Next Article in Journal
Optimizing the Surrounding Building Configuration to Improve the Cooling Ability of Urban Parks on Surrounding Neighborhoods
Previous Article in Journal
Relevance and Reliability of Outdoor SO2 Monitoring in Low-Income Countries Using Low-Cost Sensors
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 14
Issue 6
10.3390/atmos14060913
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Evaluation of ERA5-Simulated Temperature and Its Extremes for Australia

by
Dipayan Choudhury
1,*,
Fei Ji
1,2,
Nidhi Nishant
2,3 and
Giovanni Di Virgilio
1,3
1
Climate and Atmospheric Science, NSW Department of Planning and Environment, Sydney 2141, Australia
2
Australian Research Council Centre of Excellence for Climate Extremes, University of New South Wales, Sydney 2052, Australia
3
Climate Change Research Centre, University of New South Wales, Sydney 2052, Australia
*
Author to whom correspondence should be addressed.
Atmosphere 2023, 14(6), 913; https://doi.org/10.3390/atmos14060913
Submission received: 31 March 2023 / Revised: 19 May 2023 / Accepted: 21 May 2023 / Published: 24 May 2023
(This article belongs to the Section Climatology)
Browse Figures
Versions Notes

Abstract

:
Atmospheric reanalysis products offer high-resolution and long-term gridded datasets that can often be used as an alternative or a supplement to observational data. Although more accessible than typical observational data and deemed fit for climate change studies, reanalysis data can show biases resulting from data assimilation approaches. Thus, a thorough evaluation of the reanalysis product over the region and metric of study is critical. Here, we evaluate the performance of the latest generation of ECMWF reanalysis, ERA5, in simulating mean and extreme temperatures over Australia for 1979–2020 versus high-quality gridded observations. We find ERA5 generally simulates maximum and minimum temperatures reasonably well (mean bias ~1.5 °C), even though it underestimates/overestimates the daily maximum/minimum temperatures, leading to a cold bias for Tmax and a warm bias for Tmin. ERA5 also underestimates the decadal warming trend in both Tmax and Tmin compared to the observations. Furthermore, ERA5 struggles to simulate the temporal variability of Tmin, leading to a markedly worse skill in Tmin than Tmax. In terms of extreme indices, ERA5 is skilled at capturing the spatial and temporal patterns and trends of extremes, albeit with the presence of biases in each index. This can partially be attributed to the warm bias in the minimum temperature. Overall, ERA5 captures the mean and extreme temperature indices over the Australian continent reasonably well, warranting its potential to supplement observations in aiding climate change-related studies, downscaling for boundary conditions, and climate model evaluation.

1. Introduction

Observations provide a key foundation for understanding long-term variability and change in climate, and for providing the underpinning for climate model evaluation and projections [1]. However, observations are unevenly distributed, and they come with uncertainties. In Australia, the majority of in-situ observational stations are located within 200 km of the coast, and a relatively sparse density of stations exists in the arid interior and high-elevation areas. While efforts have been made to convert in-situ station data to regular gridded data sets such as the Australian Water Availability Project (AWAP) [2] and more recently the Australian Gridded Climate Data (AGCD) [3], uncertainties still exist for AWAP and AGCD over data-sparse areas [4].
Additionally, observations alone struggle to provide a complete and accurate picture of the state of the Earth’s system, even when data from remote sensing and automatic weather stations are included. As an alternative to observations, reanalysis products are often used for performing climate analyses. Reanalysis combines past short-range weather forecasts with observations through data assimilation. The process mimics the production of day-to-day weather forecasts, which use an analysis of the current state of the Earth system as their starting point. The analysis is a physically consistent blend of observations with a short-range forecast based on the previous analysis [5,6,7,8,9]. Thus, these reanalysis products fill the gaps in the observational record, and they do so in a way that is consistent in time, thus minimising any spurious signals of change. Additionally, they provide spatially continuous data that eases climate model evaluation. Reanalyses are usually produced at lower resolutions than current weather forecasts, and use the same modern data assimilation system and forecasting model configuration throughout the reanalysis period for consistency. Prior to the production of a new reanalysis, work is carried out to improve the quality and availability of past observational data, for example, by digitising old paper-based records and reprocessing existing satellite records. Careful quality control is carried out as reanalyses are produced, and their reliability is assessed by comparison with reanalyses produced at other institutes. Reanalysis data have often been used for a wide range of applications, including monitoring climate change, downscaling, climate model evaluation and research, education, policymaking, and business, as well as sectors such as renewable energy and agriculture [10,11,12,13].
Reanalysis products are often taken as an alternative solution to observational weather and climate data due to availability and accessibility problems, particularly in data-sparse regions. However, reanalysis has its own uncertainties [14]. It is vital to evaluate reanalysis products against observations to quantify their skill and uncertainties before being used for applications such as dynamical downscaling.
ERA5 is the latest generation of ECMWF (European Centre for Medium-Range Weather Forecasts) atmospheric reanalysis [15]. The technical details of ERA5 are discussed in the next section. Several studies have evaluated ERA5 reanalysis for different variables such as convective parameters [16], tropical and subtropical cyclones [17,18], solar irradiance [19], wind resources [20], and stream flow [21]. Most studies have focused on precipitation [22,23,24,25,26,27], temperature [28], and a combination of both [29,30]. Gleixner et al. (2020) evaluated the performance of ERA5 reanalysis for near-surface temperature and precipitation and compared it with ERA-Interim over East Africa. They found ERA5 reduced climatological bias in near-surface temperature and precipitation, and had improved accuracy in representing inter-annual variability. However, both ERA5 and ERA-Interim still underestimated the observed long-term trends. Yu et al. (2021) evaluated ERA5 and ERA-Interim in simulating 2-m temperature and surface temperature in the Arctic and found both ERA5 and ERA-interim to be well correlated with surface monitoring data. However, they both showed the presence of warm biases for 2-m air temperature, though the bias was smaller for ERA5 compared with ERA-Interim.
While there is a plethora of variables that can be considered for validating reanalysis products against observations, here we focus on temperature extremes. Temperature plays a key role in modulating surface hydrology and the severity of droughts [31]. Studies have shown global warming to already result in more frequent extreme temperature events [32,33,34,35], which has had a dramatic impact on human life and caused economic and environmental damage [36,37,38]. Zander et al. (2015) showed that heat-related extremes lead to an annual economic burden of around US$6.2 billion (95% CI: 5.2–7.3 billion) for the Australian workforce. This amounts to 0.33 to 0.47% of Australia’s GDP (Gross Domestic Product). ERA5 evaluations have been undertaken in North America [16,26], East Africa [29], Europe [16,24,30], and Asia [19,21,22,23,25,27,28]. However, such an evaluation has not been conducted for Australia. In this study, we evaluate the skill of the ERA5 reanalysis dataset in simulating monthly means of daily maximum and minimum temperatures and temperature extremes over Australia.

2. Data

2.1. Observation

Gridded observations are taken from the Australian Gridded Climate Data, AGCD [3], which has a gridded resolution of 0.05° × 0.05°. AGCD is the official dataset of the Australian Bureau of Meteorology, which uses advanced statistical and scientific modelling to combine data from a large array of in situ stations. These in-situ stations have a much higher density over the coastal populated regions and a lower density over the western and central parts of Australia. The two variables considered here are maximum temperature (Tmax) and minimum temperature (Tmin).

2.2. ERA5 Data

ERA5 is the fifth-generation ECMWF atmospheric reanalysis of the global climate, covering the period from January 1979 to the present [15]. ERA5 provides hourly estimates of a plethora of atmospheric, land, and oceanic climate variables. The data is on a 30 km horizontal grid and resolves the atmosphere using 137 levels from the surface up to a height of 80 km.
ERA5 is based on the Integrated Forecasting System (IFS) Cy41r2 that was operational until 2016. ERA5 thus benefits from a decade of developments in model physics, core dynamics, and data assimilation. In addition to a significantly enhanced horizontal resolution of 0.25° latitude (~31 km), compared to 80 km for the previous generation (ERA-Interim), ERA5 has hourly output throughout, and uncertainty estimate from an ensemble (3-hourly at half the horizontal resolution). Additionally, ERA5 has 137 vertical levels extending up to 0.01 hPa. ERA5 combines satellite and in-situ observations into its global estimates using advanced modelling and data assimilation techniques; therefore, ERA5 has been shown to have more accurate simulations compared to previous generations [15,39,40].

3. Methods

We re-grid ERA5 data to match the AGCD grid using bilinear interpolation. Evaluation of the mean climate is undertaken on AGCD grids. The climate extreme indices are calculated on ERA5 native grids and then interpolated to AGCD grids for evaluation. We perform our ERA5 and AGCD evaluations over the common period between the two datasets, 1979–2020.
We first evaluate the monthly mean values of the daily maximum and daily minimum temperatures simulated by ERA5 over the Australian landmass in comparison with those from AGCD. Next, we assess ERA5’s skill in capturing a handful of climate extreme indices, as described below.

3.1. Climate Extreme Indices—ET-SCI

We evaluate climate extremes based on daily temperatures as defined by the World Meteorological Organization’s Expert Team on Sector-specific Climate Indices (ET-SCI; [41]). The ET-SCI team of climate scientists, in collaboration with sector experts, defines a set of climate extreme indices that are directly relevant to each of the sectors considered. These indices are widely used in the evaluation of climate model simulations for capturing climate extremes [42,43]. We use the recommended ClimPACT version 2 software to calculate the ET-SCI indices (https://climpact-sci.org/, accessed on 23 May 2023), focusing on daily maximum and minimum temperatures.
Although ClimPACT produces more than 36 temperature-related indices, we select only six key indices based on the following considerations:
  • To capture key aspects of temperature extremes. We choose absolute indices (e.g., coldest day minimum temperature (TNn)), threshold-based indices (e.g., number of warm days (TX90P), number of warm nights (TN90p), number of days when minimum temperature is less than 2 °C (TNlt2)), and duration-based indices (e.g., warm spell duration index (WSDI) and cold spell duration index (CSDI)).
  • To capture extremes that have impacts on society and infrastructure; for example, extreme indices like TNlt2 and WSDI have clear impacts on human health [44].

3.2. Evaluation Approaches

There are multiple metrics for evaluating climate models as proposed and applied by previous studies. The metrics range from summarising features of the spatial distribution of the climatological mean state to summarising temporal variability. Some of the common and widely used metrics for evaluating the spatial distribution of models against observations are bias (difference between the modelled and observed climatological mean field; here smaller values indicate better performance of the model simulation) and pattern correlation (PCorr: correlation between the modelled and observed climatological mean field; Equation (1) [45]):
P C o r r = ∑ ( x m − x m ¯ ) ( x o − x o ¯ ) ∑ ( x m − x m ¯ ) 2     ∑ ( x o − x o ¯ ) 2
where x m and x o denote the values of the variable in the sample of model simulations and the observational data sets, respectively, and x m ¯ and x o ¯ denote the means of the variable in the sample of model simulations and the observational data sets, respectively. Here, larger PCorr values indicate better performance of the model simulation.
For the mean field and each climate extreme index in Table 1, we calculate the bias and PCorr to compare ERA5 with AGCD. To investigate the skill of ERA5 in capturing temporal variability, we compare the Coefficient of Variation (CV, Equation (2)), decadal trends, and interannual variability between the two datasets.
C V = s m
where s is the standard deviation and m is the mean of each dataset.
The decadal trends in AGCD and ERA5 are calculated using the Mann-Kendall test [46,47,48,49] to check for differences in long-term trends between the two datasets.
We perform our analyses primarily for the annual timescale but also extend them to the austral seasons. The four seasons are summer (December–January–February, DJF), autumn (March–April–May, MAM), winter (June–July–August, JJA), and spring (September–October–November, SON). Alongside presenting results over the Australian mainland and Tasmania, we also evaluate the performance over the four Australian National Resource Management (NRM) regions (Figure S1). These are Northern Australia (NA), Eastern Australia (EA), Southern Australia (SA), and the Rangelands ®. These regions are in line with the climate zones [50], are associated with the IPCC reference regions [51], and are commonly used for evaluations over the Australian region [52].

4. Results

4.1. Mean Climate (Daily Maximum and Minimum Temperatures)

This section evaluates the daily maximum and minimum temperatures from ERA5 over Australia by comparing them against AGCD. Evaluation is performed primarily using bias maps and correlations. The coefficients of variation and the decadal trends in maximum and minimum temperatures are further compared between ERA5 and AGCD. The results section in the main text primarily focuses on annual mean values, with corresponding seasonal analyses being presented in the Supplementary Information. We finish off by presenting regional analyses of inter-annual and seasonal variabilities for the NRM regions.

4.1.1. Biases and Correlations

We first evaluate Tmax and Tmin simulated by ERA5 against gridded observations (AGCD). The annual mean values of Tmax and Tmin from AGCD and ERA5 and the bias in ERA5 compared to AGCD are shown in Figure 1. Results show that ERA5 generally replicates the spatial patterns reasonably well, albeit with an underestimation of the larger values (>32 °C). Compared to AGCD, ERA5 simulates around 1–2 °C cooler values over most of the continent for Tmax (Figure 1a–c). Nevertheless, positive biases up to 2.5 °C are evident over parts of the western coast, northern Queensland, and the south-east coast. These discrepancies can largely be attributed to the sparsity of observing stations over these areas in the cases of the west/north [3], as well as the difference in elevation between ERA5 grid points and stations [53,54], such as the Great Dividing Range in the east. ERA5 shows weaker skill in simulating Tmin compared to Tmax (Figure 1d–f). Clear discrepancies in the simulated spatial patterns are evident, with Tmin values that are 3–4 °C warmer than AGCD over most of the continent. This warming bias is generally stronger in the western half of Australia compared to the eastern half. However, small cold biases are evident over small areas of northern Queensland and the Northern Territory. Similar results of ERA5 underperforming in simulating minimum temperatures compared to maximum temperatures over China have been reported by previous studies, such as Xu et al., 2022.
Analyses for the seasonal means of Tmax and Tmin are presented in Figures S2 and S3 in the Supplementary Information. The results are largely consistent with the findings shown in Figure 1, with the simulation of Tmax being generally better than Tmin. The biases in the seasonal values are in the same order of magnitude as the annual counterparts, with a couple of exceptions. The largest seasonal bias in Tmax is evident for the austral summer (Figure S2c), while the largest bias in Tmin is for the austral winter and spring seasons (Figure S3i,l). Although the largest seasonal bias in Tmax is ~2.5 °C for DJF (Figure S2c), the largest seasonal bias in Tmin is of the order of ~5 °C for JJA (Figure S3i). It is worth mentioning here that the spots of concentrated bias in Figure 1c,f (and in the seasonal biases of Figures S2 and S3) evident over South Australia can be attributed to the lack of observational stations over the South Australian lakes.
These patterns of the skill of simulated Tmax and Tmin from ERA5 are further demonstrated in the temporal correlations between ERA5 and AGCD. Figure 2 shows the temporal correlations of annual mean Tmax and Tmin between ERA5 and AGCD. Akin to Figure 1c,f, Tmin shows a widespread reduction in skill compared to Tmax. Correlations between the seasonal mean values are presented in Figure S4 in the Supplementary Information. The correlations for seasonal mean Tmax values are similar in magnitude to those for the annual means. However, correlations of the seasonal mean Tmin are worse than the annual mean across all seasons. Furthermore, the correlations of seasonal mean values (Figure S4) show a different variability than the biases of seasonal mean values (Figures S2 and S3). While Tmax shows the strongest biases for DJF (Figure S2c), the weakest correlations are evident for SON (Figure S4g). Although the worst seasonal Tmin biases as well as correlations are noted for JJA, all seasons show worse correlations than the annual mean correlations (Figure S4 compared to Figure S3).

4.1.2. Variability and Trends

To explore this discrepancy between Tmin and Tmax, the variability in annual mean Tmax and Tmin captured by ERA5 and AGCD is presented in Figure 3 using the coefficient of variation (CV). By its definition, CV helps capture the variability in the dataset relative to its mean. Tmin shows larger variability than Tmax, a pattern that is replicated across both ERA5 and AGCD. For Tmax, ERA5 captures this variability relatively well compared to observations (Figure 3b vs. Figure 3a). This is with the exception of ERA5 simulating up to 40% more variability than AGCD over Tasmania (Figure 3c). The skill of simulated variability in Tmin from ERA5 is markedly worse than Tmax (Figure 3d–f). ERA5 simulates up to 30% less variability in annual mean Tmin compared to observations (Figure 3f). This discrepancy is most pronounced in southern and central Australia. The coefficients of variation for seasonal mean Tmax and Tmin values from ERA5 and AGCD are shown in Figures S5 and S6. Similar to annual mean values, seasonal mean values of Tmin (Figure S6) show much larger variability than Tmax (Figure S5). The highest variability is seen for Tmin in the winter, with CVs of up to 30%. For seasonal mean Tmax, ERA5 generally simulates a larger variability than AGCD, which is widespread over the continent for all seasons except the austral spring (Figure S5). This signal is opposite for Tmin. ERA5 simulates widespread, weaker variability in Tmin for all seasons. The largest discrepancy is seen for Tmin for the winter season, with ERA5 simulating up to 33% smaller variability than AGCD.
To further investigate the variability evident in observations and ERA5 simulations, we assess the decadal trends in Tmax and Tmin (Figure 4). AGCD shows strong warming trends over Australia for both Tmax and Tmin (Figure 4a,d), although this signal is much stronger for Tmax. For Tmax, AGCD shows the warming signal primarily over southern Australia, with the peak of 0.85 °C/decade situated over southwestern Australia (Figure 4a). Local warming trends are also present in southern New South Wales (NSW) and Victoria. Similar to AGCD, ERA5 shows a warming trend in Tmax, albeit generally smaller over most of the continent. Nevertheless, the highest warming trend (0.92 °C/decade) is slightly higher than that from AGCD (Figure 4b). The spatial patterns are also different, with the peak warming signals being further inland than for AGCD (Figure 4b). This difference is better summarised in Figure 4c, which shows a clear underestimation of the warming trend over western Australia and the entire eastern coast of Australia. Due to the higher simulated trend over certain regions and different spatial distribution, ERA5 simulates a stronger trend of up to 0.4 °C/decade over several different regions, though the overall warming signal is weaker than AGCD (Figure 4c). Additionally, ERA5 simulates a stronger warming signal over Tasmania for Tmax (Figure 4c).
Similar to Tmax, Tmin generally shows a warming decadal trend over most of the continent for both AGCD and ERA5 (Figure 4d,e). However, unlike Tmax, Tmin shows a small cooling trend over northern Western Australia in AGCD (Figure 4d). This cooling trend is overestimated in ERA5 (Figure 4e). The decadal warming trend in Tmin simulated by ERA5 is generally weaker than that for AGCD, although there is a stronger warming signal zonally across central Australia. Figure 4f summarises these findings and shows that ERA5 generally underestimates the warming trend by up to 0.43 °C/decade, with an overestimation over the western coast of up to 0.3 °C/decade. Akin to Tmax, ERA5 simulates a stronger warming signal over Tasmania compared to AGCD for Tmin.

4.1.3. Regional Analysis

The results presented above for the Australian mainland are broadly consistent with those for the four NRM regions. Figure 5a,b shows that ERA5 underestimates both Tmax and Tmin for all the regions (akin to Figure 1c,f) across the entire timeseries. That is, the simulation of Tmax by ERA5 is colder than AGCD, and vice versa for Tmin. However, the interannual variability is captured relatively well by ERA5, suggesting a possible stationary bias between the two datasets. The biases in Tmax are smaller than Tmin for all regions. These biases are lowest for NA, followed by EA, with comparable values over SA and R for Tmax (Figure 5a). For Tmin, the lowest difference is seen for EA, with the largest differences for R (Figure 5b). Monthly variability of ERA5 compared to AGCD shows that the largest discrepancy in Tmax is noted for the summer months (Figure 5c, also Figure S1), while that for Tmin is noted for the winter months (Figure 5d, also Figure S2). These patterns are most pronounced for NA and smallest for R.

4.2. Climate Extremes

This section evaluates the six target climate extreme indices (Table 1) from ERA5 over Australia by comparing them against AGCD. Evaluations are performed primarily using spatial bias maps and temporal correlations. We also assess the interannual variability and decadal trends in the simulated ERA5 indices and compare these with AGCD to further investigate any discrepancies.

4.2.1. Observed Climate Extremes and Spatial Bias

Annual mean biases in the six temperature extremes are shown in Figure 6. For duration-related extremes (CSDI and WSDI, first two columns), there is a clear north-to-south gradient in AGCD (Figure 6a,b), which is well simulated in ERA5 (Figure 6g,h). While the spatial distributions are well captured, clear biases are evident in ERA5. Biases show a general overestimation of values in ERA5 over the domain, with an underestimation over central Australia. The largest biases are present in Central and Western Australia, which could be related to uncertainties from the limited stations available in these regions for AGCD. Nevertheless, these biases are typically small in magnitude (~2 days).
TNlt2 has a general south-to-north gradient, with more days with a minimum temperature below 2 °C located in southern Australia, especially over south-east Australia, where mountain ranges are located (third column in Figure 6). ERA5 captures the spatial distribution of TNlt2 relatively well (Figure 6i), although it underestimates cold days (Figure 6o). This bias could be attributed to the warm bias in the minimum temperature reported earlier (Figure 1f). The biases are stronger (~5 days) in the southern part of the domain as compared to the northern part of the domain (~1 day). TNn has a north-west to south-east gradient, with higher TNn values in northern Australia and lower TNn in southern Australia, especially over the high-elevation south-eastern regions (last column in Figure 6). ERA5 captures the spatial TNn pattern of AGCD relatively well but shows a warm bias throughout the domain (Figure 6r). These biases remain relatively consistent over most of the domain (2–4 °C).
Both TN90p and TX90p (fourth and fifth columns in Figure 6) generally have larger values in Northern Australia and smaller values in Southern Australia. These patterns are captured relatively well in ERA5. The biases between ERA5 and AGCD are small (<2 days) but show large spatial variability. For example, TN90p shows overestimation over most of the domain except for western Australia (Figure 6p), whereas TX90p typically shows positive biases in western Australia and negative biases over eastern Australia (Figure 6q).
In general, ERA5 captures the spatial patterns of the observed temperature extreme indices very well, with small relative biases. The warm biases noticed for colder indices can potentially be attributed to the warm biases in simulated minimum temperatures.

4.2.2. Temporal Correlations

Figure 7 presents the temporal correlations between ERA5 and AGCD for the six temperature and climate extreme indices. Similar to the strong temporal correlation between ERA5 and AGCD for mean maximum/minimum temperature (Figure 2), the temporal correlations for these extreme indices are large (Figure 7).
For extremes like WSDI and TNn, the correlation varies from reasonably good (above 0.6) to very good (above 0.9) between ERA5 and AGCD for most of the domain. WSDI shows comparatively stronger correlations than CSDI and TNn over most of the domain. TNlt2 shows a different spatial pattern of correlation as compared to the other extreme indices. It has a very strong temporal correlation for both northern and southern Australia, but relatively smaller correlations for central Australia. This could be related to warm biases in the minimum temperature. In northern Australia, the minimum temperature is relatively high, which leads to almost zero TNlt2. Additionally, the bias in TNlt2 was also found to be almost zero (Figure 6o), which would correspond to the correlation of almost 1.0 over this region in Figure 7c. For TN90p and TX90p, very large positive correlations (more than 0.9) are found between ERA5 and AGCD, with TX90P being slightly better than TN90p (Figure 7d,e).

4.2.3. Variability and Trends

Figure 8 shows the decadal trends of each of the climate extreme indices for AGCD and ERA5. The indices generally show a very weak negative to strong positive trend, with a reasonably good agreement between ERA5 and AGCD. CSDI and TNn typically show very weak negative to zero trends in both AGCD and ERA data sets, suggesting declines in cold day duration and low temperature extremes (Figure 8a,g, and Figure 8f,i). WSDI and TNlt2 show positive trends over the northern and south-eastern parts of Australia, respectively, and weak negative trends elsewhere in both datasets (Figure 8b,h, and Figure 8c,i). TN90p and TX90p show a stronger positive trend over most of the domain, suggesting an increase in the number of warm nights and warm days (Figure 8d,j, and Figure 8e,k). The biggest disagreement between the two datasets is evident for TN90p and TNn, with ERA5 simulating a negative trend over north-western Australia for TN90p and a stronger negative trend than AGCD for TNn.

4.2.4. Regional Analysis

Next, we investigate interannual variability over the four NRM regions to assess any regional discrepancies between the two datasets. Examining the performance of ERA5 over these regions is particularly important when considering the design of regional downscaling experiments or assessing regional-scale climate impacts.
Figure 9 shows a good agreement between ERA5 and AGCD for all the NRM regions for all extreme indices, except TNlt2 and TNn. ERA5 underestimates TNlt2 but overestimates TNn for most of the regions, which could result from the warm biases in minimum temperatures reported in Figure 1. Although there is a clear difference between these two indices, ERA5 generally replicates the interannual variability noticed in AGCD reasonably well.

5. Discussion and Conclusions

This paper presents an analysis of the skill of the latest generation of the ECMWF reanalysis product, ERA5, in simulating temperature and its extremes over the Australian region. Our analysis of maximum and minimum daily temperatures shows that ERA5 captures the spatial patterns relatively well, with a general underestimation of the values relative to AGCD. However, the mean annual bias in Tmin (+3.6 °C, Figure 1f) is generally larger than that for Tmax (−1.7 °C, Figure 1c) for most of the Australian mainland. While these biases are consistent across the seasons, they are further exacerbated in the austral summer for Tmax, and the austral winter and spring for Tmin. ERA5’s reduced skill in simulating Tmin compared to Tmax over the Australian mainland is clearly evident in the much smaller correlation skill of Tmin compared to Tmax across the annual as well as seasonal timescales (Figure 2 and Figure S3). We further investigate this difference between Tmin and Tmax biases by looking at the simulated variability and find that ERA5 struggles in simulating the variability of Tmin (Figure 3). This, along with Tmin having a larger variability than Tmax in AGCD (Figure 3a,d), could partly explain the larger biases in Tmin compared to Tmax. Furthermore, ERA5 underestimates the decadal warming trend in both Tmax and Tmin compared to AGCD overall, with Tmin ERA5 simulations being markedly worse than Tmax. In addition, ERA5 simulates stronger warming trends over several different regions of the Australian mainland in Tmax and shows a cooling decadal trend in Tmin over the western Northern Territory (Figure 4). A more detailed analysis over the four NRM regions shows that while ERA5 simulates colder Tmax and warmer Tmin compared to AGCD, it captures the interannual and monthly variability reasonably well, suggesting the possibility of a stationary bias between the two datasets.
ERA5 simulates the spatial distribution of observed temperature extreme indices reasonably well, although clear biases are noticed for each extreme index. Warm biases noted in the ERA5 minimum temperatures result in positive biases in all extreme indices except for TNlt2. Temporal correlations of temperature extremes between ERA5 and AGCD are generally good, especially for TX90p and TN90p. The spatial patterns and magnitudes of decadal trends of these extreme indices are generally well simulated in ERA5, albeit with small biases overall. The interannual variability of the indices is also reasonably replicated across the NRM regions, although small stationary biases can be seen in some instances. It is worth noting that the biases identified in these extreme indices would need to be considered when the indices are used for climate change impact applications across the multiple sectors mentioned in Table 1. However, instructions on how to best adopt our analyses for such applications are beyond the scope of this study and are left for sector experts to develop.
While these results shed light on the skill of the latest generation of the ERA reanalysis dataset, there are certain limitations to the study and possible avenues for future work. The AGCD dataset is treated as observations here, which are obtained by interpolating data across the in-situ stations. There is a sparsity of such stations over areas of north-western and central Australia and also over high elevations, which could contribute to uncertainties in the gridded temperature dataset. However, temperature not being as localised as precipitation could alleviate some of these concerns. Additionally, we focus our analysis on a handful of the ET-SCI-defined temperature extreme indices. This study could be extended to other extreme indices that may be necessary for specific study regions and applications. Nevertheless, our study shows that ERA5 captures the mean and extreme temperature indices over the Australian continent reasonably well. Thus, ERA5 provides immense potential to supplement observational datasets for climate change impact assessments, climate model evaluation, downscaling for boundary condition generation, and other applications across Australia.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/atmos14060913/s1.

Author Contributions

Conceptualization, D.C., F.J. and N.N.; methodology, D.C., N.N. and F.J.; software, N.N. and D.C.; validation, F.J., D.C. and G.D.V.; formal analysis, All; investigation, F.J. and D.C.; resources, N.N.; data curation, N.N.; writing—original draft preparation, D.C., F.J. and G.D.V.; writing—review and editing, D.C., F.J. and G.D.V.; visualization, D.C. and N.N.; supervision, G.D.V.; project administration, D.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the New South Wales (NSW) Government Climate Change Fund.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Details about AGCD are available at the Australian Bureau of Meteorology website (http://www.bom.gov.au/metadata/catalogue/19115/ANZCW0503900567, (accessed on 2 February 2023)). The dataset is available on the NCI (National Computational Infrastructure) server in project zv2. Detail on how to access the data can be found at http://climate-cms.wikis.unsw.edu.au/AGCD, (accessed on 2 February 2023). ERA5 data is available on the NCI in Project rt52.

Acknowledgments

This work is made possible by funding from the NSW Climate Change Fund and was supported by the NSW Department of Planning and Environment. The modelling work was undertaken on the National Computational Infrastructure (NCI) high-performance computers in Canberra, Australia, which are supported by the Australian Commonwealth Government.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Alexander, L.V. Global observed long-term changes in temperature and precipitation extremes: A review of progress and limitations in IPCC assessments and beyond. Weather. Clim. Extremes 2016, 11, 4–16. [Google Scholar] [CrossRef]
  2. Jones, D.A. William Wang, and Robert Fawcett. High-quality spatial climate data-sets for Australia. Aust. Meteorol. Oceanogr. J. 2009, 58, 233. [Google Scholar] [CrossRef]
  3. Evans, A.; Jones, D.; Smalley, R.; Lellyett, S. An Enhanced Gridded Rainfall Analysis Scheme for Australia; Australian Bureau of Meteorology: Melbourne, VIC, Australia, 2020; Volume 66, pp. 55–67. [Google Scholar]
  4. Chubb, T.; Manton, M.; Siems, S.; Peace, A. Evaluation of the AWAP daily precipitation spatial analysis with an independent gauge network in the Snowy Mountains. J. South. Hemisph. Earth Syst. Sci. 2016, 66, 55–67. [Google Scholar] [CrossRef]
  5. Poli, P.; Hersbach, H.; Dee, D.P.; Berrisford, P.; Simmons, A.J.; Vitart, F.; Laloyaux, P.; Tan, D.G.H.; Peubey, C.; Thépaut, J.-N.; et al. ERA-20C: An Atmospheric Reanalysis of the Twentieth Century. J. Clim. 2016, 29, 4083–4097. [Google Scholar] [CrossRef]
  6. Kobayashi, S.; Ota, Y.; Harada, Y.; Ebita, A.; Moriya, M.; Onoda, H.; Onogi, K.; Kamahori, H.; Kobayashi, C.; Endo, H.; et al. The JRA-55 Reanalysis: General Specifications and Basic Characteristics. J. Meteorol. Soc. Jpn. Ser. II 2015, 93, 5–48. [Google Scholar] [CrossRef]
  7. Saha, S.; Moorthi, S.; Pan, H.-L.; Wu, X.; Wang, J.; Nadiga, S.; Tripp, P.; Kistler, R.; Woollen, J.; Behringer, D.; et al. The NCEP Climate Forecast System Reanalysis. Bull. Am. Meteorol. Soc. 2010, 91, 1015–1058. [Google Scholar] [CrossRef]
  8. Kalnay, E.; Kanamitsu, M.; Kistler, R.; Collins, W.; Deaven, D.; Gandin, L.; Iredell, M.; Saha, S.; White, G.; Woollen, J.; et al. The NCEP/NCAR 40-year reanalysis project. Bull. Am. Meteorol. Soc. 1996, 77, 437–472. [Google Scholar] [CrossRef]
  9. Dee, D.P.; Uppala, S.M.; Simmons, A.J.; Berrisford, P.; Poli, P.; Kobayashi, S.; Andrae, U.; Balmaseda, M.A.; Balsamo, G.; Bauer, P.; et al. The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Q. J. R. Meteorol. Soc. 2011, 137, 553–597. [Google Scholar] [CrossRef]
  10. Frank, C.W.; Wahl, S.; Keller, J.D.; Pospichal, B.; Hense, A.; Crewell, S. Bias correction of a novel European reanalysis data set for solar energy applications. Sol. Energy 2018, 164, 12–24. [Google Scholar] [CrossRef]
  11. Muñoz-Sabater, J.; Dutra, E.; Agustí-Panareda, A.; Albergel, C.; Arduini, G.; Balsamo, G.; Boussetta, S.; Choulga, M.; Harrigan, S.; Hersbach, H.; et al. ERA5-Land: A state-of-the-art global reanalysis dataset for land applications. Earth Syst. Sci. Data 2021, 13, 4349–4383. [Google Scholar] [CrossRef]
  12. Di Napoli, C.; Barnard, C.; Prudhomme, C.; Cloke, H.L.; Pappenberger, F. ERA5-HEAT: A global gridded historical dataset of human thermal comfort indices from climate reanalysis. Geosci. Data J. 2021, 8, 2–10. [Google Scholar] [CrossRef]
  13. Wu, T.; Song, L.; Li, W.; Wang, Z.; Zhang, H.; Xin, X.; Zhang, Y.; Zhang, L.; Li, J.; Wu, F.; et al. An overview of BCC climate system model development and application for climate change studies. J. Meteorol. Res. 2014, 28, 34–56. [Google Scholar] [CrossRef]
  14. Biswas, S.; Chand, S.S.; Dowdy, A.J.; Wright, W.; Foale, C.; Zhao, X.; Deo, A. Statistical Calibration of Long-Term Reanalysis Data for Australian Fire Weather Conditions. J. Appl. Meteorol. Clim. 2022, 61, 729–758. [Google Scholar] [CrossRef]
  15. Hersbach, H.; Bell, B.; Berrisford, P.; Hirahara, S.; Horányi, A.; Muñoz-Sabater, J.; Nicolas, J.; Peubey, C.; Radu, R.; Schepers, D.; et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 2020, 146, 1999–2049. [Google Scholar] [CrossRef]
  16. Taszarek, M.; Pilguj, N.; Allen, J.T.; Gensini, V.; Brooks, H.E.; Szuster, P. Comparison of convective parameters derived from ERA5 and MERRA2 with rawinsonde data over Europe and North America. J. Clim. 2021, 34, 3211–3237. [Google Scholar] [CrossRef]
  17. Slocum, C.J.; Razin, M.N.; Knaff, J.A.; Stow, J.P. Does ERA5 Mark a New Era for Resolving the Tropical Cyclone Environment? J. Clim. 2022, 35, 7147–7164. [Google Scholar] [CrossRef]
  18. Karwat, A.; Franzke, C.L.E.; Blender, R. Long-Term Trends of Northern Hemispheric Winter Cyclones in the Extended ERA5 Reanalysis. J. Geophys. Res. Atmos. 2022, 127, e2022JD036952. [Google Scholar] [CrossRef]
  19. Sianturi, Y.; Marjuki; Sartika, K. Evaluation of ERA5 and MERRA2 reanalyses to estimate solar irradiance using ground observations over Indonesia region. In Proceedings of the AIP Conference, Yogyakarta, Indonesia, 30–31 October 2019; Volume 2223, p. 020002. [Google Scholar]
  20. Gualtieri, G. Reliability of ERA5 Reanalysis Data for Wind Resource Assessment: A Comparison against Tall Towers. Energies 2021, 14, 4169. [Google Scholar] [CrossRef]
  21. Sun, H.; Su, F.; Yao, T.; He, Z.; Tang, G.; Huang, J.; Zheng, B.; Meng, F.; Ou, T.; Chen, D. General overestimation of ERA5 precipitation in flow simulations for High Mountain Asia basins. Environ. Res. Commun. 2021, 3, 121003. [Google Scholar] [CrossRef]
  22. Jiang, Q.; Li, W.; Fan, Z.; He, X.; Sun, W.; Chen, S.; Wen, J.; Gao, J.; Wang, J. Evaluation of the ERA5 reanalysis precipitation dataset over Chinese Mainland. J. Hydrol. 2021, 595, 125660. [Google Scholar] [CrossRef]
  23. Xin, Y.; Lu, N.; Jiang, H.; Liu, Y.; Yao, L. Performance of ERA5 reanalysis precipitation products in the Guangdong-Hong Kong-Macao greater Bay Area, China. J. Hydrol. 2021, 602, 126791. [Google Scholar] [CrossRef]
  24. Bandhauer, M.; Isotta, F.; Lakatos, M.; Lussana, C.; Båserud, L.; Izsák, B.; Szentes, O.; Tveito, O.E.; Frei, C. Evaluation of daily precipitation analyses in E-OBS (v19.0e) and ERA5 by comparison to regional high-resolution datasets in European regions. Int. J. Clim. 2022, 42, 727–747. [Google Scholar] [CrossRef]
  25. Izadi, N.; Karakani, E.G.; Saadatabadi, A.R.; Shamsipour, A.; Fattahi, E.; Habibi, M. Evaluation of ERA5 Precipitation Accuracy Based on Various Time Scales over Iran during 2000–2018. Water 2021, 13, 2538. [Google Scholar] [CrossRef]
  26. Tarek, M.; Brissette, F.P.; Arsenault, R. Evaluation of the ERA5 reanalysis as a potential reference dataset for hydrological modelling over North America. Hydrol. Earth Syst. Sci. 2020, 24, 2527–2544. [Google Scholar] [CrossRef]
  27. Jiao, D.; Xu, N.; Yang, F.; Xu, K. Evaluation of spatial-temporal variation performance of ERA5 precipitation data in China. Sci. Rep. 2021, 11, 17956. [Google Scholar] [CrossRef]
  28. Yu, Y.; Xiao, W.; Zhang, Z.; Cheng, X.; Hui, F.; Zhao, J. Evaluation of 2-m Air Temperature and Surface Temperature from ERA5 and ERA-I Using Buoy Observations in the Arctic during 2010–2020. Remote Sens. 2021, 13, 2813. [Google Scholar] [CrossRef]
  29. Gleixner, S.; Demissie, T.; Diro, G.T. Did ERA5 Improve Temperature and Precipitation Reanalysis over East Africa? Atmosphere 2020, 11, 996. [Google Scholar] [CrossRef]
  30. Delhasse, A.; Kittel, C.; Amory, C.; Hofer, S.; van As, D.; Fausto, R.S.; Fettweis, X. Brief communication: Evaluation of the near-surface climate in ERA5 over the Greenland Ice Sheet. Cryosphere 2020, 14, 957–965. [Google Scholar] [CrossRef]
  31. Brunner, M.I.; Swain, D.L.; Gilleland, E.; Wood, A.W. Increasing importance of temperature as a contributor to the spatial extent of streamflow drought. Environ. Res. Lett. 2021, 16, 024038. [Google Scholar] [CrossRef]
  32. Alexander, L.V.; Zhang, X.; Peterson, T.C.; Caesar, J.; Gleason, B.; Tank, A.M.G.K.; Haylock, M.; Collins, D.; Trewin, B.; Rahimzadeh, F.; et al. Global observed changes in daily climate extremes of temperature and precipitation. J. Geophys. Res. Atmos. 2006, 111, 1042–1063. [Google Scholar] [CrossRef]
  33. Alexander, L.V.; Arblaster, J.M. Historical and projected trends in temperature and precipitation extremes in Australia in observations and CMIP5. Weather. Clim. Extrem. 2017, 15, 34–56. [Google Scholar] [CrossRef]
  34. Fischer, E.M.; Knutti, R. Anthropogenic contribution to global occurrence of heavy-precipitation and high-temperature extremes. Nat. Clim. Chang. 2015, 5, 560–564. [Google Scholar] [CrossRef]
  35. Perkins-Kirkpatrick, S.E.; Gibson, P.B. Changes in regional heatwave characteristics as a function of increasing global temperature. Sci. Rep. 2017, 7, 12256. [Google Scholar] [CrossRef] [PubMed]
  36. Kiem, A.S.; Johnson, F.; Westra, S.; van Dijk, A.; Evans, J.P.; O’donnell, A.; Rouillard, A.; Barr, C.; Tyler, J.; Thyer, M.; et al. Natural hazards in Australia: Droughts. Clim. Chang. 2016, 139, 37–54. [Google Scholar] [CrossRef]
  37. Zander, K.; Botzen, W.J.W.; Oppermann, E.; Kjellstrom, T.; Garnett, S.T. Heat stress causes substantial labour productivity loss in Australia. Nat. Clim. Chang. 2015, 5, 647–651. [Google Scholar] [CrossRef]
  38. Perkins-Kirkpatrick, S.E.; White, C.J.; Alexander, L.V.; Argüeso, D.; Boschat, G.; Cowan, T.; Evans, J.P.; Ekström, M.; Oliver, E.C.J.; Phatak, A.; et al. Natural hazards in Australia: Heatwaves. Clim. Chang. 2016, 139, 101–114. [Google Scholar] [CrossRef]
  39. Hoffmann, L.; Günther, G.; Li, D.; Stein, O.; Wu, X.; Griessbach, S.; Heng, Y.; Konopka, P.; Müller, R.; Vogel, B.; et al. From ERA-Interim to ERA5: The considerable impact of ECMWF’s next-generation reanalysis on Lagrangian transport simulations. Atmos. Chem. Phys. 2019, 19, 3097–3124. [Google Scholar] [CrossRef]
  40. Nogueira, M. Inter-comparison of ERA-5, ERA-interim and GPCP rainfall over the last 40 years: Process-based analysis of systematic and random differences. J. Hydrol. 2020, 583, 124632. [Google Scholar] [CrossRef]
  41. Alexander, L.; Herold, N. ClimPACT2: Indices and Software; Version 2. 2016. Available online: https://climpact-sci.org (accessed on 30 March 2023).
  42. Ji, F.; Nishant, N.; Evans, J.P.; Di Virgilio, G.; Cheung, K.K.; Tam, E.; Beyer, K.; Riley, M.L. Introducing NARCliM1.5: Evaluation and projection of climate extremes for southeast Australia. Weather. Clim. Extrem. 2022, 38, 100526. [Google Scholar] [CrossRef]
  43. Nishant, N.; Di Virgilio, G.; Ji, F.; Tam, E.; Beyer, K.; Riley, M.L. Evaluation of Present-Day CMIP6 Model Simulations of Extreme Precipitation and Temperature over the Australian Continent. Atmosphere 2022, 13, 1478. [Google Scholar] [CrossRef]
  44. Zivin, J.G.; Shrader, J. Temperature Extremes, Health, and Human Capital. Futur. Child. 2016, 26, 31–50. [Google Scholar] [CrossRef]
  45. Cohen, I.; Huang, Y.; Chen, J.; Benesty, J. Pearson Correlation Coefficient. In Noise Reduction in Speech Processing; Springer: Berlin/Heidelberg, Germany, 2009. [Google Scholar]
  46. Kendall, M.G. Rank Correlation Methods, 4th ed.; Charles Griffin: London, UK, 1975. [Google Scholar]
  47. Mann, H.B. Nonparametric tests against trend. Econometrica 1945, 13, 245–259. [Google Scholar] [CrossRef]
  48. Hussain, M.; Mahmud, I. pyMannKendall: A python package for non parametric Mann Kendall family of trend tests. J. Open Source Softw. 2019, 4, 1556. [Google Scholar] [CrossRef]
  49. Yue, S.; Wang, C.Y. The Mann-Kendall Test Modified by Effective Sample Size to Detect Trend in Serially Correlated Hydrological Series. Water Resour. Manag. 2004, 18, 201–218. [Google Scholar] [CrossRef]
  50. Whetton, P.; Ekström, M.; Gerbing, C.; Grose, M.; Bhend, J.; Webb, L.; Risbey, J.; Holper, P.; Clarke, J.; Hennessy, K.; et al. CSIRO and Bureau of Meteorology 2015, Climate Change in Australia Information for Australia’s Natural Resource Management Regions: Technical Report; CSIRO and Bureau of Meteorology: Melbourne, VIC, Australia, 2015. [Google Scholar]
  51. Iturbide, M.; Gutiérrez, J.M.; Alves, L.M.; Bedia, J.; Cerezo-Mota, R.; Cimadevilla, E.; Cofiño, A.S.; Di Luca, A.; Faria, S.H.; Gorodetskaya, I.V.; et al. An update of IPCC climate reference regions for subcontinental analysis of climate model data: Definition and aggregated datasets. Earth Syst. Sci. Data 2020, 12, 2959–2970. [Google Scholar] [CrossRef]
  52. Di Virgilio, G.; Ji, F.; Tam, E.; Nishant, N.; Evans, J.P.; Thomas, C.; Riley, M.L.; Beyer, K.; Grose, M.R.; Narsey, S.; et al. Selecting CMIP6 GCMs for CORDEX Dynamical Downscaling: Model Performance, Independence, and Climate Change Signals. Earth’s Futur. 2022, 10. [Google Scholar] [CrossRef]
  53. Xu, W.; Lei, X.; Chen, S.; Yu, T.; Hu, Z.; Zhang, M.; Jiang, L.; Bao, R.; Guan, X.; Ma, M.; et al. How Well Does the ERA5 Reanalysis Capture the Extreme Climate Events Over China? Part II: Extreme Temperature. Front. Environ. Sci. 2022, 10. [Google Scholar] [CrossRef]
  54. Velikou, K.; Lazoglou, G.; Tolika, K.; Anagnostopoulou, C. Reliability of the ERA5 in Replicating Mean and Extreme Temperatures across Europe. Water 2022, 14, 543. [Google Scholar] [CrossRef]
Atmosphere 14 00913 g001 550
Figure 1. Annual mean biases in ERA5 compared to AGCD for mean Tmax and Tmin. (a) Mean Tmax from AGCD; (b) mean Tmax from ERA5; and (c) difference in mean Tmax between ERA5 and AGCD. (d–f) Same as (a–c) but for Tmin. Notice the different colorbar limits for Tmax and Tmin.
Figure 1. Annual mean biases in ERA5 compared to AGCD for mean Tmax and Tmin. (a) Mean Tmax from AGCD; (b) mean Tmax from ERA5; and (c) difference in mean Tmax between ERA5 and AGCD. (d–f) Same as (a–c) but for Tmin. Notice the different colorbar limits for Tmax and Tmin.
Atmosphere 14 00913 g001
Atmosphere 14 00913 g002 550
Figure 2. Correlation between annual means of ERA5 and AGCD for (a) Tmax and (b) Tmin.
Figure 2. Correlation between annual means of ERA5 and AGCD for (a) Tmax and (b) Tmin.
Atmosphere 14 00913 g002
Atmosphere 14 00913 g003 550
Figure 3. Coefficient of variation for annual mean Tmax and Tmin in ERA5 compared to AGCD. Coefficient of variation of annual mean Tmax values from (a) AGCD, (b) ERA5, and (c) difference between ERA5 and AGCD (d–f); same as (a–c) but for Tmin. Notice the different colorbar limits for Tmax and Tmin.
Figure 3. Coefficient of variation for annual mean Tmax and Tmin in ERA5 compared to AGCD. Coefficient of variation of annual mean Tmax values from (a) AGCD, (b) ERA5, and (c) difference between ERA5 and AGCD (d–f); same as (a–c) but for Tmin. Notice the different colorbar limits for Tmax and Tmin.
Atmosphere 14 00913 g003
Atmosphere 14 00913 g004 550
Figure 4. Decadal trends for Tmax in the top row and Tmin in the bottom row from AGCD (a,d), ERA5 (b,e), and the difference between ERA5 and AGCD (c,f).
Figure 4. Decadal trends for Tmax in the top row and Tmin in the bottom row from AGCD (a,d), ERA5 (b,e), and the difference between ERA5 and AGCD (c,f).
Atmosphere 14 00913 g004
Atmosphere 14 00913 g005 550
Figure 5. Inter-annual and monthly variability for the National Resource Management regions. (a) Inter-annual variability of Tmax; (b) inter-annual variability of Tmin; (c) monthly variability of Tmax; and (d) monthly variability of Tmin. Values from AGCD in solid lines and those from ERA5 in dashed lines for Southern Australia (SA—blue), Eastern Australia (EA—red), Rangeland (R—grey), and Northern Australia (NA—black).
Figure 5. Inter-annual and monthly variability for the National Resource Management regions. (a) Inter-annual variability of Tmax; (b) inter-annual variability of Tmin; (c) monthly variability of Tmax; and (d) monthly variability of Tmin. Values from AGCD in solid lines and those from ERA5 in dashed lines for Southern Australia (SA—blue), Eastern Australia (EA—red), Rangeland (R—grey), and Northern Australia (NA—black).
Atmosphere 14 00913 g005
Atmosphere 14 00913 g006 550
Figure 6. Annual mean biases in six temperature-based climate extreme indices. Mean values from AGCD in the top row, ERA5 in the second row, and biases in the third row. Different columns for different indices: (a,g,m) Cold Spell Duration Indicator (CSDI, in days); (b,h,n) Warm Spell Duration Indicator (WSDI, in days); (c,i,o) Number of days when the minimum temperature is below 2 °C (TNlt2, in days); (d,j,p) Number of warm nights (TN90p, in days); (e,k,q) Number of warm days (Tx90p, in days); (f,l,r) Coldest daily minimum temperature (TNn, in °C).
Figure 6. Annual mean biases in six temperature-based climate extreme indices. Mean values from AGCD in the top row, ERA5 in the second row, and biases in the third row. Different columns for different indices: (a,g,m) Cold Spell Duration Indicator (CSDI, in days); (b,h,n) Warm Spell Duration Indicator (WSDI, in days); (c,i,o) Number of days when the minimum temperature is below 2 °C (TNlt2, in days); (d,j,p) Number of warm nights (TN90p, in days); (e,k,q) Number of warm days (Tx90p, in days); (f,l,r) Coldest daily minimum temperature (TNn, in °C).
Atmosphere 14 00913 g006
Atmosphere 14 00913 g007 550
Figure 7. Temporal correlation between ERA5 and AGCD of (a) Cold Spell Duration Indicator (CSDI), (b) Warm Spell Duration Indicator (WSDI), (c) Number of days when minimum temperature is below 2 °C (TNlt2), (d) Number of warm nights (TN90p), (e) Number of warm days (Tx90p), and (f) Coldest daily minimum temperature (TNn).
Figure 7. Temporal correlation between ERA5 and AGCD of (a) Cold Spell Duration Indicator (CSDI), (b) Warm Spell Duration Indicator (WSDI), (c) Number of days when minimum temperature is below 2 °C (TNlt2), (d) Number of warm nights (TN90p), (e) Number of warm days (Tx90p), and (f) Coldest daily minimum temperature (TNn).
Atmosphere 14 00913 g007
Atmosphere 14 00913 g008 550
Figure 8. Decadal trends of temperature extremes in AGCD (top row) and ERA5 (bottom row). (a,g) Cold Spell Duration Indicator (CSDI, in days); (b,h) Warm Spell Duration Indicator (WSDI, in days); (c,i) Number of days when minimum temperature is below 2 °C (TNlt2, in days); (d,j) Number of warm nights (TN90p, in days); (e,k) Number of warm days (Tx90p, in days); (f,l) Coldest daily minimum temperature (TNn, in °C).
Figure 8. Decadal trends of temperature extremes in AGCD (top row) and ERA5 (bottom row). (a,g) Cold Spell Duration Indicator (CSDI, in days); (b,h) Warm Spell Duration Indicator (WSDI, in days); (c,i) Number of days when minimum temperature is below 2 °C (TNlt2, in days); (d,j) Number of warm nights (TN90p, in days); (e,k) Number of warm days (Tx90p, in days); (f,l) Coldest daily minimum temperature (TNn, in °C).
Atmosphere 14 00913 g008
Atmosphere 14 00913 g009 550
Figure 9. Interannual variability over the four NRM regions for (a) Cold Spell Duration Indicator (CSDI), (b) Warm Spell Duration Indicator (WSDI), (c) Number of days when minimum temperature is below 2 °C (TNlt2), (d) Number of warm nights (TN90p), (e) Number of warm days (Tx90p), and (f) Coldest daily minimum temperature (TNn). Values from AGCD in solid lines and those from ERA5 in dashed lines for Southern Australia (SA—blue), Eastern Australia (EA—red), Rangeland (R—grey), and Northern Australia (NA—black).
Figure 9. Interannual variability over the four NRM regions for (a) Cold Spell Duration Indicator (CSDI), (b) Warm Spell Duration Indicator (WSDI), (c) Number of days when minimum temperature is below 2 °C (TNlt2), (d) Number of warm nights (TN90p), (e) Number of warm days (Tx90p), and (f) Coldest daily minimum temperature (TNn). Values from AGCD in solid lines and those from ERA5 in dashed lines for Southern Australia (SA—blue), Eastern Australia (EA—red), Rangeland (R—grey), and Northern Australia (NA—black).
Atmosphere 14 00913 g009
Table
Table 1. List of ET-SCI Indices assessed in this study. Note that the last column of Sectors Affected is adopted from https://climpact-sci.org/ (accessed on 30 March 2023) to suggest the multi-sector relevance of the selected indices.
Table 1. List of ET-SCI Indices assessed in this study. Note that the last column of Sectors Affected is adopted from https://climpact-sci.org/ (accessed on 30 March 2023) to suggest the multi-sector relevance of the selected indices.
NoIndexDefinitionUnitsTimescaleSectors Affected
1.CSDICold spell duration indicator (annual count of days with at least six or more consecutive days when minimum temperature < 10th percentile)daysAnnualHealth, agriculture and food security, coasts, disaster risk reduction, energy, fisheries, forestry/GHGs, and cryosphere
2. TNlt2Number of days when the minimum temperature is below 2 °CdaysMonthly/AnnualAgriculture and food security, forestry/GHGs, and cryosphere
3. TN90pNumber of warm nightsdaysAnnualEnergy
4. TX90pNumber of warm daysdaysAnnualEnergy
5.WSDIWarm spell duration indicator (annual number of days contributing to events where six or more consecutive days experience maximum temperature > 90th percentile)daysAnnualHealth, agriculture and food security, water resources and food security, coasts, disaster risk reduction, energy, fisheries, forestry/GHGs, and cryosphere
6.TNn Coldest daily minimum temperature°CAnnual/MonthlyAgriculture and food security, energy, forestry/GHGs, and cryosphere
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Choudhury, D.; Ji, F.; Nishant, N.; Di Virgilio, G. Evaluation of ERA5-Simulated Temperature and Its Extremes for Australia. Atmosphere 2023, 14, 913. https://doi.org/10.3390/atmos14060913

AMA Style

Choudhury D, Ji F, Nishant N, Di Virgilio G. Evaluation of ERA5-Simulated Temperature and Its Extremes for Australia. Atmosphere. 2023; 14(6):913. https://doi.org/10.3390/atmos14060913

Chicago/Turabian Style

Choudhury, Dipayan, Fei Ji, Nidhi Nishant, and Giovanni Di Virgilio. 2023. "Evaluation of ERA5-Simulated Temperature and Its Extremes for Australia" Atmosphere 14, no. 6: 913. https://doi.org/10.3390/atmos14060913

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop