Climate: Past, Present & Future

Climate: Past, Present & Future

Ostracods, the sentinels of past oceanic circulation

Ostracods, the sentinels of past oceanic circulation
Name of the proxy


Type of proxy

Paleoenvironment proxy


All types of aquatic environments but here we will focus on marine waters

Period of time investigated


How does it work?

Ostracoda are crustacean of millimetre size which have inhabited all types of marine environments from the Ordovician to today (e.g. Salas et al. 2007) and colonized continental water bodies during the Carboniferous (Bennett et al. 2012). They are characterised by their bivalve calcified carapace articulated dorsally which encloses and protects the soft parts and appendages of the animal (Figure 1). The majority of Ostracoda live on or in the sediments: they are consequently highly sensitive to their environment.

What are the key findings that have been done using this type of proxy?

Throughout their history, marine Ostracoda inhabiting deep seas had very different morphologies from the contemporary shallow water species: thin shells, long, hollow and delicate spines and no eye spots (although this point is discussed; Figure 2). Based on the study of sediments, associated organisms and analogies with modern-days Ostracoda, ostracodologists concluded that those animals developed in low energy environments ranging from 500 to 5000 m depth in connection with global ocean cold water supplied by ice-caps (Lethiers & Feist 1991). This discovery provided a unique window into the oceanic circulation through geological times and the existence of a cold deep-water layer. The presence and characteristics of these Ostracoda have been cornerstones in understanding that the thermohaline circulation has not been constant through the Phanerozoic but rather existed only during the Late Ordovician, the Carboniferous-Permian interval and from the Eocene to today (Benson 1975).

Figure 2. Simplified geological time scale with Eras and Periods of the Phanerozoic. On the right are reported some archetypal deep-sea Ostracoda from the literature (for all photos, scale bar is 100 µm). A: Processobairdia spinanterocerata Bless & Michel, 1987; B: Cristanaria katyae Crasquin-Soleau, 2008; C: Gencella taurensis Forel, work in progress; D: Pedicythere klothopetasi Yasuhara et al., 2009.

Today, this field of research is very active as Ostracoda are the only metazoans regularly fossilized in deep-sea sediments over an extremely long period of the history of Earth. Their long fossil record spanning 5 mass extinctions and periods of extreme climatic changes make them precious tools to unravel the response of deep-water ecosystems to past climatic changes and the rhythms of their recovery. The extreme sensitivity and history of these peculiar animals make them sentinels of deep-sea ecosystems facing ongoing global temperature increase and acidification of marine waters.

  • Bennett, C.E., Siveter, D.J., Davies, S.J., Williams, M., Wilkinson, I.P., Browne, M., Miller, C.G. 2012. Ostracods from freshwater and brackish environments of the Carboniferous of the Midland Valley of Scotland: the early colonisation of terrestrial water bodies. Geological Magazine, 149, 366-396.
  • Benson, R.H. 1975. The origin of the psychrosphere as recorded in change of deep sea Ostracode assemblages. Lethaia, 8, 69-83.
  • Bless, M.J.M., Michel, M.P. 1967. An ostracode fauna from the Upper Devonian of the Gildar-Monto region (NW Spain). Leidse Geologische Mededelingen, 39, 269-271.
  • Crasquin-Soleau, S., Carcione, L., Martini, R., 2008. Permian ostracods from the Lercara Formation (Middle Triassic to Carnian?, Sicily, Italy). Palaeontology, 51, 537-560.
  • Lethiers, F., Feist, R. 1991. Ostracodes, stratigraphie et bathymétrie du passage Dévonien–Carbonifère au Viséen Inférieur en Montagne Noire (France). Geobios, 24, 71-104.
  • Salas, M.J., Vannier, J., Williams, M. 2007. Early Ordovician Ostracods from Argentina: their bearing on the origin of Binodicope and Palaeocope clades. Journal of Paleontology, 81, 1384-1395.
  • Yasuhara, M., Okahashi, H., Cronin, T.M. 2009. Taxonomy of Quaternary Deep-Sea Ostracods from the Western North Atlantic Ocean. Palaeontology, 52, 879-931.

Written by Marie-Béatrice Forel

Edited by Célia Sapart and Caroline Jacques

Hot towns, summer in the city!

Hot towns, summer in the city!

Cities obviously experience a different climate than natural landscapes. Already in 1810 the British meteorologist Luke Howard documented that the air temperature in the city of London was several degrees higher than in its surroundings. This so called urban heat island has several causes. In general the relatively dark surfaces of asphalt and roofs absorb solar radiation very efficiently and this heat is stored in building material during the day. At night this heat is released to the atmosphere, which keeps the city warm. Moreover, heat produced by air conditioning, traffic and industry contributes substantially to a city’s heat load. With a decreased amount of vegetation, cities also lose the shade and cooling effect of trees (Figure 1).

Currently already over 50% of the world’s population is residing in urban areas and this number is foreseen to increase even further in the future. Moreover, climate model projections indicate that heat waves will occur more often in the future. Together these developments will make many citizens potentially vulnerable to  urban heat. Although a slight temperature increase might look appreciable at first glance, elevated temperatures affect human health, since hospital visits and mortality are enhanced in warm conditions (about 2% per degree Celsius, e.g. Hajat et al., 2002). In addition, labour productivity in warm periods is reduced, resulting in economic losses. E.g. for Australia this was estimated to be about $650 per capita (Zander et al, 2015) which is a substantial contribution to the national income. Also, the urban energy demand needed for heating purposes in winter and cooling in summer is governed by urban weather. Finally cities are vulnerable to flooding in case of extreme precipitation by peak showers, when the sewage system capacity is hampered. Hence how do cities manage urban heat and keep dry feet?

Behind the general picture on the urban heat island, several scientific questions do remain. E.g. what is the temperature variability within a city, and how can we monitor temperatures? Also, can we make special weather forecasts for cities? Monitoring urban weather and climate is challenging since traditional weather stations are not suitable for urban areas since they require undisturbed terrain. Crowdsourcing, i.e. the collection of weather data by citizens has now become popular. Many hobby meteorologists have installed weather stations at home, and distribute their data directly via several websites as,, and These crowdsourced observations were crucial in estimation of the urban heat island effect in the Netherlands, a west-European country with a mild maritime climate were little attention was paid to urban climate. On hot summer days the urban heat island was over 6 degrees (Steeneveld et al., 2011, Heusinkveld et al., 2014)!

The urban climate can be efficiently monitored by tricycle traverse measurements (Figure 2). Bikes are excellent modules to measure the urban climate, since a wide variety of urban morphology (vegetation cover, building design and material) can be explored, especially outside the main roads. And it is carbon free, all driven by electricity generated by solar panels. Moreover, the bike is appealing for the general public.

Figure 2: A cargo tricycle equipped with a weather station to measure temperature, humidity, wind speed, solar and thermal radiation (source: Wageningen University; design Bert Heusinkveld).

What do we learn from such bike traverses? Figure 3 shows the variety of temperatures observed during a heat wave in a mid-size town in the Netherlands. Obviously, local temperature differences at the end of the afternoon may reach up to 3.5 ºC in this case. In general the town centre is relatively warm, though more surprisingly the relatively young neighbourhoods at the city edges appear to be warm too. In these neighbourhoods the vegetation is young, resulting in limited shadowing and therefore efficient heat absorption in roads and building walls. Local cool spots appear in parks and at small lakes. Luckily, enough room to escape from the heat!

Figure 3: Air temperature observations in the mid-size town Wageningen (the Netherlands, August 2nd 2013, 17.00 local time) obtained from two bike traverses. Source land cover maps:

No weather station in your garden? We still catch you via your smartphone! Smartphone users with the OpenSignal App that is intended to monitor wireless network capacity provide as a by product the smartphone battery temperature. Recently it was discovered that the temperature of your smartphone battery follows the air temperature outdoor (Overeem et al., 2013). Inversely this means that if we know the smartphone battery temperature, we can estimated the outdoor temperature. This insight offers a high potential for recording urban temperatures in areas where observations are rather scarce. The open question remains whether spatial and temporal scales these observations are applicable. Is it possible to get a reliable temperature record for your neighbourhood via available smartphones. Also, will these modern Big Data techniques change the paradigm on performing traditional measurements?

Weather forecasts on TV and radio never focus on the detailed weather for cities, which is somewhat surprising since many human activities, human health and critical infrastructures depend on the city temperature. Since computer power has rapidly grown the last decades, and still is, weather forecast models have refined their grid spacing. With this refinement urban areas “become visible” for these models. On one hand this offers a great potential for city specific forecasts. On the other hand information about the urban morphology is needed to feed these weather forecast models. For example they need to know whether urban districts contain skyscrapers or just three-story residential areas, and for example how much vegetation is present. Thereto the World Urban Database and Portal Tool is set up in which local experts document their city. You are welcome to join to describe your city and inform our weather forecast models how your city looks like!

  • Heusinkveld, B.G., G.J. Steeneveld, L.W.A. van Hove, C.M.J. Jacobs, and A.A.M. Holtslag 2014: Spatial variability of the Rotterdam urban heat island as influenced by urban land use, J. Geophys. Res, 119, 677–692.
  • Overeem, A.,  J. C. R. Robinson,  H. Leijnse,  G. J. Steeneveld,  B. K. P. Horn, and  R. Uijlenhoet (2013), Crowdsourcing urban air temperatures from smartphone battery temperatures, Geophys. Res. Lett., 40, 4081–4085, doi:10.1002/grl.50786
  • Steeneveld, G.J., S. Koopmans, B.G. Heusinkveld, L.W.A. van Hove, and A.A.M. Holtslag, 2011: Quantifying urban heat island effects and human comfort for cities of variable size and urban morphology in The Netherlands, J. Geophys. Res., 116, D20129, doi:10.1029/2011JD015988.
  • Zander, K.K., W.J.W. Botzen, E. Oppermann, T. Kjellstrom, S.T. Garnett, 2015: Heat stress causes substantial labour productivity loss in Australia, Nature Climate Change  5, 647–651.

Written by Gert-Jan Steeneveld, associate Professor at Wageningen University, The Netherlands

Edited by Célia Sapart and Caroline Jacques

The Climate Tango of ENSO and CO2

In 1904, the Swedish chemist Svante Arrhenius suggested that the burning of fossil fuels to satiate our hunger for energy would increase the percentage of carbon dioxide (CO2) in the atmosphere, which would change the Earth’s temperature. Regular measurements of atmospheric CO2, started in the late 1950’s at remote locations such as Mauna Loa in Hawaii and the South Pole, confirmed his hypothesis about increasing CO2, with one important caveat. The rate at which CO2 was accumulating in the atmosphere did not match the rate at which it was being produced by fossil fuel burning. In fact, the atmosphere was apparently retaining only half of what was pouring in. Figure 1 shows the observed atmospheric mole fraction of CO2 – moles of CO2 per mole of dry air, usually expressed in “parts per million” – changing with time, along with the change we would have expected if all our fossil fuel emissions would have stayed in the atmosphere.

Figure 1: Measured (CO2) mole fraction (moles of (CO2) per mole of dry air) at Mauna Loa, Hawaii in blue, which is a good approximation to the global average atmospheric (CO2). In red is what the global average mole fraction would have been if all the fossil fuel and land use change (mostly biomass burning) emissions since preindustrial times had accumulated in the atmosphere. The green shaded area, therefore, is (CO2) that was emitted due to human activity but did not stay in the atmosphere. Compared to the preindustrial baseline of 280 parts per million (ppm), the (CO2) accumulated in the atmosphere is roughly half of what was emitted. Monthly mean Mauna Loa (CO2) measurements from the Scripps Institute of Oceanography (SIO, and National Oceanic and Atmospheric Administration (NOAA,, fossil fuel emissions from the Carbon Dioxide Information Analysis Center (CDIAC,, and land use change emissions from the Global Carbon Project.

Today, we know that the other half – the fossil fuel CO2 emission “missing” from the atmosphere – is taken up by the land biosphere and oceans. The land biosphere accumulates carbon from the atmosphere by increasing plant mass, while the oceans dissolve atmospheric CO2 as the accumulating fossil fuel CO2 in the atmosphere drives the ocean-atmosphere carbon equilibrium out of balance. Beyond this large-scale picture, however, much still remains uncertain. Which part of the land biosphere – the tropics, the temperate latitudes, or the boreal forests and grasslands – takes up the most carbon? How does that uptake change from, say, a drought year to a rainy year? Do old growth forests and young forests take up carbon differently? To what extent do the land and ocean uptakes respond to natural climate cycles such as El Niño or the Pacific Decadal Oscillation?

Answering these questions accurately will not only tell us how the carbon cycle works today, but also how it might respond to a changing climate in the future. Almost all climate models used today to predict future climate contain a model to simulate the response of the carbon cycle (such as the land uptake of CO2) to future climate forcings such as droughts, floods, and elevated temperatures. The prediction skill of a climate model depends crucially on the fidelity of these carbon cycle responses built into the model. However, for the future land carbon uptake these models do not even agree on the sign of the response. Natural climate variations such as the El Niño Southern Oscillations (ENSO) provide us with experiments to evaluate and improve our understanding of these carbon cycle responses.

ENSO is a climate pattern that involves periodic oscillation in winds and sea surface temperatures over the tropical eastern Pacific Ocean. It represents a major control on the year-to-year variation in temperature and precipitation in the Tropics, going through its El Niño and La Niña phases in an irregular fashion that is still difficult to predict but repeating in roughly four-year cycles on average. Less known is the control of ENSO on the atmospheric chemical composition. The global abundance of several gases including greenhouse gases of the natural atmosphere, such as CO2 and CH4, show a clear relation with ENSO. In the case of CO2, both the land and the ocean contribute to this variation in ways that are not well quantified yet, making ENSO an excellent test for models. Closest to the imagination are probably pictures with vague contours of Indonesian farmland covered in thick smoke during El Nino. Indeed, fire is an important mechanism connecting precipitation variability to CO2 variability.

Figure 2 shows the carbon cycle response to the ENSO cycle, as manifest in the atmospheric mole fraction of CO2. Atmospheric CO2 (as well as other greenhouse gases such as CH4) is measured cooperatively by multiple laboratories at a global network of sampling sites, an effort that began more than fifty years ago with Mauna Loa and South Pole. Our knowledge of the carbon cycle response to ENSO – such as the amount of additional carbon in the atmosphere during a strong El Niño, or the partitioning of that signal into contributing factors such as fires in Tropical Asia versus drought in Amazonia – derives to a large degree from these measurements.

Figure 2: The (CO2) growth rate as measured by the increment of one month over the same month in the previous year. The inter-annual variations in the (CO2) growth rate show a clear imprint of ENSO. Large El Niño events such as the 1997-98 and 2015-16 ones show up as anomalously large spikes in the growth rate. Most of the additional (CO2) in the atmosphere during and after an El Niño comes from the Tropics, and therefore the response measured at Mauna Loa (red) is usually larger than the global average response across a network of background sites (blue). (CO2) data taken from the NOAA (CO2) trends page at

The atmospheric growth rate of the CO2 mole fraction spikes right after a big El Niño event, such as after 1997-98 and 2015-2016 in Figure 2. Since we know the total mass of air in the atmosphere, we can translate between CO2 mole fraction spikes of Figure 2 and mass of carbon added to the atmosphere. The 1997-98 El Niño added ~2 Petagrams carbon (PgC) to the atmosphere, while preliminary estimates suggest that the more recent 2015-16 event injected ~3 PgC (for comparison, our current global fossil fuel emission is ~10 PgC/year). Having a global network of sites – instead of just background sites such as Mauna Loa – allows us to drill down into the mechanisms behind each of these CO2 increments. For example, we know now that most of the additional 2 PgC carbon from the 1997-98 El Niño was injected from extended fires in the Tropics. Due to the sheer magnitude of the carbon cycle response to El Nino, with the year 2015 setting the record in global CO2 growth to just above 3 ppm/yr, ENSO events present natural experiments against which we can verify our understanding of the interaction between climate and the carbon cycle.

Even for the large changes in CO2 as observed during strong El Nino’s the attribution to specific processes remains a challenge, because of the various coupled responses in the Earth system. For example, the ocean-atmosphere exchange of CO2 is also influenced by ENSO, as shifting patterns in tropical sea surface temperature change the mixing rate of deep and surface waters, influencing gas exchange. Therefore, to get the process-attribution correct, scientist try to disentangle the various influences on atmospheric CO2, which requires a lot of measurements.

Over the past couple of decades, it has become clear that our cooperative network of atmospheric measurements has large gaps over areas that play very important roles in determining the climate impact on the carbon cycle. The gaps are usually due to logistical reasons, such as the difficulty of maintaining measurement sites in Tropical forests, or the expense of making regular shipboard measurements to cover the oceans. To fill this data gap, space based measurements have emerged as a promising yet challenging alternative.

Carbon cycle gases such as CO2 and CH4 (and to a lesser extent CO), by virtue of being greenhouse gases, absorb and emit electromagnetic radiation at certain specific frequencies, usually in the infrared (IR). In theory, a downward-looking IR sensor at the top of the atmosphere should be able to estimate the amount of these gases by measuring the strength of IR radiation at those frequencies. Figure 3 shows the average CO2 mole fraction between the surface and the top of the atmosphere (“column average CO2”) retrieved from the Orbiting Carbon Observatory 2 (OCO2) satellite over Equatorial Africa, showing elevated values due to biomass burning.

Figure 3: Column average CO2 over Central Africa in 2015 from the Orbiting Carbon Observatory 2 (OCO2) satellite. The bright red band over Equatorial Africa due to biomass burning is visible in contrast to lower (CO2) elsewhere. Column average CO2 from the ACOS algorithm available at

In practice, IR measured from space is sensitive to many interfering species other than CO2 or CH4, such as the amount of water vapor and dust particles, complicating the estimation of CO2 and CH4 from space. If these complications can be resolved in the near future, space-based measurements could potentially fill the data gap between surface measurement sites (such as Equatorial Africa, which has almost no surface measurements) and enrich our knowledge of the carbon cycle. With such improvements to our measurement capabilities, we hope to better understand today’s carbon cycle and its response to climate. The clearer we see this carbon-climate tango, the better we will be able to predict its imprint on tomorrow’s climate. The most recent climax was one of the most well observed in history. Data are pouring in from multiple sources, and the coming few years promises a lot of interesting analysis as we try to decipher the steps of this complicated dance. Keep your eyes peeled for updates on this blog!

This post has been written by:

Dr Sourish Basu,  NOAA Earth System Research Laboratory, USA
Dr Sander Houweling, SRON Netherlands Institute for Space Research, NL

and edited by the new editor of this blog Célia Julia Sapart, Université Libre de Bruxelles, B.

EGU, Vienna 2015: the round-up

EGU by numbers

In April, the EGU returned to Vienna for their annual Congress meeting. Over 11,837 scientists from 108 countries descended in the Vienna International Centre for the six-day conference. Delegates enjoyed over 4,870 oral presentations, 8,489 posters, and 705 PICO presentations. That’s a lot of science!

The Vienna International Centre – where the magic happens


Science and ice cream for everyone!

As always, the EGU were excellent hosts and organised over 300 side events that included a range of activities, workshops, and networking events for all delegates. Among this year’s favourites were the Geo Cinema, Science Communication workshops, and Ask the Expert panel sessions. Another added bonus this year was the ice cream stall in the foyer – my personal favourite was the pumpkin seed flavour (tastier than it sounds, I promise).


With 23% of delegates being students, and a strong Early Career Scientist contingent, the Young Scientists’ lounge proved very popular, and provided a great social space to grab a coffee, talk to friends, finalise PowerPoint slides, and peruse the latest job adverts.


The exhibition hall where delegates can peruse the latest journals, textbooks, and equipment. And buy their ice creams!

The exhibition hall where delegates can peruse the latest journals, textbooks, and equipment. And buy their ice creams!


The best bits of Climate: Past, Present and Future

The Climate division hosted over 75 scientific sessions, covering climate science research from the poles to the tropics. We also enjoyed some excellent workshops including an ‘Introduction to climate modelling’ session and a ‘Meet the Editors’ panel discussion with Professor Carlo Barbante (Editor of Climate of the Past) and Professor Axel Kleidon (Editor of Earth System Dynamics).


Thanks EGU and see you next year!

Amid the scientific sessions and workshops, the EGU is always a great place to catch up with old friends and colleagues, and make new collaborations. There is also time to sample the delights of Vienna, while discussing research projects over a Wiener Melange. After another impressively productive and enjoyable conference, we are all looking forward to EGU Vienna 2016!


Get every new post on this blog delivered to your Inbox.

Join other followers: