Time Series Of Daily Maximum Temperatures Research Articles

Statistical properties of the entropy from ordinal patterns.

The ultimate purpose of the statistical analysis of ordinal patterns is to characterize the distribution of the features they induce. In particular, knowing the joint distribution of the pair entropy-statistical complexity for a large class of time series models would allow statistical tests that are unavailable to date. Working in this direction, we characterize the asymptotic distribution of the empirical Shannon's entropy for any model under which the true normalized entropy is neither zero nor one. We obtain the asymptotic distribution from the central limit theorem (assuming large time series), the multivariate delta method, and a third-order correction of its mean value. We discuss the applicability of other results (exact, first-, and second-order corrections) regarding their accuracy and numerical stability. Within a general framework for building test statistics about Shannon's entropy, we present a bilateral test that verifies if there is enough evidence to reject the hypothesis that two signals produce ordinal patterns with the same Shannon's entropy. We applied this bilateral test to the daily maximum temperature time series from three cities (Dublin, Edinburgh, and Miami) and obtained sensible results.

Monitoring observed changes in warm-days extremes over Turkey

This study aimed to analyze warm-days changes extracted from daily maximum temperature time-series of 71 stations in Turkey during 1961–2016. First, the trend analysis of warm-days events indicated that the annual count of warm-days occurrences has been significance rising by a rate of 1.4 days per decade over more than 90 percent of the studied stations. Thus, there are consistent patterns in daily warming throughout the study area. The spatial maps showed that the lowest frequency in the minimum annual number of warm-days occurred in western areas by a total number of 21 days and its highest occurred in the eastern area by a total number of 35 days. Moreover, the highest frequency in the mean and minimum annual count of warm-days observed in the northwestern lowlands and their frequency has decreased toward the eastern mountainous. Also, the highest range of differences between the annual count of maximum and minimum values has happened in the eastern regions which are characterized by high elevation and the lower existence of atmospheric humidity. Therefore, a higher moisture content of the atmosphere in lowland coastal stations favors severe warm-days and the increased risk of human health.

Investigating dynamical features in the long-term daily maximum temperature time series recorded at Adrián Jara, Paraguay

In this study, we analyse the daily series of maximum temperature measured at Adrian Jara, Paraguay, to disclose features possibly linked with climatological indices that are thought to regulate, in general, the climatic variability of South America. This station is well suited to perform this investigation because it is located in a rural not urbanized area. Our findings reveal that the analyzed series is modulated by several cycles: the annual and half-annual cycle; the 8-year cycle that could be driven by 8-year cycle of the baroclinic waves at the Southern Hemisphere; the 2–3-year periodicity, which seems linked with the Pacific Decadal Oscillation; the 15–16 years periodicity that seems to be induced by the low-frequency fluctuations of SST in the North Atlantic. It is also found an apparent shift in the maximum temperature in the late 1989 that seems to be in agreement with similar findings by other studies that documented such potential jump in the climate system.

Climate change: a new metric to measure changes in the frequency of extreme temperatures using record data

Consensus on global warming is the result of multiple and varying lines of evidence, and one key ramification is the increase in frequency of extreme climate events including record high temperatures. Here we develop a metric—called “record equivalent draws” (RED)—based on record high (low) temperature observations, and show that changes in RED approximate changes in the likelihood of extreme high (low) temperatures. Since we also show that this metric is independent of the specifics of the underlying temperature distributions, RED estimates can be aggregated across different climates to provide a genuinely global assessment of climate change. Using data on monthly average temperatures across the global landmass we find that the frequency of extreme high temperatures increased 10-fold between the first three decades of the last century (1900–1929) and the most recent decade (1999–2008). A more disaggregated analysis shows that the increase in frequency of extreme high temperatures is greater in the tropics than in higher latitudes, a pattern that is not indicated by changes in mean temperature. Our RED estimates also suggest concurrent increases in the frequency of both extreme high and extreme low temperatures during 2002–2008, a period when we observe a plateauing of global mean temperature. Using daily extreme temperature observations, we find that the frequency of extreme high temperatures is greater in the daily minimum temperature time-series compared to the daily maximum temperature time-series. There is no such observable difference in the frequency of extreme low temperatures between the daily minimum and daily maximum.

Summer heat waves over western Europe 1880–2003, their relationship to large-scale forcings and predictability

We investigate the large-scale forcing and teleconnections between atmospheric circulation (sea level pressure, SLP), sea surface temperatures (SSTs), precipitation and heat wave events over western Europe using a new dataset of 54 daily maximum temperature time series. Forty four of these time series have been homogenised at the daily timescale to ensure that the presence of inhomogeneities has been minimised. The daily data have been used to create a seasonal index of the number of heat waves. Using canonical correlation analysis (CCA), heat waves over western Europe are shown to be related to anomalous high pressure over Scandinavia and central western Europe. Other forcing factors such as Atlantic SSTs and European precipitation, the later as a proxy for soil moisture, a known factor in strengthening land–atmosphere feedback processes, are also important. The strength of the relationship between summer SLP anomalies and heat waves is improved (from 35%) to account for around 46% of its variability when summer Atlantic and Mediterranean SSTs and summer European precipitation anomalies are included as predictors. This indicates that these predictors are not completely collinear rather that they each have some contribution to accounting for summer heat wave variability. However, the simplicity and scale of the statistical analysis masks this complex interaction between variables. There is some useful predictive skill of summer heat waves using multiple lagged predictors. A CCA using preceding winter North Atlantic SSTs and preceding January to May Mediterranean total precipitation results in significant hindcast (1972–2003) Spearman rank correlation skill scores up to 0.55 with an average skill score over the domain equal to 0.28 ± 0.28. In agreement with previous studies focused on mean summer temperature, there appears to be some predictability of heat wave events on the decadal scale from the Atlantic Multidecadal Oscillation (AMO), although the long-term global mean temperature is also well related to western European heat waves. Combining these results with the observed positive trends in summer continental European SLP, North Atlantic SSTs and indications of a decline in European summer precipitation then possibly these long-term changes are also related to increased heat wave occurrence and it is important that the physical processes controlling these changes be more fully understood.

Assessment of the probability of extreme weather events and their potential effects in large conurbations

The likelihood of occurrence of extreme high-temperature run events is estimated for different values of the event intensity and persistence from very long synthetic time series of daily maximum temperatures generated by Monte Carlo simulations using a first-order autoregressive or Markov model. A theoretical analysis reveals a higher relative sensitivity of the simulated extreme event probabilities to changes in the variability of climate than to changes in its mean state. Moreover, this sensitivity relatively increases at a nonlinear rate the more extreme the event. The developed probabilistic model is applied in order to derive local scenarios of extreme high-temperature run events for a large conurbation like the city of Berlin assuming both arbitrary hypothetical and physically based new climate states described by changes in the model parameters (e.g. the mean, the standard deviation and the first-order autocorrelation of the daily maximum temperature time series). As a consequence of a 1.7°C increase in the mean as well as a 19% increase in the temperature variability in July as predicted by the climate model ECHAM_1/LSG assuming an unrestricted future increase in the global atmospheric concentration of climate relevant greenhouse gases according to the IPCC Scenario A (“Business as usual”) the intensity as well as the persistence of extreme high-temperature run events will rise considerably up to the end of the next century. In particular, intense hot spells characterized by at least five consecutive daily maximum temperatures equaling or exceeding 33°C are expected to occur every eight years under the new climate conditions compared to a current repetition time of about 47 years. The potential environmental effects might be a significant increase in the heat-stress-related morbidity and mortality rate, an aggravation of the summer smog situation and a destabilization of the urban ecosystems.