Bayesian space–time model to analyse frost risk for agriculture in Southeast Australia

Abstract

Despite a broad pattern of warming in minimum temperatures over the past 50years, regions of southeastern Australia have experienced increases in frost frequency in recent decades, and more broadly across southern Australia, an extension of the frost window due to an earlier onset and later cessation. Consistent across southern Australia is a later cessation of frosts, with some areas of southeastern Australia experiencing the last frost an average 4weeks later than in the 1960s (i.e. mean date of last frost for the period 1960-1970 was 19 September versus 22 October for the period [2000][2001][2002][2003][2004][2005][2006][2007][2008][2009]. We seek to model the spatial changes in frosts for a region exhibiting the strongest individual station trends, i.e. northern Victoria and southern New South Wales. We identify statistically significant trends at low-lying stations for the month of August and construct and validate a Bayesian space-time model of minimum temperatures, using rates of greenhouse gas (GHG) emissions, as well as other well-understood causal factors including solar radiation, the El Niño Southern Oscillation (ENSO 3.4) and times series data relating to the position (STRP) and intensity (STRI) of subtropical highs and blocking high pressure systems. We assess the performance of this modelling approach against observational records as well as against additive and linear regression modelling approaches using root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) as well as false alarm and hit rate metrics. The spatiotemporal modelling approach demonstrated considerably better predictive skill than the others, with enhanced performance across all the metrics analysed. This enhanced performance was consistent across each decade and for temperature extremes below 2°C. Crimp, S., Bakar, K., Kokic, P., Jin, H., Nicholls, N. & Howden, M. (2015) ABSTRACT: Despite a broad pattern of warming in minimum temperatures over the past 50 years, regions of southeastern Australia have experienced increases in frost frequency in recent decades, and more broadly across southern Australia, an extension of the frost window due to an earlier onset and later cessation. Consistent across southern Australia is a later cessation of frosts, with some areas of southeastern Australia experiencing the last frost an average 4 weeks later than in the 1960s (i.e. mean date of last frost for the period 1960-1970 was 19 September versus 22 October for the period [2000][2001][2002][2003][2004][2005][2006][2007][2008][2009]. We seek to model the spatial changes in frosts for a region exhibiting the strongest individual station trends, i.e. northern Victoria and southern New South Wales. We identify statistically significant trends at low-lying stations for the month of August and construct and validate a Bayesian space-time model of minimum temperatures, using rates of greenhouse gas (GHG) emissions, as well as other well-understood causal factors including solar radiation, the El Niño S...

spTimer: Spatio-Temporal Bayesian Modeling UsingR

“…We observe that the MSE for the Bayesian space-time GP model is reduced by about 56% compared to the generalized additive models, showing superiority of the spatio-temporal models implemented in the package spTimer. Further similar comparison studies have been performed by Crimp, Bakar, Kokic, Jin, Nicholls, and Howden (2015), where spTimer GP model also shows better predictive performance compared to the additive models for analyzing frost levels in south-east Australia.…”

An investigation of some unexpected frost day increases in southern Australia

“…Instead they suggest a growing influence from the tropical dry season associated with a poleward shift of the ocean and atmosphere circulation, particularly after 1980. A recent high resolution modelling study (Delworth and Zeng, 2014) also finds that anthropogenic forcing is the main cause of the rainfall decline observed in southwest Western Australia.…”

“…Crimp et al (2014) analysed changes in frost occurrences in northern Victoria and southern New South Wales over the months August to November for the period 1961 to 2009. They found increases in the number of frost days at low-lying stations over the August to November period and a lengthening of the frost season.…”

Drought forecasting in eastern Australia using multivariate adaptive regression spline, least square support vector machine and M5Tree model

“…Thus, accurate models can assist with the management of risk and promote economic returns (Timbal and Hendon, 2011;Koehn, 2015;Williams et al, 2015;Qureshi et al, 2016). Consequently, drought has been forecasted using many approaches, for example, with hydrological models (Brown et al, 2015), Markov chain (Rahmat et al, 2016), Bayesian space-time models (Crimp et al, 2015) and recently, data-driven models (Abbot and Marohasy, 2012, 2014; Şahin, 2015b, a, 2016;Deo et al, 2016b). Notwithstanding this, other than Rahmat et al (2016) who applied a Markov chain model for SPI modelling, to the best of our …”