David E. Robertson

Evidence for Using Lagged Climate Indices to Forecast Australian Seasonal Rainfall

AbstractLagged oceanic and atmospheric climate indices are potentially useful predictors of seasonal rainfall totals. A rigorous Bayesian joint probability modeling approach is applied to find the cross-validation predictive densities of gridded Australian seasonal rainfall totals using lagged climate indices as predictors over the period of 1950–2009. The evidence supporting the use of each climate index as a predictor of seasonal rainfall is quantified by the pseudo-Bayes factor based on cross-validation predictive densities. The evidence strongly supports the use of climate indices from the Pacific region with weaker, but positive, evidence for the use of climate indices from the Indian region and the extratropical region. The spatial structure and seasonal variation of the evidence for each climate index is mapped and compared. Spatially, the strongest supporting evidence is found for forecasting in northern and eastern Australia. Seasonally, the strongest evidence is found from August–October to Nove...

Merging Seasonal Rainfall Forecasts from Multiple Statistical Models through Bayesian Model Averaging

AbstractMerging forecasts from multiple models has the potential to combine the strengths of individual models and to better represent forecast uncertainty than the use of a single model. This study develops a Bayesian model averaging (BMA) method for merging forecasts from multiple models, giving greater weights to better performing models. The study aims for a BMA method that is capable of producing relatively stable weights in the presence of significant sampling variability, leading to robust forecasts for future events. The BMA method is applied to merge forecasts from multiple statistical models for seasonal rainfall forecasts over Australia using climate indices as predictors. It is shown that the fully merged forecasts effectively combine the best skills of the models to maximize the spatial coverage of positive skill. Overall, the skill is low for the first half of the year but more positive for the second half of the year. Models in the Pacific group contribute the most skill, and models in the ...

Seasonal Forecasts of Australian Rainfall through Calibration and Bridging of Coupled GCM Outputs

AbstractCoupled general circulation models (GCMs) are increasingly being used to forecast seasonal rainfall, but forecast skill is still low for many regions. GCM forecasts suffer from systematic biases, and forecast probabilities derived from ensemble members are often statistically unreliable. Hence, it is necessary to postprocess GCM forecasts to improve skill and statistical reliability. In this study, the authors compare three methods of statistically postprocessing GCM output—calibration, bridging, and a combination of calibration and bridging—as ways to treat these problems and make use of multiple GCM outputs to increase the skill of Australian seasonal rainfall forecasts. Three calibration models are established using ensemble mean rainfall from three variants of the Predictive Ocean Atmosphere Model for Australia (POAMA) version M2.4 as predictors. Six bridging models are established using POAMA forecasts of seasonal climate indices as predictors. The calibration and bridging forecasts are merge...

Bayesian networks for decision analyses — an application to irrigation system selection

Farmers are under continual pressure from Government and industry to change farm practices to meet productivity and environmental targets. In response to these pressures, farmers will make decisions to adopt practices that reflect their motivations and priorities. However, where the changes of practice are major, there may be considerable uncertainty associated with the decision-making process. Decision support tools are one method that may assist in reducing the uncertainty associated with decisions about changes in farm practices. Bayesian networks provide a useful tool to assist in the structuring and analysis of decision problems. A Bayesian network is a decision analysis framework, based on Bayesian probability theory, which allows the integration of scientific and experiential knowledge, and the uncertainty associated with this knowledge. The approach involves describing a system in terms of variables and linkages, or relationships between variables, at a level appropriate to the decision making. This is achieved through representing linkages as conditional probability tables and propagating probabilities through the network to give the likelihood of variable outcomes. Therefore, the approach ensures that treatment of risks and uncertainties is an intrinsic part of the decision-making processes. The Bayesian network is dynamic and interactive, and hence if a network previously developed does not fit a users conceptual understanding of the system, it can be adapted quickly and simply to the cognitive understanding of the user. A case study Bayesian network has been developed for decisions associated with the selection of irrigation systems for irrigated dairy farms in Northern Victoria. This case study demonstrates that the most appropriate irrigation system for a dairy farm is dependent on factors including the amount of irrigation water available and soil types. Analysis of the Bayesian network indicates that the appropriate irrigation system is more sensitive to the income generated from pasture than to the price of water. The Bayesian network can demonstrate the impacts of decisions on the farmers system and can allow the farmer to evaluate these impacts according to their own priorities and criteria. This information can then be used by the natural resource manager to assess the appropriate level of incentive or penalty required if the farmer is to adopt the preferred option that will also achieve preferable outcomes from a natural resource management persepective.

Reliable long‐range ensemble streamflow forecasts: Combining calibrated climate forecasts with a conceptual runoff model and a staged error model

We present a new streamflow forecasting system called forecast guided stochastic scenarios (FoGSS). FoGSS makes use of ensemble seasonal precipitation forecasts from a coupled ocean-atmosphere general circulation model (CGCM). The CGCM forecasts are post-processed with the method of calibration, bridging and merging (CBaM) to produce ensemble precipitation forecasts over river catchments. CBaM corrects biases and removes noise from the CGCM forecasts, and produces highly reliable ensemble precipitation forecasts. The post-processed CGCM forecasts are used to force the Wapaba monthly rainfall-runoff model. Uncertainty in the hydrological modelling is accounted for with a 3-stage error model. Stage 1 applies the log-sinh transformation to normalize residuals and homogenize their variance; Stage 2 applies a conditional bias-correction to correct biases and help remove negative forecast skill; Stage 3 applies an autoregressive model to improve forecast accuracy at short lead-times and propagate uncertainty through the forecast. FoGSS generates ensemble forecasts in the form of time series for the coming 12-months. In a case study of two catchments, FoGSS produces reliable forecasts at all lead-times. Forecast skill with respect to climatology is evident to lead-times of about 3 months. At longer lead-times, forecast skill approximates that of climatology forecasts; that is, forecasts become like stochastic scenarios. Because forecast skill is virtually never negative at long lead-times, forecasts of accumulated volumes can be skillful. Forecasts of accumulated 12-month streamflow volumes are significantly skillful in several instances, and ensembles of accumulated volumes are reliable. We conclude that FoGSS forecasts could be highly useful to water managers. This article is protected by copyright. All rights reserved.

Improving Precipitation Forecasts by Generating Ensembles through Postprocessing

AbstractThis paper evaluates a postprocessing method for deterministic quantitative precipitation forecasts (raw QPFs) from a numerical weather prediction model. The postprocessing aims to produce calibrated QPF ensembles that are bias free, more accurate than raw QPFs, and reliable for use in streamflow forecasting applications. The method combines a simplified version of the Bayesian joint probability (BJP) modeling approach and the Schaake shuffle. The BJP modeling approach relates raw QPFs and observed precipitation by modeling their joint distribution. It corrects biases in the raw QPFs and generates ensemble forecasts that reflect the uncertainty in the raw QPFs. The BJP modeling approach is applied to each lead time and each forecast location separately. The Schaake shuffle is then employed to produce calibrated QPFs with appropriate space–time correlations by linking ensemble members generated by the BJP modeling approach.Calibrated QPFs are produced for 10 Australian catchments that cover a wide ...

How Suitable is Quantile Mapping For Postprocessing GCM Precipitation Forecasts

AbstractGCMs are used by many national weather services to produce seasonal outlooks of atmospheric and oceanic conditions and fluxes. Postprocessing is often a necessary step before GCM forecasts can be applied in practice. Quantile mapping (QM) is rapidly becoming the method of choice by operational agencies to postprocess raw GCM outputs. The authors investigate whether QM is appropriate for this task. Ensemble forecast postprocessing methods should aim to 1) correct bias, 2) ensure forecasts are reliable in ensemble spread, and 3) guarantee forecasts are at least as skillful as climatology, a property called “coherence.” This study evaluates the effectiveness of QM in achieving these aims by applying it to precipitation forecasts from the POAMA model. It is shown that while QM is highly effective in correcting bias, it cannot ensure reliability in forecast ensemble spread or guarantee coherence. This is because QM ignores the correlation between raw ensemble forecasts and observations. When raw foreca...

Estimating hydraulic parameters for a surface irrigation model from field conditions

Border-check irrigation is the most common method of irrigating pastures in Northern Victoria. To make the best use of a border-check irrigation system, consideration needs to be given to the irrigation schedule and irrigation event management. Surface irrigation models can provide an inexpensive and rapid method for identifying optimal irrigation event performance. The most common difficulty encountered when using surface irrigation models is determining appropriate hydraulic parameters. Two experiments were conducted to investigate the relationship between hydraulic parameters of the Analytical Irrigation Model and easily observable field conditions. The field experiments were performed at Tatura, Victoria, on 12 irrigation bays characterised by a Lemnos loam, a red duplex soil, sown to perennial pasture. For each experiment, 3 replicates of 4 treatments were applied. The first experiment found a linear relationship between field soil water deficit, approximated by crop water use less effective rainfall, and the initial infiltration depth. The second experiment found no relationship between pasture height and the model surface roughness parameter. An alternative to estimate the surface roughness parameter is suggested, which involves making an early observation of irrigation advance and solving for the unknown roughness parameter. The parameter estimation method developed in this paper can assist in improving the management of border-check irrigation on Lemnos loam soil, which covers about 125 000 hectares in the Goulburn Valley. However, field-testing of the approach on commercial farms and other soil types is required.

Calibrating hourly rainfall-runoff models with daily forcings for streamflow forecasting applications in meso-scale catchments

The absence of long sub-daily rainfall records can hamper development of continuous streamflow forecasting systems run at sub-daily time steps. We test the hypothesis that simple disaggregation of daily rainfall data to hourly data, combined with hourly streamflow data, can be used to establish efficient hourly rainfall-runoff models. The approach is tested on four rainfall-runoff models and a range of meso-scale catchments (150-3500?km2). We also compare our disaggregation approach to a method of parameter scaling that attains an hourly parameter-set from daily data.Simple disaggregation of daily rainfall produces hourly streamflow models that perform almost as well as those developed from hourly rainfall data. Rainfall disaggregation performs at least as well as parameter scaling, and often better. For the catchments and models we test, simple disaggregation is a very straightforward and effective way to establish hydrological models for continuous sub-daily streamflow forecasting systems when sub-daily rainfall data are unavailable. Daily rainfall is disaggregated to hourly to calibrate hourly hydrological models.Models perform almost as well as models calibrated with observed hourly rainfall.Disaggregation performed at least as well as parameter scaling.A way to develop hourly river forecast systems with daily rainfall.

Seasonal Forecasts of Unregulated Inflows into the Murray River, Australia

Water users along the Murray River, Australia, have traditionally used climatology forecasts of river flows for intra-annual planning of water use and trading. In this paper, we develop and assess the performance of statistical models for forecasting three-month inflow totals for the Murray River. Predictors are selected to represent the influence of initial catchment conditions and future climate on streamflows. These predictors vary with season and location, but are dominated by antecedent streamflows and indices describing the El Nino–Southern Oscillation. For all seasons, the forecasts are skilful with respect to climatology forecasts, and the forecast probability distributions appear to be reliable. Forecast skill is highest for forecasts made between September and December. The forecasts appear to be robust with respect to event size and time, except for the austral autumn seasons for which none of the predictors can forecast the decline in seasonal rainfall over the most recent decade.