Search

A large literature on the sensitivity of streamflow response to climate variations has emerged over the past several decades, but the underlying mechanisms are not fully understood. The usual approaches to this problem have been simple and do not fully address its complexity, which involves the individual and joint effects of large‐scale climate modes and smaller‐scale weather features on streamflow volumes, the presence of persistence in these time series and their effects on antecedent conditions. Ongoing improvements in observational and reanalysis datasets and online access allow for better quantification of the connections between streamflow response, climate variability modes and regional weather features. The purpose of this paper is to determine whether Bayesian networks can be used to better identify key factors and their associated pathways leading to streamflow generation. A Markov blanket approach is described and illustrated using monthly streamflow series recorded at eight gauging stations within or immediately adjacent to the coastal region of eastern mainland Australia. The method is compared and contrasted with conventional multiple linear regression with discussion around this type of approach as part of a growing interest in machine learning applications. One example is used to illustrate the application of both approaches to a streamflow series exhibiting strong non‐stationarity. Results for the other streamflow series are included in a concise synthesis. Overall, the results suggest that Bayesian networks have several desirable features in terms of transparency, interpretability and explanatory insight. The findings from this study lend support to the use of Bayesian networks for modelling connections between streamflow volume and the variability of climate and regional weather. This improved understanding of the key controls of streamflow variability is intended to help address growing needs around informing social, cultural, economic and ecological aspects of water planning and management. [ABSTRACT FROM AUTHOR]

Understanding the relationships between large-scale, low-frequency climate variability modes, fire weather conditions and lighting-ignited wildfires has implications for fire-weather prediction, fire management and conservation. This article proposes a Bayesian network framework for quantifying the influence of climate modes on fire weather conditions and occurrence of lightning-ignited wildfires. The main objectives are to describe and demonstrate a probabilistic framework for identifying and quantifying the joint and individual relationships that comprise the climate-wildfire system; gain insight into potential causal mechanisms and pathways; gauge the influence of climate modes on fire weather and lightning-ignition relative to that of local-scale conditions alone; assess the predictive skill of the network; and motivate the use of techniques that are intuitive, flexible and for which user‐friendly software is freely available. A case study illustrates the application of the framework to a forested region in southwest Australia. Indices for six climate variability modes are considered along with two hazard variables (observed fire weather conditions and prescribed burn area), and a 41-year record of lightning-ignited wildfire counts. Using the case study data set, we demonstrate that the proposed framework: (1) is based on reasonable assumptions provided the joint density of the variables is converted to multivariate normal; (2) generates a parsimonious and interpretable network architecture; (3) identifies known or partially known relationships between the variables; (4) has potential to be used in a predictive setting for fire weather conditions; and (5) climate modes are more directly related to fire weather conditions than to lightning-ignition counts. [ABSTRACT FROM AUTHOR]

Abstract An exploratory analysis of lightning-ignited wildfire data for the Warren Region of Western Australia was carried out for the period from April 1976 to December 2016. Temporal patterns in the series were examined in terms of characterizing the seasonal cycle, and detecting long-term trends and changes in seasonality over time. A generalized additive modelling approach was used to ensure that temporal features were determined by the data rather than a priori assumed mathematical forms (e.g. linear or low-order polynomial functions). The spatial organization of the data was evaluated using concepts from the theory of stochastic point processes. Results indicate a strong seasonality in the monthly lightning ignition series, the presence of a long-term trend and an interaction between trend and seasonality. There is also strong evidence of spatial variation in the number of ignitions per unit area in terms of location and distance from nearest ignition. Within the Warren Region, observation platforms for fire detection and reporting protocols have remained stable over the period of record, and changes in land use are unlikely to have altered the pattern of lightning ignition. Thus, the above results might reflect an interplay between: landscape attributes (e.g. vegetation classes, elevation, slope, aspect); changes in rainfall and fuel moisture; changes in fuel management practices; and, perhaps, an increase in the frequency of dry thunderstorms and fire weather conditions. Highlights • Lightning-ignited wildfires exhibit a changing seasonal cycle and long-term trend. • Sizeable proportion of lightning ignitions originated from strikes on or near trees. • Point pattern of ignitions does not exhibit complete spatial randomness. • Point pattern is consistent with an inhomogeneous Poisson process. [ABSTRACT FROM AUTHOR]

Lightning is a natural hazard that can lead to the ignition of wildfires, disruption and damage to power and telecommunication infrastructures, human and livestock injuries and fatalities, and disruption to airport activities. This paper examines the ability of six statistical and machine-learning classification techniques to distinguish between nonlightning and lightning days at the coarse spatial and temporal scales of current general circulation models and reanalyses. The classification techniques considered were 1) a combination of principal component analysis and logistic regression, 2) classification and regression trees, 3) random forests, 4) linear discriminant analysis, 5) quadratic discriminant analysis, and 6) logistic regression. Lightning-flash counts at six locations across Australia for 2004-13 were used, together with atmospheric variables from the ERA-Interim dataset. Tenfold cross validation was used to evaluate classification performance. It was found that logistic regression was superior to the other classifiers considered and that its prediction skill is much better than using climatological values. The sets of atmospheric variables included in the final logistic-regression models were primarily composed of spatial mean measures of instability and lifting potential, along with atmospheric water content. The memberships of these sets varied among climatic zones. [ABSTRACT FROM AUTHOR]

Lightning accompanied by inconsequential rainfall (i.e., 'dry' lightning) is the primary natural ignition source for wildfires globally. This paper presents a machine-learning and statistical-classification analysis of dry and 'wet' thunderstorm days in relation to associated atmospheric conditions. The study is based on daily data for lightning-flash count and precipitation from ground-based sensors and gauges and a comprehensive set of atmospheric variables that are based on ERA-Interim for the period from 2004 to 2013 at six locations in Australia. These locations represent a wide range of climatic zones (temperate, subtropical, and tropical). Quadratic surface representations and low-dimensional summary statistics were used to characterize the main features of the atmospheric fields. Four prediction skill scores were considered, and 10-fold cross validation was used to evaluate the performance of each classifier. The results were compared with those obtained by adopting the approach used in an earlier study for the U.S. Pacific Northwest. It was found that both approaches have prediction skill when tested against independent data, that mean atmospheric field quantities proved to be the most influential variables in determining dry-lightning activity, and that no single classifier or set of atmospheric variables proved to be consistently superior to its counterpart for the six sites examined here. [ABSTRACT FROM AUTHOR]

Highlights: [•] We propose and demonstrate an index for similarity in the storm event response of catchments to rainfall. [•] The similarity index is based on the steady state assumption. [•] The index is applicable to catchments where runoff is generated by saturation excess overland flow. [•] The index was applied to 236 independent storm events on eight catchments in southeast Australia. [•] The index’s effectiveness was assessed by comparing its groupings with those obtained from a flood frequency analysis. [ABSTRACT FROM AUTHOR]

A hierarchical spatial model for daily rainfall extremes that characterizes their temporal variation due to interannual climatic forcing as well as their spatial pattern is proposed. The model treats the parameters of at-site probability distributions for rainfall extremes as “data” that are likely to be spatially correlated and driven by atmospheric forcing. The method is applied to daily rainfall extremes for summer and winter half years over the Swan–Avon River basin in Western Australia. Two techniques for the characterization of at-site extremes—peaks-over-threshold (POT) analysis and the generalized extreme value (GEV) distribution—and three climatic drivers—the El Niño–Southern Oscillation as measured by the Southern Oscillation index (SOI), the Southern Hemisphere annular mode as measured by an Antarctic Oscillation index (AOI), and solar irradiance (SI)—were considered. The POT analysis of at-site extremes revealed that at-site thresholds lacked spatial coherence, making it difficult to determine a smooth spatial surface for the threshold parameter. In contrast, the GEV-based analysis indicated smooth spatial patterns in daily rainfall extremes that are consistent with the predominant orientation of storm tracks over the study area and the presence of a coastal escarpment near the western edge of the basin. It also indicated a linkage between temporal trends in daily rainfall extremes and those of the SOI and AOI. By applying the spatial models to winter and summer extreme rainfalls separately, an apparent increasing trend in return levels of summer rainfall to the northwest and decreasing trends in return levels of winter rainfall to the southwest of the region are found. [ABSTRACT FROM AUTHOR]

Summary: Digital elevation models (DEMs) are widely used in topographically-based hydrology models. Most researchers dealing with DEMs have worked exclusively with either grid or contour based techniques. Intercomparisons of these techniques are empirical in nature, few in number and limited to a small number of catchments. We examine the relative effects of these techniques on the distribution of the topographic index (TI) for nine idealised hillslope types and eight natural catchments (4.4–83km2) in southeast Australia. In particular we investigate the underlying causes for the frequently observed increase in the TI with grid size. A simple analytical model for a parallel-planar hillslope indicates that median TI is asymptotic with respect to the number of grid cells and a linear function of grid size. We demonstrate that the latter result holds for hillslopes with more complex geometry. Using a multiple flow direction algorithm, we find that hillslope geometry and flow partitioning affect the spatial pattern of the TI and the rate of change (M) of median TI with grid size. This makes it impractical to recommend general guidelines for the value of the directional parameter in flow partitioning algorithms. For the natural catchments we show that median TI is also a linear function of grid size and that there is some evidence for a linear relationship between M and the inverse of the median topographic slope. Throughout the investigation we find that contour DEMs are less sensitive to changes in resolution than grid DEMs. [Copyright &y& Elsevier]

The impact of climate change on hydrology and water resources is one of the most critical issues facing the world in the next few decades. In particular, there is a need to quantify the risks associated with maintaining the security of resource quantity and quality, and to assess the effectiveness of potential management strategies. In this paper, we assess the impacts of climate variability and change on one aspect of river health. A simple model of Anabaena algal bloom occurrence at a weir pool in the lower Murrumbidgee River, Australia, has been coupled to a catchment model that is used to simulate streamflow, irrigation demand and diversions, dam water storage and releases, and decision-making by both irrigators and managers. Long-term climate data are obtained from a statistical downscaling algorithm, which, when applied to global climate model predictions can provide climate data suitable for driving the coupled model under a variety of climatic scenarios. The coupled model is then used to assess the impact of climate variability and projected climate change on the frequency, duration and magnitude of Anabaena blooms. The impact of two management strategies for bloom control are also assessed and it is shown that even a single, quite simple, resource-neutral, adaptive management strategy has the potential to substantially reduce the occurrence and impact of algal blooms and to more than compensate for the deleterious impacts of climate change. This result supports the notion that planning for the future can lead to positive outcomes in the present. [ABSTRACT FROM AUTHOR]