We do not have a theory to guide their calculation, in general or specific contexts, particularly with the amount of resources usually available in such cases. As a result, these measures are often calculated with little scientific justification and high subjectivity, and such doubtful approximations are used for critical decision-making. This problem applies particularly to countries with weak economies, such as small island states, where the most precious environmental resources are often concentrated. This paper discusses frameworks for a “least disappointing,” approximate quantification of environmental vulnerability. After a review of recent research and recent attempts to quantify environmental vulnerability, we discuss models and theoretical frameworks for obtaining an approximate, standardizable vulnerability indicator of minimal subjectivity and maximum generality. We also discuss issues of empirical testing and comparability between indicators developed for different environments. To assess the state of the art, we describe an independent ongoing project developed in the South Pacific area and aimed to the comparative evaluation of the vulnerability of arbitrary countries.     

Foraging areas of king penguins from Macquarie Island in relation to a marine protected area

Twenty-three king penguins (Aptenodytes patagonicus) from Macquarie Island were tracked by satellite during the late incubation period in 1998–1999 to determine the overlap of the foraging zone of king penguins with an area to be declared a marine protected area (MPA) near the island. While all penguins left the colony in an easterly direction and traveled clockwise back to the island, three penguins foraged in the northern parts of the general foraging area and stayed north of 56°S. The remaining 20 penguins ventured south and most crossed 59°S before returning to the island. The total foraging area was estimated to be 156,000 km² with 36,500 km² being most important (where penguins spent >150 hr in total). North-foraging penguins reached on average 331 ± 24 km from the colony compared to 530 ± 76 km for the south-foraging penguins. The latter traveled an average total distance of 1313 ± 176 km, while the northern foragers averaged 963 ± 166 km. Not only did the penguins spend the majority of their foraging time within the boundaries of the proposed MPA, they also foraged chiefly within the boundaries of a highly protected zone. Thus, the MPA is likely to encompass the foraging zone of king penguins, at least during incubation.     

A GIS-based approach for assessing the regional conservation status of genetic diversity: an example from the southern Appalachians

Conserving genetic diversity requires an assessment of the distribution of genetic variants in relation to patterns of land use and environmental variation at a regional scale. This assessment requires a novel approach to integrating and analyzing the genetic and environmental data across spatial scales. To explore the integration of genetic data with other geospatial data sets, we developed a GIS-based approach for examining patterns of genetic diversity for several species of salamanders in southern Appalachians. The genetic data, from allozyme surveys in the genetics literature, were integrated into a GIS database along with related attributes including population identifications and spatial locations. Using existing geospatial data, we classified sample locations as being either protected from anthropogenic disturbance (e.g., National Parks, Wilderness Areas) or as unprotected (e.g., private lands, multiple-use lands in National Forests). We used multidimensional scaling of allelic frequencies and contributions of populations to interpopulation differences in allelic richness to determine which populations had genetic characteristics most different from other populations in the sample. Measures of genetic differentiation were integrated into the GIS database to facilitate spatial analysis and visualization of the indices in relation to land use. This approach was useful for both identification of populations with components of genetic variation that were not well represented at protected sites and for identifying areas of species distributions where more genetic sampling would be necessary to make informed management decisions. Our approach could be readily adapted for use by managers and geneticists working with other species and types of genetic markers.     

Stream restoration and enhancement projects: is anyone monitoring?

Declines in salmon stocks and general watershed health in Washington State, USA, have led to an increase in stream restoration and enhancement projects initiated throughout the state. The increasing number of projects has also raised questions regarding the monitoring of these efforts. Project managers receiving hydraulic project approvals (HPAs) were surveyed to determine whether monitoring was taking place on their projects. About half the project managers surveyed reported the collection of baseline data and the use of biological, physical, chemical, or other water quality measures for their projects. Of those who reported collection of monitoring data, only 18% indicated that monitoring was required. Respondents were also asked to rank the importance of various project goals on a Likert scale. Project managers with projects focusing on “engineering” goals (e.g., roadbed stabilization) were less likely than other project managers to collect baseline monitoring data. Project managers with projects focusing on “restoration/ecological” or “fisheries” goals were more likely than other project managers to collect monitoring measures. Although monitoring appears to be taking place in slightly more than half of the projects surveyed, the nature of the data collected varies widely across projects, and in most cases the monitoring effort is voluntary. This suggests that project sponsors, funders, and managers must consider the issues involved in requiring appropriate monitoring, establishing standardized monitoring guidelines, the time frames in which to monitor, providing other incentives for conducting monitoring, and ensuring adequate funding for monitoring efforts.     

Maintaining Volunteer Commitment to Local Watershed Initiatives

Australia's Landcare program is advanced as a successful international example of local watershed groups and governments working together to improve natural resource management. One of the aspects considered critical in the success of watershed groups is engaging widespread participation. This paper draws on two regional surveys that explored burnout, or loss of engagement, among Landcare participants in the state of Victoria using the Maslach Burnout Inventory. Survey findings indicated that a large proportion of respondents were experiencing high burnout in terms of low personal accomplishment and suggested that there was potential for burnout to increase. The authors suggest that the expectations of watershed groups must be based around a realistic assessment of the capacity for volunteer groups to deliver improved environmental and social outcomes.