Weather and Climate @ Reading

By: Prof. David Ferreira

Recently, a colleague pointed out that, in the Northern Hemisphere subtropics, the winds blow eastward around 30-40N (the jet stream), and westward around 20-10N (the trade winds), and these winds make the ocean spin in the clockwise direction. This is akin to thinking of the ocean as a solid plate rotating around an axis: if one pushes the plate in one direction on one side, and the opposite direction on the other side, the plate will spin.

Fig. 1: The observed Mean Ocean topography estimated from a combination of satellite measurements and ocean drifters from Maximenko et al. (2009). Data, figure, and further details can be found here https://apdrc.soest.hawaii.edu/projects/DOT/. The pink arrows represent the dominant winds, while the blue arrows show the directions of the ocean current averaged over the full depths of the ocean.

Fig. 1 shows the Dynamical Ocean Topography (DOT). The currents tend to flow along the contours of DOT. Focusing on the North Pacific, the figure highlights the most striking example of a subtropical gyre, with a clockwise motion between about 10 and 40N (but these gyres are found in all basins). Applying the solid plate reasoning to the southern hemisphere suggests that the gyres spin counterclockwise (which they do! See Fig. 1).  

A “correlation” analysis supports the solid plate explanation. All subtropical gyres spin in the direction imparted by the winds, and indeed on some level, this is a decent explanation of subtropical gyres.  However, the remark of my colleague provides us with an opportunity to emphasize why we bother with fluid dynamics, equations and idealized modelling. 

The solid plate reasoning hides a complex dynamic. When the wind blows over the ocean, the first effect is not to make it spin. The winds first create horizontal currents confined to the upper 50-100 m of the oceans. Somewhat unintuitively, these currents, called Ekman currents, are at right angle of the wind and to their right (in the Northern Hemisphere). The meeting of the Ekman currents around 25N creates a bulge of the ocean surface between the Jet stream and the trade winds, but also a descending motion w_e (water cannot accumulate in the surface layer forever). This is where we use a version of the Sverdrup balance: 

βV = f w_e

where f is the Coriolis parameter (Earth is spinning on itself), V is the north–south flow in the deep ocean and β represents the variations of the Coriolis parameter with latitude.

Fundamentally, the parameter β represents a geometrical effect: Earth is a sphere. If Earth was flat, β would be zero. A direction that particularly matters for the dynamics is that along the axis of rotation. The sphericity of Earth implies that, even in a flat bottom ocean, columns of water aligning with the axis of rotation are smaller near the pole than near the Equator (Fig. 2 a).

Fig. 2: Schematic showing the variations of the height of water column in the direction of the rotation axis. Case a) corresponds to a flat-bottom ocean where the depth of the ocean is the same everywhere in the direction of gravity. Cases b) and c) illustrate what happens when the bottom of the ocean is tilted in some region of the planet.

We can now interpret the Sverdrup balance (1). The wind pushes water downward into the ocean interior (w_e <0) making water columns taller. As a result, water columns must move southward where they can accommodate their new height. Indeed, we see in the real ocean (Fig. 1) that the southward motions of the ocean spread over the full width of the basin and is only broken in a very narrow region on the western side of the basin (in the Kuroshio current, the Gulf Stream, the Agulhas current etc.).

Obviously, we cannot experiment with the real ocean to test our understanding. However, we can do it in numerical models that solve more complex equations than Eq. (1) (where nonetheless we can afford some simplifications to make things easier).

This is illustrated in Fig. 3 using a square basin with a flat bottom. The ocean is assumed to be of uniform density. It is driven by eastward wind to the north of the basin and westward wind to the south, mimicking the real-world wind pattern. Fig. 3a shows that the circulation is southward throughout the basin, except in a very narrow current on the western boundary.

Fig. 3: Simulations with a numerical ocean model (the MITgcm, mitgcm.org). The pink arrows in the top left panel show the direction of the winds. The blue arrows indicate the direction of the currents. The upper row has a flat bottom ocean with (a) and without (b) beta effect. In the bottom row, the sea floor is tilted, shallower in the northern part of the basin (corresponds to Fig. 2c)) and shallower in the southern part of the basin (corresponds to Fig. 2b)

What happens if we set β to zero, i.e. make a flat Earth? This is shown Fig. 3b (same winds as in Fig. 3a): the circulation becomes circular around the center of the basin. The flow is southward over half of the basin, and northward over the other half, nothing like the real ocean.

The geometrical interpretation of β suggests that the shape of the ocean bottom could play a role here. Imagine that, in some region of the ocean, the sea floor is getting deeper as one moves northward. Thus, seen from the perspective on the sphere, water columns in that region have similar heights (Fig. 2b). The tilted bottom floor makes Earth “flatter” in this region. This effect is called the topographic β-effect. In this case, it partially cancels effect. For a given wind, columns need to travel southward faster than before to accommodate for the pilling up of water by the wind. So, the gyre must be stronger.

Conversely, if the ocean floor is getting shallower moving northward in some region (Fig. 2c), the height of water columns varies much more, the Earth is more “spherical”. The topographic β-effect reinforces the normal β-effect, and we predict that the gyre should be weaker than for a flat bottom ocean.

Again, we can test our predictions through numerical experiments. Fig. 3c) and d) show the circulation in our simplified geometry, but now with tilted bottoms. As predicted, the gyre in Fig. 3c (corresponding to the tilt in Fig. 2b) is stronger than with a flat bottom while in Fig. 3d (corresponding to the tilt in Fig. 2c) it is weaker than with a flat bottom.

The β-effect is therefore critical to explain the characteristic pattern of the subtropical gyres. Something we could not get from the “correlation” analysis. One could do much more with the theory. It is possible to make more analytical advances, which reveals that β also controls the width of the narrow western boundary current (the smaller β, the wider the boundary current). Incidentally, the analytical solution can be used to test numerical models (and be happy that discretization, used in climate predictions, can be trusted). Retrospectively, it appears that the correlation analysis of wind patterns and gyres did not tell us much that was interesting.

The gyre dynamics discussed above was done decades ago and is pretty much textbook material nowadays, but the need to develop theory for new research remains. Recently, we explored the connection between Arctic sea ice decline and Ocean heat transport under climate change (Aylmer et al. 2024). Climates models that predict larger sea ice decreases in the future also have larger increases in ocean heat transport into the Arctic. This sounds intuitively OK, but is it what happens? Is it even quantitatively possible? That is, is the additional heat to the Arctic ocean large enough to explain the additional melting in these models? Less intuitive, models that predict more heat to the Arctic by the atmosphere tend to show relatively less sea ice retreat. These correlations have been noted for decades, but their interpretation has remained speculative. Aylmer et al. (2024) provides a theoretical framework that predicts, quantitatively, the behavior of the climate models, providing firm ground for interpretation.

Interestingly, this work started in 2019 with the development of an analytical model of climate by Dr Jake Aylmer during the first year of his PhD (published in Aylmer et al., 2020).

Yes, it takes time. As our field turns more and more toward statistical analysis, big data, and is pushed toward more applied, impactful research by funding agencies (with cascading effect on doctoral training programs), there is less and less room for this type of work.

As I am polishing this blog, I am listening to a seminar by David Karoly discussing Hoskins and Karoly (1981), a paper which had a massive impact in the field of meteorology and climate dynamics. This paper has been cited more than 3000 times and is still cited 100 times per year 40 years after publication! David Karoly pointed out that this work had major limitations and should be critically evaluated before being cited. Hoskins and Karoly (1981) is grounded in mathematics, geophysical fluid dynamics, and idealized experiments. One needs to be trained in such things to critically assess and carry on fundamental research in climate science. Unfortunately, the number of PhD students trained with these skills is gradually decreasing.

Correlation analysis is cool, but a bit of theory on top is cooler.

References:

Aylmer, J., D. Ferreira, and D. Feltham, 2020: Impacts of oceanic and atmospheric heat transports on sea-ice extent. J. Climate, 33, 7197–7215. https://doi.org/10.1175/JCLI-D-19-0761.1 

Aylmer, J., D. Ferreira, and D. Feltham, 2024: Ocean heat transport regulates Arctic sea ice loss. In press in Nature Communications.  https://doi.org/10.21203/rs.3.rs-3249087/v1 

Hoskins, B. J., and D. J. Karoly, 1981: The Steady Linear Response of a Spherical Atmosphere to Thermal and Orographic Forcing. J. Atmos. Sci., 38, 1179–1196, https://doi.org/10.1175/1520-0469(1981)038<1179:TSLROA>2.0.CO;2. 

Maximenko, N., P. Niiler, M.-H. Rio, O. Melnichenko, L. Centurioni, D. Chambers, V. Zlotnicki, and B. Galperin, 2009: Mean dynamic topography of the ocean derived from satellite and drifting buoy data using three different techniques. J. Atmos. Oceanic Tech., 26 (9), 1910-1919. DOI: https://doi.org/10.1175/2009JTECHO672.1 

By: Prof. Keith Haines and Samantha Petch

Atmospheric CO2 levels are rising every year, primarily due to human activities, Fig 1. However, the rate of this increase does vary significantly from year to year, not because of human emissions, but mostly due to changes in how much CO2 is absorbed by land ecosystems. 

Fig 1. Recent changes in Atmospheric CO2 concentrations. Interannual changes can just about be seen.

The land’s ability to absorb CO2 is determined by the balance between photosynthesis and respiration and is the least understood part of the global carbon cycle. This land carbon sink plays a crucial role in offsetting anthropogenic CO2 emissions, accounting for approximately 30 ±10% of these emissions each year (Friedlingstein et al., 2020). The controls on this land surface carbon sink are still poorly understood and would have a strong impact on future CO2 growth rates (CGR) if they were to change significantly.  There are notable increases in CGR during El Niño events and decreases during La Niña events (Keeling & Revelle, 1985) and there is also a widespread consensus that variations in tropical ecosystems exert the most significant influence on global CGR.  

Tropical temperatures have a strong positive correlation with CGR (Wang et al., 2013) and given that tropical temperatures are usually considered optimal for photosynthesis, any elevation in these temperatures under global warming could amplify the CGR. However it has recently been realised that water availability is also playing a really important role, see Fig 2.   

Figure 2 shows detrended interannual variations in the atmospheric CGR in black, and the total terrestrial water storage (TWS on a reversed axis) in blue, as measured from the GRACE satellite which detects water through mass redistributions over the Earth’s surface, (from Petch et al. 2024 following Humphreys et al. 2018).

Less water over land e.g. during the El Niño in 2015/6, goes with more rapid increases in atmospheric CO2 as the land surface absorption is reduced.  This range of uptake (up to 4GtC/yr) is really large and to understand this relationship better we really need to regionalise the surface response, see Fig 3.   

Figure 3 shows the regional interannual correlations in GRACE surface water storage and the global CGR (Petch et al. 2024).

Petch et al. (2024) found that the tropics alone can explain the entire global TWS-CGR relationship seen in Fig 2, with only minor offsetting contributions from the northern and southern hemisphere extratropics. Additionally, tropical forests, e.g. over the Amazon, were found to be the dominant land cover type governing this relationship, despite occupying only a relatively small portion of the land surface. While the negative correlations in Fig 3 (and other evidence) point to the importance of the tropics in the global relation seen in Fig 2, it would be important to have direct regional surface CO2 flux data to back this up.   

Having a verifiable way of comparing surface CO2 fluxes is of course of the highest scientific and political importance. Currently national anthropogenic CO2 emissions all rely on national inventories which are hard to verify or compare. CO2 in the atmosphere gets mixed relatively quickly, at least at the hemispheric level, and so only frequent reliable satellite based atmospheric measurements stand a chance of achieving this, e.g. ESA’s CO2M-(A,B,C) multi-satellite mission, launching 2025/6. In the meantime, less frequent satellite measurements e.g. GOSAT (2009 onwards) can be assimilated into atmospheric transport models which may then be used to infer surface sources and sinks which best match observed atmospheric CO2 concentrations. Given the importance of these estimates many teams produce such surface CO2 flux products, although many still rely on in situ measurements alone and show a lot of disagreement between regional results (Fig 4).

Figure 4 shows % contributions to global land carbon flux interannual variability over 2001-2023 from four atmospheric CO2 inverse model products. Note, the NISMON-CO2 map is depicted over a larger range due to higher contributions from tropical forests. While these products agree well with the global CGR from Fig 2, the regional agreements can be seen to be fairly poor, Petch et al. (2024).

We believe there is potential to bring in additional satellite-based information from the land surface such as the GRACE water data shown earlier, and perhaps other datasets such as ESA’s Climate Change Initiative Land surface temperature measurements (https://climate.esa.int/en/projects/land-surface-temperature/), to provide additional constraints to these atmospheric CO2 transport models and bring greater confidence to space-based estimates of regional CO2 fluxes.  This would help both improve understanding of ecosystem responses to climatic changes as well as provide better verification tools for monitoring anthropogenic emissions in a more verifiable way.  

References 

Friedlingstein, P., O’Sullivan, M., Jones, M. W., Andrew, R. M., Hauck, J., Olsen,  A., . . . Zaehle, S. (2020). Global carbon budget 2020. Earth Syst. Sci. Data,  12 , 3269–3340. https://doi.org/10.5194/essd-12-3269-2020 

Humphrey, V., Zscheischler, J., Ciais, P., Gudmundsson, L., Sitch, S., & Seneviratne, S. I. (2018). Sensitivity of atmospheric co2 growth rate to observed changes in terrestrial water storage. Nature, 560 , 628–631. https://doi.org/10.1038/s41586-018-0424-4 

Keeling, C. D., & Revelle, R. (1985). Effects of el nino/southern oscillation on the atmospheric content of carbon dioxide. Meteoritics, 20 , 437-450. https://doi.org/10.1093/nsr/nwab150 

Petch, S., Feng, L., Palmer, P. King, R. P., Quaife, T. & Haines, K.  (2024) Strong relation between atmospheric CO2 growth rate and terrestrial water storage in tropical forests on interannual timescales. Submitted to Glob. Biogeochem. Cyc. Also ESS Open Archive . https://doi.org/10.22541/essoar.172132358.83519012/v1  

Wang, W., Ciais, P., Nemani, R. R., Canadell, J. G., Piao, S., Sitch, S., . . ., Myneni, R. B. (2013). Variations in atmospheric CO2 growth rates coupled  with tropical temperature. Proc. Natl Acad. Sci., 110 , 13061–13066. https://doi.org/10.1073/pnas.1219683110 

By: Prof. Joy Singarayer

It is broadly acknowledged that colonialism and imperialism hold much responsibility in the causes of the climate crisis. The aggressive and unsustainable extraction of resources from colonized regions laid the ground for today’s economic reliance on fossil fuels and general overconsumption. It is also widely recognised that the industrialized nations responsible for the majority of historical carbon emissions, benefited most from colonial exploitation, while the Global South now bears the brunt of climate impacts (Evans et al., 2023). Further, the power imbalances and political structures that emerged from colonialism continue to sideline the most vulnerable nations in current climate policies.  

Exploring the impacts of Empire is something that is interesting to me from a personal point of view, as I can directly reach out and touch that history through my father, who grew up under colonial rule in Sri Lanka (then Ceylon1). I’ve also had many and varied conversations with colleagues in relation to ‘decolonising the curriculum’ or inequalities in climate change impacts, but far fewer about colonialism and research itself. This gap is echoed in the IPCC assessment reports. In the sixth set of reports, Working Group II (Impacts, Vulnerability, and Adaptation) outlines how colonial histories have shaped current vulnerabilities and gives attention to indigenous knowledge in adaptation strategies. Working Group III (Mitigation) acknowledges inequalities in responsibility and the need to consider equity in mitigation. However, the Working Group I report (The Physical Science Basis) does not engage with these themes apart from tangentially in reference to differences in regional data availability.  

Two colleagues, climate dynamicist Ted Shepherd and historian Rohan Deb Roy, and I are seeking to address questions of the links between colonialism and climate science. How does colonialism, past and present, shape contemporary practices in climate science? What are the ways in which scientific institutions both perpetuate and challenge these practices? To that end we started by convening historians of science, geographers, and climate scientists, at a symposium in May this year to promote an exchange of knowledge centred on critically examining the methodologies and institutional frameworks within climate science.  

Several recurring themes became evident during the symposium, though the breadth of topics discussed far exceeded what can be covered here. One significant theme concerns who is included and recognized for their contributions to climate science. The dominant narrative in the literature, including in the latest IPCC reports, still often centres on individual scientists (mostly white, male) advancing the field with solitary efforts. However, this narrative neglects the invisible labour performed by enslaved or indentured labourers, and indigenous peoples. These individuals frequently possessed skills and knowledge that European scientists relied on, all while their knowledge systems and contributions were systematically erased. The incorporation of indigenous knowledge, which is often portrayed as a recent development, has a long history intertwined with the evolution of climate science (Mercer and Simpson, 2023). For example, during his tenure in India, Sir Gilbert Walker relied heavily on the labour and knowledge of Indian clerks, assistants, and local observers. Yet, local knowledge was rarely acknowledged in publications or recognized in the careers of those who contributed to it.  

Similar issues persist today with practices such as ‘helicopter research’, where scientists from the privileged and powerful countries undertake research in low-income countries with little involvement from local researchers and communities in the research or publication thereof. This contributes to underrepresentation of authors from the Global South in scientific papers. For example, of the top 100 climate science papers over five years up to 2021, less than 1% of authors were based in Africa (Tandon, 2021; Figure 1).   

Fig. 1 The percentage of authors from the Top 100 most-cited climate science papers during 2016-20, from each continent. Modified from Carbon Brief (Tandon, 2021).

The lack of access to data, models, and exorbitant journal publication fees further exacerbates these challenges, perpetuating the exclusion of Global South scientists from key areas like mitigation scenario modelling for the IPCC WGI report. This exclusion has led to future scenarios derived using Integrated Assessment Models where the Global South bears the burden of land conversion for emissions mitigation, while the Global North is not asked to make comparable sacrifices in lifestyle (Ketchum, 2021). This dynamic reflects the enduring colonial power structures within climate science. Indeed, colonial regions have historically served as testing grounds for engineering climate “improvements” (Mahoney and Endfield, 2018), a pattern that continues today with climate geoengineering projects across Africa, often initiated and funded by the US and Europe. 

Another critical theme discussed is the subjectivity inherent in the collection of observational data. Colonial powers historically gathered meteorological data that served their specific interests such as ensuring safe and fast shipping routes and later air routes, and for agricultural needs in the colonies. In this sense meteorological measurement developed very much as part of the tools of Empire. As a result, the length and consistency of climate and weather records vary greatly by region, leading to significant implications for the present. This disparity affects the ability to attribute current climate changes and extreme events, contributing to uncertainty and, consequently, potentially influencing access to funding for damage or adaptation. This represents a continuing injustice, as the historical biases in data collection can directly impact modern decisions and resources. Moreover, the areas of focus in climate science and the questions that researchers pursue are likely influenced by these historical data biases. It is important to reflect critically on the distribution of new data and the spatial scales we consider, particularly regarding where, why, and for whose benefit.  

Reflecting on the symposium discussions, in current climate science there are clear parallels with, and legacies of, colonial approaches. As I imagine is common especially in the UK, my experience of higher education contained little exposure to the history of the subject. With the small amount I have learned over the last couple of months, I feel compelled to take time to absorb more of the history of our discipline, to reflect on my own research practice, engage in more interdisciplinary conversations with colleagues and students, and organise future events. For anyone else interested in reading work of much more knowledgeable scholars, I have included a short reading list below. 

1 Ceylon gained independence in 1948, when my father was already in his twenties.  

Reading: 

Bhambra, G. K., and Newell, P., 2023. More than a metaphor: ‘climate colonialism’ in perspective. Global Social Challenges Journal, 2(2), 179-187. Retrieved May 27, 2024, from https://doi.org/10.1332/EIEM6688 

Coen, D.R., 2018. Climate in Motion: Science, Empire, and the problem of scale. The University of Chicago Press. 

Evans S. et al., 2023. Revealed: how colonial rule radically shifts historical responsibility for climate change. At https://www.carbonbrief.org/revealed-how-colonial-rule-radically-shifts-historical-responsibility-for-climate-change/ (Accessed 1/9/2024) 

Ketchum, C., 2022. How scientists from the ‘Global South’ are sidelined at the IPCC. At: https://theintercept.com/2022/11/17/climate-un-ipcc-inequality/ (Accessed 27/5/2024). 

Mahony, M., 2021, Meteorology and Empire. The Routledge Handbook of Science and Empire. Goss, A. (ed.). Routledge, p. 47-58 12 p. 

Mahoney, M., and Endfield, G, 2018. Climate and Colonialism. Wiley Interdisciplinary Reviews: Climate Change, 9:e510, https://doi.org/10.1002/wcc.510 

Mercer, H. and Simpson T., 2023. Imperialism, colonialism, and climate science. Wiley Interdisciplinary Reviews: Climate Change, 14:e851, https://doi.org/10.1002/wcc.851  

Rodrigues, R.R. and Shepherd, T.G., 2022. Small is beautiful: climate-change science as if people mattered, PNAS Nexus, 1, pgac009, https://doi.org/10.1093/pnasnexus/pgac009 

Tchekwie Deranger, E., Sinclair, R., Gray, B., McGregor, D., Gobby, J., 2022. Decolonizing Climate Research and Policy: making space to tell our own stories, in our own ways, Community Development Journal, 57(1), Pages 52-73, https://doi.org/10.1093/cdj/bsab050  

Tandon, A., 2021. Analysis: the lack of diversity in climate-science research. At: https://www.carbonbrief.org/analysis-the-lack-of-diversity-in-climate-science-research/ (Accessed 27/5/2024). 

by: Dr. James Weber

Air quality (AQ), and the policies enacted to improve it, is becoming an increasingly important issue. It is also becoming increasingly politicised; exemplified by arguments over clean air zones like London’s ULEZ, low traffic neighbourhoods and even moorland burning (Weber et al. 2023). The negative impact of poor AQ on health, particularly for the most vulnerable people, is well established yet understanding the drivers of air quality and, thus the ways it can be improved, is challenging.  

AQ is generally defined in terms of the concentrations of key pollutants which negatively impact human health, for example nitrogen dioxide (NO2), ozone (O3) and fine particulate matter (PM2.5). Their concentrations are determined by the balance of pollutant sources (local emissions, longer range transport and production in the atmosphere) and sinks (loss to terrestrial or aqueous surfaces and dispersion in the atmosphere). Thus, understanding AQ requires knowledge of meteorology, atmospheric chemistry and aerosol science. While an extreme example, the London Smog of 1952 which resulted in at least 10,000 deaths, cannot be understood without considering both the anti-cyclonic behaviour which led to a temperature inversion, trapping air close to the ground, and the chemistry which converted the sulphur dioxide emitted by the burning of low quality coal into sulphuric acid. 

The dependence on prevailing meteorology can make evaluation of interventions, such as clean air zones, challenging, particularly in the short term because it can be hard to determine the extent to which any AQ change (or lack thereof) following an intervention is due to changes in local emissions (which the intervention can influence) and/or the extent to which it is due to prevailing meteorology and longer-range transport of pollution.  

The assessment of clean air zones is beyond the scope of a single blog post but here I present a summary of how some key, widely measured pollutants have changed in Reading over the last 10 years, explore how simple analysis can point to their source(s) and demonstrate how machine learning can be used to assess the influence of different variables on their concentrations. I focus on Reading, but this analysis could be done with data from any of the ~270 air quality monitoring sites maintained by the Department of the Environment, Food and Rural Affairs (DEFRA) around the UK as well as those maintained by local authorities and the ever-growing number of air quality monitoring sites around the world.  

Comparison to AQ Targets 

I use the DEFRA air quality monitoring site located in Reading New Town, an urban background site (i.e. not next to a busy road) in this case and make use of the excellent Openair R package from David Carslaw and colleagues at the University of York. It is important to recognise that the concentrations of shorter-lived pollutants can vary across an urban region due to varying proximity to local sources and the (lack of) mixing due to orography, for example the street canyon effect. Therefore, to construct a complete picture, multiple sites across a city or town should be considered.  

The most recent WHO air quality targets recommend that daily mean NO2 and PM2.5 concentrations should not exceed 25 μg m-3 and 15 μg m-3 respectively for more than 3-4 days per year (~1%)viii. Plots of daily mean NO2 and PM2.5 for 2014-2024 (Fig 1) demonstrate just how frequently these pollutants have exceeded this limit over the last 10 years (30% and 10% of the time respectively). Of course, care must be taken when applying a limit proposed in 2021 to years prior but it is nevertheless informative.  

Daily mean NO2 and PM2.5 concentrations with 2021 WHO target (black line) and 2005 WHO target (dashed black line, PM2.5 only).

Longer Term Trends 

Analysing the longer term trend however paints a slightly different picture. NO2 exhibits a consistent decrease (p<0.001) over 2014-2024 (Fig 2a) while PM2.5 shows a much slower decline (p<0.05) (Fig 2b). The reduction in NO2 is likely driven by improvements to the vehicle fleet given traffic’s role as a major NO2 source (see later).  Early 2020 does appear to exhibit relatively low NO2 which could be attributed solely to the COVID19 lockdown’s reduced traffic flow. However, this lower NO2 starts before lockdown began and is likely due, at least initially, to the meteorological conditions . January, February and early March 2020 experienced more westerlies (i.e. air originating from less polluted regions) and higher wind speeds than late March and April 2020 which also had more easterlies (air coming from more polluted areas). Therefore, pre-lockdown meteorological conditions were conducive to relatively low NO2 while after the start of lockdown, meteorological conditions favoured higher NO2, suggesting the observed low values were due more to emission reductions. This one example highlights the complexity of understanding the drivers of air quality. 

The rate of PM2.5 decrease is around an order of magnitude lower than that of NO2 which is due in part to traffic contributing a smaller fraction to PM2.5 emissions than NO2.  

While NO2 and PM2.5 show decreases, O3 exhibits a steady increase (p<0.001) (Fig 2c). This increase is in part due to the reduction in NOx(=NO + NO2); in particular, the increase in O3 levels in early 2020 coincided with the Covid lockdown. This highlights a key challenge in AQ policy; under certain chemical environments, reducing NOx will increase O3 (Grange et al., 2021).  

Monthly mean concentrations and trend for (a) NO2, (b) PM2.5 and (c) O3 from the Reading New Tow sensor. Numbers in green show slope and values at 95% confidence.

These trends are also seen in the exceedances. NO2 drops from ~50% exceedance in 2014-2017 to <10% in 2023 (with a COVID19 dip also visible) while PM2.5 shows slower improvement (Fig 3).  

Percentage of years each year when daily mean NO2 and PM2.5 exceeded the WHO 2021 limit for Reading New Town.

The Detective Work Begins  

Of course, from a policy perspective, the key aim is to understand the relative importance of difference sources of air pollution so that measures – local, regional, national, or even international – can be designed to improve the situation.   

The diurnal and weekly cycles of NO2 and PM2.5 (Fig 4) provide some clues as to their sources and so how they might be affected by policies. During the week, NO2 shows strong peaks in the morning and evening rush hour, supporting the dominant role of traffic. In contrast, PM2.5’s morning peak is much smaller, and the evening peak occurs slightly later, suggesting the reduction in boundary layer height and greater confinement of pollution closer to the surface is more important for PM2.5 than NO2 (local emissions from domestic heating may also play a small role in the winter). Over the course of a week, NO2 also exhibits a much greater reduction at the weekend as the flow of commercial vehicles, a major source of NO2, is greatly reduced; however, PM2.5 exhibits little weekday-weekend variation, further supporting the argument that a greater fraction of PM2.5 comes from non-traffic sources.  

Diurnal, weekly and annual variation in normalised NO2 and PM2.5 concentrations.

Local Emissions vs. Longer Range Transport  

If we combine the AQ data with meteorological data from the University of Reading’s atmospheric observatory, the influence of wind speed and wind direction on pollutant levels can be examined. The polar plots in Figure 5 show normalised NO2 (left) and PM2.5 (right) concentrations as a function of wind speed and direction. NO2 concentrations are highest at very low wind speeds (i.e. centre of plot) while stronger winds from East-North-East (ENE, ~the Greater London region) are also associated with higher NO2. Higher wind speeds from other directions are associated with low NO2, suggesting the dispersion of local emissions is outweighing any longer range transport. The story is quite different for PM2.5: low wind speeds do yield higher than average concentrations but, by far the highest pollution arises when there are strong winds from ENE. This presents strong evidence that NO2 is primarily governed by local emissions, with a smaller contribution from longer range transport, while PM2.5 is driven much more by transport of pollution. Therefore, policies to reduce local emissions (e.g. a Reading clean air zone) are more likely to improve NO2 (which is already decreasing steadily anyway) than PM2.5. Of course, the situation may be very different in another town or city; just because it appears (at least from preliminary analysis) that a clean air zone in Reading would have little impact on PM2.5, this does not mean such a policy would be ineffective for PM2.5 everywhere.  

Polar plots of NO2 and PM2.5

The Rise of the Machine (Learning) 

As in many fields, machine learning can be used *with great care* to understand the variability of a particular parameter (e.g. NO2) to a range of explanatory variables. In this case, I use the deweather package developed by David Carslaw and colleagues to construct a statistical model of pollutant concentration from contemporaneous meteorological and temporal data (e.g. wind speed, hour of the day).  

After building the model, I explore the dependency of a pollutant to each variable in turn – the so-called partial dependency – by sampling the model with many different values of that variable while holding all other variables at their mean value. This provides information as to how each variable in isolation can affect air quality and the magnitude of its influence (Fig 6). Of course, this requires the user to have included all the important factors and, in this case, we have used a basic set of explanatory variables, omitting more complex ones such as air mass origin which can be useful for tracking long range transport of pollution.  

The plots for PM2.5 and NO2 (for 2021-2024) show that both pollutants are predicted to decrease with higher wind speeds (U10) but note the slight increase at high speed for PM2.5, most likely driven by the behaviour seen in Fig 5. Increasing temperatures (Td) are also associated with lower pollution, possibly due to a higher boundary layer and therefore greater mixing of pollutants away from the surface, although the cause of the uptick in concentration from PM2.5 remains unclear.  

Higher levels of pollution are also associated with winds from the east (i.e. those which have passed over Greater London) but this is more influential for PM2.5 than NO2 (in agreement with Fig 5). The diurnal cycles seen in the observational data are broadly reproduced but this factor is more important for NO2 than PM2.5, reflecting the greater role of traffic in NO2 production. The model returns a similar weekday-weekend pattern as observed in Figure 4.  

For both pollutants, the trend is the single most influential component. This can be thought of as the variability not captured by the other explanatory variables and will include the impact of longer term emission changes but could also include the impact of factors not included in the explanatory variables, such as varying air mass origin in this case, and so should be interpreted with care.  

An obvious use of this statistical model is to predict pollutant concentrations under counterfactual situations. For example, if a clean air zone is implemented, such a model can be used to predict the concentration of pollutants which would have occurred had no such policy been put in place (Grange et al., 2021). The difference between the modelled counterfactual concentration and measured concentration at a given time is the true impact of the policy and a better metric than the oft-used approach of comparing air quality values at an (often arbitrary) time before and after a policy’s implementation. These models are thus powerful tools for analysing the impact for policies but great care must be taken to ensure model biases – which are inevitable – are not conflated with a policy’s impact. 

Partial dependencies for PM2.5 (left) and NO2 (right). U10 is 10 m wind speed, Dirn10 10 m wind direction and Td dry bulb temperature.

References and Further Reading

For analysing air quality data, the Openair package offers a wide range of analysis and data visualisation tools. Grange et al (2018) presents the use of deweather in analysing PM while Grange et al (2021) demonstrates predicting the counterfactual to analyse the impact of changing emissions.  The R scripts used to generate the plots are available on request from the author.  

Weber, James; Val Martin, Maria; Bryant, Robert (2023). Impact of Moorland Fires on Sheffield Air Quality on 9th October 2023. The University of Sheffield. Report. https://doi.org/10.15131/shef.data.24356629.v1  

Stuart K. Grange, James D. Lee, Will S. Drysdale, Alastair C. Lewis, Christoph Hueglin, Lukas Emmenegger, and David C. Carslaw, (2021). COVID-19 lockdowns highlight a risk of increasing ozone pollution in European urban areas. Atmospheric Chemistry and Physics. https://doi.org/10.5194/acp-21-4169-2021

Stuart K. Grange, David C. Carslaw, Alastair C. Lewis, Eirini Boleti, and Christoph Hueglin, (2018). Random forest meteorological normalisation models for Swiss PM10 trend analysis. Atmospheric Chemistry and Physics. https://doi.org/10.5194/acp-18-6223-2018

Dacre, H.F., Mortimer, A.H. and Neal, L.S., 2020. How have surface NO2 concentrations changed as a result of the UK’s COVID-19 travel restrictions?. Environmental Research Letters, 15(10), p.104089. https://doi.org/10.1088/1748-9326/abb6a2

By: Dr. Thomas Wilder

The term “numerical grey zone” might seem abstract to many, but for those involved in atmospheric and oceanic modeling, it represents a challenging predicament. The numerical grey zone describes any numerical model that can resolve processes in some regions but not others, e.g. mesoscale eddies in low and high latitudes, respectively. Mesoscale eddies are energetic rotating currents with length scales of 10 – 100 km that populate the global ocean. Similar difficulties surrounding the grey zone arise in convective permitting atmospheric models.   

Eddy-permitting ocean models suffer from this numerical grey zone, making it inextricably difficult to accurately represent eddies. For example, the Met Office Hadley Centre global coupled medium resolution climate model used in CMIP6 performed poorly in the Southern Ocean. In particular, the current flowing through Drake passage, the Antarctic Circumpolar Current, was took weak, owing to a warm temperature bias (Kuhlbrodt et al., 2018). The cause of these issues was reasoned to be because of the poor representation of mesoscale processes in the NEMO ocean model component. Eddy-permitting ocean models have the potential to resolve more important processes while not being too computationally expensive to run. Ideally, higher-resolution models would be used to navigate this problem. Unfortunately, we are constrained by computational limits.    

The Southern Ocean is a major region of water mass transformations, where deep waters upwell and interact with the atmosphere and cryosphere. The Southern Ocean is also home to the Antarctic Circumpolar Current that hosts a vigorous eddy field. In this region, eddies transport heat polewards, acting as a potential mechanism for the cross-shelf transport of warm waters onto ice shelves. Moreover, eddies also contribute to the ventilation of surface and interior ocean waters, influencing the air-sea exchange of properties like heat and carbon, and even affecting cloud properties and rainfall. The Southern Ocean’s importance is clear, with it being vital to accurately represent mesoscale eddy processes.   

As part of the project “Earth System Models for the Future 2025” our task was to try and improve Southern Ocean circulation in the eddy-permitting NEMO model by implementing a new eddy parameterisation. A parameterisation is an equation that approximates the effect of processes that take place below the model’s grid resolution. Here, we are interested in the Leith viscosity parameterisations, which more faithfully represent mesoscale turbulence compared with other more common closures (Bachman et al., 2017). There are two Leith schemes: 2D Leith is proportional to relative vorticity; QG Leith is proportional to quasi-geostrophic vorticity. A benefit of the Leith closures is their utility in being used as the Gent-McWilliams (GM) diffusivity coefficient. The GM parameterisation works by mimicking the eddy transport of oceanic tracers like temperature and salinity. Employing the typical GM scheme at eddy-permitting resolution degrades eddies that have been resolved explicitly by the model. With the Leith schemes used in GM, they are argued to simultaneously not weaken resolved eddies while parameterising unresolved eddies. Added developments have also been carried out by the Met Office (MO), which include a weak GM coefficient used when the model cannot explicitly resolve eddies. These MO developments have shown promising results (Guiavarc’h et al., 2024). 

To date, we have carried out simulations examining the impact of the Leith viscosity parameterisations in the eddy-permitting NEMO model, ORCA025. More specifically, we run the Met Office’s Global Ocean Sea Ice 9 configuration. The above figure shows the results of the transport through Drake passage during the spin-up cycle. We intend to analyse the second cycle soon. The line labelled biharm is the standard simulation developed by the MO. Its respective dotted line is the same simulation without the MO changes. The other lines show the results of the Leith schemes. With the Leith schemes, we see an increase in transport of around 10-20 Sverdrups (Sv), which is exactly what we want to see, considering that observations are around 170 Sv.  

Our analysis is still in its infancy with many questions still requiring an answer. Do the Leith schemes reduce the Southern Ocean temperature biases? Do we see any changes to the formation of water masses? In the Southern Ocean, do we see any reduction in sensitivity of zonal transport and overturning circulation to wind speed changes? By improving the ocean circulation with the Leith closures, we hope to better utilise eddy-permitting models for long-range climate projections. Watch out for a publication later in the year. 

Further reading: 

Bachman, S. D., Fox-Kemper, B., & Pearson, B. (2017). A scale-aware subgrid model for quasi-geostrophic turbulence. J. Geophys. Res. Oceans, 122 (2), 1529–1554. https://doi.org/10.1002/2016JC012265

Guiavarc’h, C., Storkey, D., Blaker, A. T., Blockley, E., Megann, A., Hewitt, H. T., . . . An, B. (2024, May). GOSI9: UK Global Ocean and Sea Ice configurations. EGUsphere, 1–38. https://doi.org/10.5194/egusphere-2024-805

Kuhlbrodt, T., Jones, C., Sellar, A., Storkey, D., Blockley, E., Stringer, M., . . . Walton, J. (2018, November). The Low-Resolution Version of HadGEM3 GC3.1: Development and Evaluation for Global Climate. J. Adv. Model. Earth Syst.,10 (11), 2865–2888. https:doi.org/10.1029/2018ms001370

By: Dr. Patrick C. McGuire 

After a hiatus of 10 years, the Met Department has held a Research Away-Day once again. Over 150 Away-Day participants sauntered all the way to the Palmer Building. The Palmer Building is still on the University of Reading Whiteknights campus, but critically *away* from the Brian Hoskins and Harry Pitt buildings, where we might have otherwise been distracted by our normal working-day activities. 

The reincarnation of the Met Dept. Research Away-Day was the led by the now-outgoing Head of Department, Prof. Joy Singarayer. Following consultations that she held with department staff and students, it was clear that there was an enthusiasm to bring back the in-person vibe to the Met Dept. after it partially disappeared during the COVID pandemic.  

Photo 1: At the beginning of the morning main session in the lecture room (Photo credit: Joy Singarayer)

The in-person vibe had been partially lost to hybrid or completely-online Teams and Zoom meetings and to many of us working from home (WFH) for multiple days per week. Yes, the hybrid option (or recording option) in the meetings is so nice sometimes, when we need to take care of family members, when we are off at a conference, or when we really need to focus for a deadline, but the in-person vibe of the Met Dept had not really made its full return until we all met for the extended day-long Research Away-Day over in the Palmer Building last Thursday. 

Photo 2: Dhirendra Kumar giving a 4-minute short talk about a Simple loss model for European windstorms (Photo credit: Ankit Bhandekar)

We had 30 short 4-minute talks and two 12-minute keynote talks during the main sessions in a fit-for-purpose 150-seat auditorium. Dr. Andrea Dittus and Dr. James O’Donoghue were the keynote speakers. Andrea regaled us with her insights into What happens to the climate during stabilisation scenarios. James went on to amaze us with his stories about the Aurorae of Jupiter and their relation to heat flow. The 30 other speakers had a ‘long’ 4 minutes to wow the audience with their research vignettes on topics ranging from Urban-scale modelling, to Melt-ponds on sea ice, and further to Mineral dust in the atmosphere. 

 We also had morning and afternoon breakout sessions, with a total of 7 different topics. Participants stated their preferences for the breakout topics during registration. I helped to organize two of the breakout sessions: New software tools in climate and weather; and Statistics of climate risk, finance, & insurance. And I helped to facilitate the discussion in one of the breakout groups of the session on New software tools in climate and weather. In that discussion, we did spend almost 40% of the time talking about using large language models such as ChatGPT to serve as programming assistants. 

Photo 3: At the end of the afternoon main session in the lecture room (Photo credit: David Brayshaw)

I also was able to attend a different breakout session, about Using AI to help your research, led by Prof. Singarayer and Dr. Mark Muetzelfeldt. They did a superb job leading that session, and they gave us a group project to do at the end of the session, wherein we asked ChatGPT to produce Python code to do a first-pass trend analysis and visualization of the temperature and precipitation records from over 200 years of data from the Oxford weather stations. It was rather amazing to find out what improvements to ChatGPT have been made since it first came out over a year and a half ago. 

Photo 4: Research strategy & culture session, led by 3 Research Division Leads (Profs. Emily Black, Paul Williams, & Danny Feltham). (Photo credit: Joy Singarayer)

We did have a very-interesting, dedicated session where the 3 Research Division Leads (Profs. Emily Black, Paul Williams, & Danny Feltham) discussed Research strategy & culture. The audience opinions on how well the department is doing in providing excellent and inclusive research environments were all recorded and tabulated by the Mentimeter website. 

The poster session and scientific socializing were also both top notch. I was able to present my poster on Growing virtual crops in Peru during climate change, and I had several interested customers, including Dr. Martin Airey and Dr. Robin Hogan. I think the catering by Venue Reading was superb, and the food and drink were part of the reason the in-person vibe and the scientific socializing were able to return to our department to such a degree after the COVID & WFH hiatus. It’s been a long time since I’ve seen so many different groups of people in our department happily chatting away. It’s too bad that we don’t have a few photos of the poster hall.  

Photo 5: Water@Reading meetup (Harshita Gupta, Dr. Helen Hooker, David Richardson, Prof. Hannah Cloke, & Prof. Liz Stephens) at the Met Dept. Research Away-Day (Photo Credit: Hannah Cloke’s camera)

I found out about the return of the Research Away-Day from Prof. Singarayer when it was in its infancy, when she mentioned it while giving her perspective as Head of Department at one of our meetings of the Senior Researcher Forum. I raised my hand and said that I wanted to help organize the Research Away-Day because I had organized a similar event for the Geosciences Dept. when I was working at the Free University of Berlin, called the Internal GeoSymposium. I had also previously attended events of a similar nature at the University of Chicago’s Geophysical Sciences Dept. (the Noon Balloons), in the University of Arizona’s Astronomy Dept. (internal symposia), and in Bielefeld University’s Computer Science Dept. (an overnight retreat to a meeting place in a small village).  

Some of you have probably attended either our department’s Research Away-Days over 10 years ago or other universities’ departments’ Research Away-Days of some sort. These Research Away-Days have a long tradition, apparently. I would speculate that Research Away-Days go back many centuries, maybe even to the time of the ancient Greeks. 

 The other co-organizers (Dr. Ambrogio Volonté, Dr. Daniel Shipley, Dr. Thomas Wilder, and Dr. Holly Ayres) of the University of Reading Met Dept. Research Away-Day also found out about the initiative from Joy while attending either the Senior Researcher Forum or the Post-doctoral Forum. Together with Joy, we started having organizational meetings for the Research Away-Day in November 2023, meeting every few weeks, with an increasing cadence of weekly meetings starting in early May 2024. 

Figure 1: Preliminary, incomplete polling results from Research Away-Day participants about how often the future Away-Day conferences should be. The single participant who selected Other suggested a 3-year cadence.

I’m personally hoping that we can have another Met Dept. Research Away-Day in one year, but I have heard some people suggest that every second year would be better. Alas, I have also heard an enthusiastic endorsement of holding another Research Away-Day in 6 months’ time. Results from a more-complete, yet-ongoing survey are shown in Fig.1, above. Twenty-three (23) participants suggested every year; sixteen (16) participants suggested every two years; 1 participant suggested every three years; 1 participant suggested every 6 months; and nobody has suggested never having it again. In the comments section, some people have suggested having a two-day Away-Day, and some people have suggested having the weekly Lunchtime Seminars sometimes feature multiple short talks. Regardless, let’s keep this in-person vibe going! 

By: Dr. Yumeng Chen

“The world is not perfect. Every measurement should come with an error bar.” This is what I learned before I stepped into the fluid dynamics lab as a student many years ago. This statement still echoes now when I work on data assimilation (DA). Because neither observations nor model forecasts are perfect, based on their errors/uncertainties, DA combines both observations and forecasts to provide an estimate of the most likely state of the modelled system. This estimate also comes with an error bar that should give reduced uncertainties compared to both the model forecast and observations. 

Arctic sea ice, as an important component of the climate system, regulates solar radiation, provides habitants for marine life, and influences human activities. Like a variety of fields such as numerical weather prediction, marine ecosystems, and land surface modelling, the Arctic sea ice community adopts the DA technique operationally to provide an estimate of the state of the Arctic sea ice for better prediction. The seasonal forecast of the Arctic sea ice is improved by better initialisation of the sea ice concentration (percentage of the Arctic sea ice in each grid cell)  and sea ice thickness (Kimmritz et al., 2019). However, initial conditions are not the only source of uncertainty of the Arctic sea ice. Numerical models can also suffer from erroneous parameters. These errors can cause biases in Arctic sea ice prediction especially in long-term simulations. 

Fortunately, DA can also provide estimates of unobserved model fields and model parameters. How can DA estimate something that we do not observe? This is based on the relationship between the errors of parameters and observed model forecasts. This means that if we know the error of the model forecast from observations, using this relationship, we can infer and reduce the error in the model parameters. In most operational DA methods, the relationship between parameter and forecast errors is represented by error covariances which are derived from the numerical models. Thus, the performance of parameter estimation using DA depends heavily on the dynamics of the numerical models. 

This dependence on model dynamics can cause problems for the parameter estimation. For example, DA could provide wrong estimates when multiple sets of model parameters lead to the same model forecast – if you know, this implies an ill-posed problem. Also, parameter estimation may fail when the parameters are not so sensitive to the observed model fields compared to other sources of uncertainties. To demonstrate the potential issue of parameter estimation when forecast errors of observed model fields do not dominantly come from parameter errors, we can apply DA to estimate parameters in a novel ‘dynamics-only’ sea ice model developed in the scale-aware sea ice project (Chen et al., 2023).  

We set up an idealised experiment where a block of sea ice is forced by periodic random storms (Figure 1a). In this idealised setup, no thermodynamics exists, and the wind is the only external forcing. In this dynamics-only sea ice model, depending on the state of the sea ice, the sea ice can deform elastically like a spring, deform permanently like viscous fluid, and break abruptly under its internal dynamics or external forcing which cause damage and fracture of the sea ice. The initial state of the sea ice is not damaged and the main errors of the experiment setup come from the wind of the random storms. 

Figure 1: a) An illustration of the experiment setup where the quivers show the wind field, and the sea ice is thicker in the middle of the domain than the boundaries with decreasing the sea ice concentration due to the constant wind forcing; time series of b) air drag coefficient estimation using sea ice velocity observations, c) damage parameter estimation using sea ice velocity observations, and d) damage parameter estimation using sea ice concentration observations.

In such an idealised setup, we can decide the true model state and parameters and assign errors to our chosen parameters. One important model parameter of the sea ice model is the air drag coefficient. This coefficient decides how the wind influences the sea ice velocity. The error in the wind field can be magnified or reduced by this coefficient in the sea ice velocity.  

Let us assume that the air drag coefficient is erroneous and all other model parameters are correct. In this case, we can get very accurate estimates of the air drag coefficient using DA when we use sea ice velocity observations (Figure 1b). Another important model parameter of the sea ice model is called damage parameter. This parameter determines the response of sea ice motion to the forces exerted on them when the sea ice is damaged. With low value of the damage parameter, the sea ice can behave like an elastic spring; with high value of the damage parameter, the sea ice is more sluggish like viscous fluid in response to the forces. However, sea ice velocity observations cannot provide a reliable estimation of the damage parameter (Figure 1c) when the damage parameter is the only erroneous model parameter. As sea ice velocity is mostly influenced by the wind field, the sea ice velocity cannot be used to infer the damage parameter reliably. Instead, we see improved parameter estimation with a combination of sea ice concentration (Figure 1d). 

Here, our example shows that parameter estimation using DA could be challenging and must be performed carefully. However, DA is still a powerful tool for improving errors in models based on available observations. 

References and Further Reading 

Kimmritz, M., Counillon, F., Smedsrud, L. H., Bethke, I., Keenlyside, N., Ogawa, F., & Wang, Y. (2019). Impact of ocean and sea ice initialisation on seasonal prediction skill in the Arctic. Journal of Advances in Modeling Earth Systems, 11, 4147–4166. https://doi.org/10.1029/2019MS001825 

Chen, Y., Smith, P., Carrassi, A., Pasmans, I., Bertino, L., Bocquet, M., Finn, T. S., Rampal, P., and Dansereau, V.: Multivariate state and parameter estimation with data assimilation on sea-ice models using a Maxwell-Elasto-Brittle rheology, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-1809, 2023. 

By: Dr. Hripsime Mkrtchyan, Prof. Giles Harrison, Prof. Keri Nicoll 

AtmosEleC – Atmospheric Electricity for Climate is a digitisation project designed to help researchers investigate the connections between atmospheric electricity and climate change. It has recently been launched on Zooniverse and is seeking volunteers to help with digitisation. You can be part of a citizen science project by clicking here: AtmosEleC – Atmospheric Electricity for Climate — Zooniverse 

Zooniverse is recognized as the largest and most popular platform globally for citizen science projects. This platform enables volunteers to support scientists in various research tasks that seek human input. These tasks include classifying galaxies, transcribing ancient manuscripts, or monitoring wildlife and now, digitizing handwritten atmospheric electricity records. Volunteers engaging with Zooniverse have the unique opportunity to participate in cutting-edge research, significantly contributing to advancements in these fields.   

Our project AtmosEleC focuses on rediscovered historical atmospheric electricity data from Lerwick Observatory in Shetland, UK. 

Figure 1. Observatories in the UK.

Lerwick Observatory (see the map above) was established in 1921 in Shetland, Scotland, and it became an important site for atmospheric electricity measurements in 1925. These measurements were continued systematically by the observatory staff until 1984.  

The Observatory’s northerly location was originally chosen for magnetic measurements and studying auroral and meteorological phenomena at high latitudes, following assistance sought by the Norwegian government. Lerwick observatory has recently celebrated its 100th anniversary and continues to make important atmospheric and magnetic measurements. More information about atmospheric observations made in Lerwick is given here. 

Almost all of the Lerwick hourly measurements during 1925-1984 have now been recovered, including many original handwritten records. Such long datasets are now recognised as precious resources for modern atmospheric science. However, to unlock the scientific opportunities of this remarkable information source, the data must first be accurately transcribed and keyed. It is for this that we need volunteers to help. 

The Earth’s atmosphere is continually electrified, and during fair weather, it is positively charged with respect to the ground. This has been known since the time of Benjamin Franklin, and scientific investigation has been motivated to explain it. The most common quantity in atmospheric electricity is the vertical electric field, which is a measure of the strength of the electrification. It is quantified as the Potential Gradient (PG). Near the surface, the PG is, conceptually, the voltage difference between the ground and a point one metre vertically above it. The PG can be measured using a potential probe (also known as a collector or equaliser), at a fixed height above the surface. 

Figure 2. Concept of a potential probe. A conducting electrode placed at a height z above the surface will acquire the potential of the atmosphere (V0) which can be measured using a sensitive voltmeter.

The PG is determined from the electric potential measured at a fixed point above the surface, using a sensing electrode of some kind and a voltage registering device. This is a demanding measurement, requiring high quality insulation and a sensitive electrostatic voltmeter. It is difficult to sustain the good insulation required in all weather conditions, and the conditions at Lerwick are often quite variable. At Lerwick, a radioactive probe was used as the sensing electrode. This was connected to an electrometer and chart recorder, so that continuous recordings could be made. This recording paper apparatus was known as  Benndorf “electrograph”(used from 1925-1960, see fig 3).  

Figure 3 The Benndorf electrometer (Adopted from Harrison 2022). (a) Schematic view of a Benndorf device with recording paper (b) Internal view of the one of the Benndorf electrometers (c) Benndorf electrometer paper chart, from Lerwick, for 2 November 1960.

The PG values on the record sheets were averaged and tabulated as monthly sets of daily values in the annual volumes of the Observatories’ Yearbook until 1967, and thereafter on individual summary sheets until 1984, stored in the National Meteorological Archive (Harrison 2022). 

Figure 4 PG data record sheet of archive data (Image credit: Lerwick Observatory Archive)

At any local site, the PG is influenced by local meteorological conditions (such as lightning, fog, rain, snow, aerosols), space weather, and radioactivity. PG measurements are particularly good at detecting and monitoring radioactive deposition (e.g. from nuclear tests or nuclear power plant releases), and data from Lerwick during the 1950s and 1960s has already demonstrated the effect of the distant detonations of nuclear weapons on the PG at the site.  

An emerging new application is that PG measurements are closely linked to global thunderstorm activity, and any changes in it. This occurs through the Global Electric Circuit (GEC), which connects distant disturbed weather with fair weather regions. One of the things we aim to investigate with the Lerwick dataset is the influence of Pacific Ocean temperatures on the GEC through El Niño events (a warming of sea surface temperature). We have already established that the Lerwick PG is related to El Niño, demonstrating an atmospheric link extending over a distance of 8000 km. 

We will be very glad for any help you can offer to digitise this important dataset and lease join our Zooniverse project if you are interested in learning more (https://www.zooniverse.org/projects/hripsi-19/atmoselec-atmospheric-electricity-for-climate)! 

Reference and Further Reading

Harrison, R. G. and Riddick, J. C.: Atmospheric electricity observations at Lerwick Geophysical Observatory, Hist. Geo Space. Sci., 13, 133–146, https://doi.org/10.5194/hgss-13-133-2022, 2022.

By: Prof. Remi Tailleux 

As is well known, atmospheric winds and ocean currents ultimately derive their energy from the Sun. In general, this involves a two-step process, whereby the solar energy is first transformed into potential energy (PE) before finding its way into kinetic energy (KE). Some aspects of this transformation remain mysterious and poorly understood, however, which are to do with the so-called `cooling paradox’. In the oceans, cooling at high latitudes creates dense waters that sink and give rise to the Atlantic meridional overturning circulation (AMOC). The puzzle here is that the cooling decreases the PE of the oceans. How then can it be converted into KE?  

The cooling paradox has been a recurring source of confusion and controversies in energetic studies of the atmosphere and oceans, so much so that one of the most famous textbooks on buoyancy-driven flows (Turner, 1973) stays away from energetics altogether. It is only with the work of Lorenz (1955) that a resolution of the cooling paradox started to emerge. Lorenz’s key contribution was to remark that some configurations, such as the equilibrium resting state depicted in the right panel of Fig. 1, possesses lots of potential energy, yet none of it is readily available for conversions into KE.

Figure 1. Schematics depiction of the potential temperature field in the atmosphere. Left panel shows that in the actual state, potential temperature varies with latitude, with cold air at in the surface layers in the polar regions (indicated in purple). Right panel shows the reference or global equilibrium state of the atmosphere, obtained by allowing each layer to relax towards their equilibrium state while retaining their mass and potential temperature. The equilibrium state is horizontally uniform and therefore a function of height or pressure only. Reproduced from Tailleux (2013).

Another way to look at the problem is by considering the two fluid configurations illustrated in Fig. 2. Both configurations have identical amounts of potential energy, but only the one in the right panel is expected to develop motions. Lorenz therefore posited that potential energy must come into two distinct flavours: Available potential energy (APE) and background potential energy (BPE): 

                                                           PE = APE + BPE 

Figure 2. Schematics illustrating two idealised stratifications having the same potential energy (PE), but a different partition APE/BPE. Intuitively, the fluid on the left is purely static and has no APE. No motion is expected to develop. The fluid on the right, however, is clearly unstable, as intuitively, one expects the fluid on the top right to start sinking owing to being denser than the other parcels. The fluid on the right has APE, while the fluid on the left has none, despite the two fluids having the same PE. Reproduced from Hughes et al. (2009).

Lorenz APE theory resolved the cooling paradox because although high latitudes cooling decreases both the PE and BPE of the oceans, it increases the APE, which is what matters, as this is the part of the PE that can be converted into KE.  

While the Lorenz discovery was revolutionary and dramatically altered energetics studies in the oceans and atmosphere, this is not to say that everybody was happy with Lorenz formulation of APE theory. Indeed, unlike kinetic energy, which can be defined for individual fluid parcels, Lorenz APE could only be defined for the fluid as whole. This was a major issue, which prompted the quest for local principle. Understanding how to do this took over two decades, as the first satisfactory local APE theory only appeared in 1981 as reviewed in Tailleux (2013). These early formulations, however, were still relatively obscure and complicated. It took over two decades for the local APE theory to be digested and reformulated before it started to be used in practical applications. One key process that can only be discussed with the local APE theory is the advection or transport of APE between different regions. As it turns out, advection of APE was recently established to be of key importance for understanding the energetics of atmospheric storm tracks (Novak and Tailleux, 2008) and of tropical cyclone intensification (Harris et al, 2022) by two PhD students in the meteorology department, thus highlighting the usefulness and importance of the local APE framework.   

Another advantage of the locally defined APE is that it can be partitioned into `mean’ and `eddy’ components, similarly as what is commonly done with kinetic energy. In the atmosphere, the mean APE and kinetic energy (KE) can be seen in Fig. 3 to characterise the large-scale circulation of the atmosphere, whereas the eddy component characterise the storm tracks, that is, the regions dominated by low pressure systems. In the oceans, one may similarly diagnose the eddy APE, which similarly characterise the storm tracks of the oceans, as depicted in Fig. 4.  

Figure 3. Mean and eddy components of Available Potential Energy and Kinetic Energy in the Northern Hemisphere. (a) Mean APE, (b) Mean KE, (c) Eddy APE, (d) Eddy KE. Reproduced from Fig. 5.8 of Novak (2016).

So far, the power and usefulness of the local APE framework has only been used in a few studies and therefore remain under-exploited. The local theory of APE is not completely settled yet and continue to evolve (Tailleux, 2018). Nevertheless, it is now clear that it represents a much better framework than Lorenz global APE theory, but more work is needed to unlock its full potential.    

Figure 4. Eddy APE at 175 m estimated from the monthly Armor3D data. Regions of intense meso-scale eddy activity appear in red. Notable regions are the Gulf Stream and Kuroshio regions and the Southern Ocean.

References and Further Reading: 

Harris, B.L., R. Tailleux, C. E. Holloway, and P.L. Vidale, 2022. A moist available potential energy budget for an axisymmetric tropical cyclone. J. Atm. Sci., 79, 2493—2513 

Hughes, G., O, A. M. Hogg, and R. W. Griffiths, 2009: Available potential energy and irreversble mixing in the meridional overturning circulation. J. Phys. Oceanogr., 39, 31300—3146. 

Lorenz, E. N., 1955. Available potential energy and the maintenance of the general circulation. Tellus, 7, 157—167. 

Novak, L., 2016. The lifecycle of storm tracks. PhD Thesis. University of Reading.  

Novak, L, and R. Tailleux, 2008: On the local view of atmospheric potential energy. J. Atm. Sci., 75, 1891—1907.  

Tailleux, R., 2013. Available potential energy and exergy in stratified fluids. Annual Review of Fluid Mechanics. 45, 35—58.  

Tailleux, R., 2018. Local available energetics of multicomponent compressible stratified fluids. J. Fluid Mec., 842, R1.  

Turner, JS, 1973. Buoyancy effects in fluids. Cambridge University Press.  

By: Dr. Xiaocen Shen

Teleconnections are usually manifested as recurring patterns which link weather and climate anomalies (departures from long-term average) over large distances across the globe (e.g., Wallace and Gutzler 1981). Therefore, they play an important role in shaping climate variability and regional climate change. One striking example is El Niño-Southern Oscillation (ENSO) teleconnection (Figure 1). ENSO is the most prominent year-to-year internal variability in the climate system, although ENSO itself occurs in the tropical Pacific and is reflected as a fluctuation between unusually warm and cold conditions, it excites wave-trains propagating into the extratropics which then strongly affect the precipitation and temperature over mid-high latitudes (e.g., Trenberth et al. 1998). Moreover, the extratropical circulation response to ENSO can further influence the upward propagation of planetary waves in the midlatitudes, thereby leading to significant changes of the stratospheric polar vortex (SPV) via the wave-mean flow interaction (e.g. Domeisen et al., 2019). The induced SPV anomaly, in return, can further descend into the troposphere, modulating the weather condition in the midlatitudes. For instance, this ENSO-SPV linkage is shown to strongly contribute to the cooling over northern Europe in late winter following El Niño (e.g., Ineson and Scaife 2009). Hence, studying teleconnections is not only helpful to better understand the atmospheric circulation variability, but is also key to improving the prediction skill. 

Figure 1. Schematic of ENSO teleconnections (from Domeisen et al. 2019)

While the observations are the starting point for studying teleconnections, the limited records make it difficult to draw robust conclusions based on observations alone. On the one hand, the climate system is complex, and the relationships sometimes vary in different time periods (e.g., Dong and McPhaden 2017). On the other hand, the simple correlation between two circulations does not always indicate a physical teleconnection, it can sometimes instead reflect co-varying changes in response to other factors (Kretschmer et al. 2021).  

To address this problem, climate model simulations are widely used as they can provide more samples. However, climate models do not always agree with the observed teleconnections, in which case the discrepancy is known as model bias.. There are two main potential reasons for the discrepancy. First, the climate models may fail to capture the key physical processes and therefore cannot reproduce the teleconnections. Second, due to the limited sample size in the observations, the discrepancy may reflect the internal variability of the climate system. Therefore, to confidently use models to study teleconnections and the related aspects, scientific judgement is needed to justify the suitability of models when there are apparent discrepancies (e.g. Jain et al. 2023). In our recent research, we have advocated the use of a forensic investigation approach to understand the discrepancies between climate models and observations, which then helps to make the decision on whether the model outputs can be trusted.  

The literal definition of a forensic investigation is the scientific analysis of physical evidence from a crime scene. The logic behind it is to conduct a thorough examination of the evidence to establish the facts and uncover the truth. In the context of assessing model discrepancies in teleconnections, a physically-based analysis is required to understand their origin, which can then provide an evidential basis for deciding whether a model is appropriate for a given scientific purpose. Since the logic is similar with that of solving crime puzzles, the term forensic is used here to characterize our approach (Figure 12). In the following, the case of ENSO-SPV relationship will be shown as an example of how to extract reliable information from the model output using the forensic investigation approach.

Figure 2. The logic of forensic investigation to unravel the climate model biases (adapted from images by Freepik)

In a climate model called the MIROC6, the ENSO-SPV relationship is opposite to observations. At the first glance, it seems that this model is not suitable for studying the ENSO-SPV relationship and related scientific questions. However, according to the physically-based analysis, we found that MIROC6 model actually well captures the relevant dynamical processes, including the extratropical response to ENSO, the anomalous upward propagation of planetary waves, and the wave-mean flow interaction in the stratosphere. The discrepancy is further shown to be mainly related to the wave propagation within the stratosphere, which eventually lead to the different SPV response. This reflects that the causal linkage between ENSO and SPV is shaped by other factors and/or background states, known as the state dependence. Therefore, although the model shows an opposite ENSO-SPV relationship to the observation, it is physically reasonable.  

Furthermore, the observations show a state dependence similar to that of the model results, in that the observed ENSO-SPV relationship is not stable and is shaped by other factors, such as the ocean background condition (e.g. Rao et al. 2019). The observational evidence neither supports nor contradicts this state dependence found in the model due to the limited sample size. Thus, depending on the specific purpose of the research, different choices can be made in how to use the model simulations.  

If the study does not require a stable teleconnection, for example, if it is used to study non-stationarity and state dependence, then the model can be used directly to provide conditional information. On the other hand, if the study requires a stable teleconnection consistent with observations, then the model should be used only after the application of a physically-based bias adjustment. In the ENSO-SPV relationship case, under the assumption that the state-dependance is spurious, a physically-based bias adjustment is applied to SPV, which effectively aligns the modelled ENSO-SPV relationship with the observations, thereby removes the model-observations discrepancy in the surface air temperature response. 

This case gives an example of how the forensic approach could help us to better understand the difference between models and observations, allowing us to make full use of climate model outputs (Figure 3). Similar physically-based approaches have been widely used in the climate research in recent decades (e.g., Kretschmer et al. 2020; Shepherd 2021), providing us with more opportunities to gain a more holistic view of model performance and to extract more information from models.  

Figure 3. The forensic investigation processes. The direction of the arrows indicates the order in which actions are taken. The bubbles enclosed by the black contours indicate the conclusions about whether we can trust the model and how to use the model outputs.

Further Reading/References: 

Domeisen, D. I. V., Garfinkel, C. I., & Butler, A. H. (2019). The teleconnection of El Nino South-ern Oscillation to the stratosphere. Reviews of Geophysics, 57(1), 5-47. https://doi.org/10.1029/2018rg000596 

Dong, L., & McPhaden, M. J. (2017). Why has the relationship between Indian and Pacific ocean decadal variability changed in recent decades? Journal of Climate, 30(6), 1971-1983. https://doi.org/10.1175/jcli-d-16-0313.1 

Ineson, S., & Scaife, A. A. (2009). The role of the stratosphere in the European climate response to El Nino. Nature Geoscience, 2(1), 32-36. https://doi.org/10.1038/ngeo381 

Jain, S., Scaife, A. A., Shepherd, T. G., Deser, C., Dunstone, N., Schmidt, G. A., Trenberth, K. E., & Turkington, T. (2023). Importance of internal variability for climate model assessment. npj Cli-mate and Atmospheric Science, 6(1), 68. https://doi.org/10.1038/s41612-023-00389-0 

Kretschmer, M., Adams, S. V., Arribas, A., Prudden, R., Robinson, N., Saggioro, E., & Shepherd, T. G. (2021). Quantifying causal pathways of teleconnections. Bulletin of the American Meteorological Society, 102(12), E2247-E2263. https://doi.org/10.1175/bams-d-20-0117.1 

Kretschmer, M., Zappa, G., & Shepherd, T. G. (2020). The role of Barents–Kara sea ice loss in projected polar vortex changes. Weather Climate Dynamics, 1(2), 715-730. https://doi.org/10.5194/wcd-1-715-2020 

Rao, J., Garfinkel, C. I., & Ren, R. C. (2019). Modulation of the Northern winter stratospheric El Nino-Southern Oscillation teleconnection by the PDO. Journal of Climate, 32(18), 5761-5783. https://doi.org/10.1175/jcli-d-19-0087.1 

Shepherd, T. G. (2021). Bringing physical reasoning into statistical practice in climate-change sci-ence. Climatic Change, 169(1-2), 2. https://doi.org/10.1007/s10584-021-03226-6 

Trenberth, K. E., Branstator, G. W., Karoly, D., Kumar, A., Lau, N. C., & Ropelewski, C. (1998). Progress during TOGA in understanding and modeling global teleconnections associated with tropical sea surface temperatures. Journal of Geophysical Research-Oceans, 103(C7), 14291-14324. https://doi.org/10.1029/97jc01444 

Wallace, J. M., & Gutzler, D. S. (1981). Teleconections in the geopotential height field during the Northern Hemisphere winter. Monthly Weather Review, 109(4), 784-812. https://doi.org/10.1175/1520-0493(1981)109<0784:Titghf>2.0.Co;2