Conrad Wasko

Quantile regression for investigating scaling of extreme precipitation with temperature

The consensus in the scientific community is that the intensity of extreme precipitation will increase in a warmer climate. However, as there is limited observational evidence to this effect, there is a growing body of research which focuses on directly investigating the relationship between temperature and precipitation. This is currently performed by binning precipitation data in temperature bins and then investigating the trend in the precipitation percentiles in each bin with temperature. In this paper, we highlight limitations in the binning approach and present quantile regression as an alternative to the above process. Quantile regression allows estimation of this scaling directly and, unlike binning, is unbiased with sample size. Moreover, quantile regression presents a natural framework for investigation into other factors (covariates) that may be affecting the nature of the scaling relationship. Results using subdaily rainfall data for Australia show the efficacy of the proposed quantile regression method, as well as the presence of season indicators as significant covariates that affect the scaling relationship of precipitation with temperature. A general increase in the scaling coefficient in winter versus summer is observed.

Reduced spatial extent of extreme storms at higher temperatures

Extreme precipitation intensity is expected to increase in proportion to the water-holding capacity of the atmosphere. However, increases beyond this expectation have been observed, implying that changes in storm dynamics may be occurring alongside changes in moisture availability. Such changes imply shifts in the spatial organization of storms, and we test this by analyzing present-day sensitivities between storm spatial organization and near-surface atmospheric temperature. We show that both the total precipitation depth and the peak precipitation intensity increases with temperature, while the storms spatial extent decreases. This suggests that storm cells intensify at warmer temperatures, with a greater total amount of moisture in the storm, as well as a redistribution of moisture toward the storm center. The results have significant implications for the severity of flooding, as precipitation may become both more intense and spatially concentrated in a warming climate.

Does storm duration modulate the extreme precipitation‐temperature scaling relationship?

Predicting future precipitation extremes is difficult, and therefore, many studies have used the historical relationship between precipitation intensity and temperature to consider what might occur in a future warmer climate. In general, extreme precipitation intensity is expected to increase as temperatures increase. However, in tropical areas it has been observed that, for higher temperatures, lower precipitation intensities occur, contradicting the expected relationship. This has been thought to be due to limits in moisture availability. In this work we show that the negative scaling found in previous studies may be a result of the analysis methods. By conditioning the precipitation intensity and temperature relationship on storm duration, we demonstrate that positive scaling of precipitation intensity with temperature in tropical regions of Australia is possible. We argue that methods for estimating scaling relationships should be modified to include storm duration.

Representing low‐frequency variability in continuous rainfall simulations: A hierarchical random Bartlett Lewis continuous rainfall generation model

Low-frequency variability, in the form of the El Nino-Southern Oscillation, plays a key role in shaping local weather systems. However, current continuous stochastic rainfall models do not account for this variability in their simulations. Here a modified Random Pulse Bartlett Lewis stochastic generation model is presented for continuous rainfall simulation exhibiting low-frequency variability. Termed the Hierarchical Random Bartlett Lewis Model (HRBLM), the model features a hierarchical structure to represent a range of rainfall characteristics associated with the El Nino-Southern Oscillation with parameters conditioned to vary as functions of relevant climatic states. Long observational records of near-continuous rainfall at various locations in Australia are used to formulate and evaluate the model. The results indicate clear benefits of using the hierarchical climate-dependent structure proposed. In addition to accurately representing the wet spells characteristics and observed low-frequency variability, the model replicates the interannual variability of the antecedent rainfall preceding the extremes, which is known to be of considerable importance in design flood estimation applications.

Global assessment of flood and storm extremes with increased temperatures

There is overwhelming consensus that the intensity of heavy precipitation events is increasing in a warming world. It is generally expected such increases will translate to a corresponding increase in flooding. Here, using global data sets for non-urban catchments, we investigate the sensitivity of extreme daily precipitation and streamflow to changes in daily temperature. We find little evidence to suggest that increases in heavy rainfall events at higher temperatures result in similar increases in streamflow, with most regions throughout the world showing decreased streamflow with higher temperatures. To understand why this is the case, we assess the impact of the size of the catchment and the rarity of the event. As the precipitation event becomes more extreme and the catchment size becomes smaller, characteristics such as the initial moisture in the catchment become less relevant, leading to a more consistent response of precipitation and streamflow extremes to temperature increase. Our results indicate that only in the most extreme cases, for smaller catchments, do increases in precipitation at higher temperatures correspond to increases in streamflow.

A quasi‐global assessment of changes in remotely sensed rainfall extremes with temperature

The dependence between extreme rainfall and temperature is used to understand climatic relationships, constrain model predictions and evaluate future changes to rainfall. Understanding this dependence, however, is limited by the fact that many areas worldwide lack gauged data, particularly at short time scales. The advent of remote sensing allows a new insight into this dependence quasi-globally. Here, we address whether remotely sensed daily rainfall and temperature can be used in ungauged areas to understand extreme rainfall scaling with temperature. Using the multi-sensor Tropical Rainfall Measuring Mission 3B42 (v7) rainfall product and remotely sensed air temperature we examine the spatial homogeneity in remotely sensed rainfall scaling with temperature and demonstrate that it replicates the spatial variation in the scaling observed in ground data. Finally, changes to duration and percentile are examined showing that the scaling response is climatologically sensitive.

A comprehensive urban floodplain dataset for model benchmarking

ABSTRACT Floodplain managers require accurate and reliable information quantifying flood flow behaviour to support effective land-use planning and flood emergency planning. Two-dimensional numerical models, typically solving the shallow water approximation of the Navier–Stokes equations, have become a de-facto standard for predicting design flow behaviour. In urban floodplains, the built environment can have a profound influence on the passage and distribution of floodwaters. Obstacles such as buildings, fences and walls can block and redistribute flows overriding the gradient of the topography and locally increasing the flood hazard. The development and application of numerical models for urban floodplains is open to considerable interpretation and numerous modelling techniques have been proposed to represent the buildings and other obstacles to flow. Here, a comprehensive dataset for an urban overland flow path is developed to help practitioners assess numerical model performance. A physical model of the Morgan–Selwyn floodway in Merewether, Newcastle, Australia was developed and validated against the historical extreme June 2007 ‘Pasha Bulker’ storm. Detailed measurements of the flow behaviour were then collected. The comprehensive dataset of the physical model topography, flood flow boundary conditions as well as detailed measurements of flow depth and velocity are freely available to practitioners who wish to further investigate the dataset or apply the dataset in their numerical modelling.