Where do oceanographers study?
The part of the ocean that oceanographers choose to study is determined not just by scientific interest but also by geo-political factors. In this section we see how JPO started out with studies focused on the areas of most interest to North American and European nations before on one hand becoming more global while also showing a shift to areas of most interest to Asian nations. While the Indian and Arctic Oceans have been studies more regularly they remain marginal seas overall. We also discover an exciting new ocean basin that has gradually emerged in the pages of JPO over the decades.
We first divide the global ocean into the main basins with their associated marginal seas. We look for references to locations in these basins to understand the overall distribution. For example, any reference to the Labrador or Mediterranean Seas is counted towards the Atlantic.
The Atlantic and Pacific Oceans have been the setting for the most articles in every year bar one. The proportion of articles set in each of the Atlantic and Pacific have been similar in most years at the start and end of the series. In the middle part of the series, however, the Atlantic was the most studied basin with up to 40% of articles being set there.
Inspection of article titles show that the topics studied in the Atlantic and Pacific are diverse. The main difference is perhaps that equatorial studies are much more common in the Pacific than the Atlantic.
The Southern Ocean was not a major focus of research in JPO in the 1970s and 1980s with around 5% of articles referencing the Southern Ocean each year. From 1990 onwards there has been a steady increase in interest until the recent years when there has been a similar proportion of articles on the Southern Ocean as on the Atlantic and Pacific Oceans.
Around 10% of articles in the first decade of JPO were based on lake research - principally the American Great Lakes. These studies subsequently declined to less than 1% of the total. This decline is presumably due to the creation of more limnology journals and the increasing specialisation of both limnology and oceanography. Estuaries (not shown in plot) are typically referenced in 1-5% of articles per year, with values typically in the upper end of that range in the last decade.
My expectation had been that the Arctic and Indian Oceans had become hot topics in recent years and I was expecting a big increase in studies on these regions. The reality has been somewhat different. The Indian Ocean accounted for less than 5% of the articles up to 1990 and 5-10% of articles in the period since. Typically about 2-3% of articles published per year referred to the Arctic Ocean until around 2010 which has since increased to around 5-6% of the total each year.
How many oceans?
JPO has always contained articles that are specific to a particular location (e.g. “A Determination of Horizontal Divergence in the Gulf Stream off Cape Lookout” in 1971), articles that relate to a type of region (e.g. “Zonal Penetration Scale of Model Midlatitude Jets”).
We can compare the number of articles that refer to no locations compared to articles that refer to one or more than one basins. In the first 25 years 40-50% of the articles had no specific location. From the late 1990s onwards this proportion decreased to 30-40% of the total. This decrease may relate to a change in how oceanographers motivate their studies to be explored in a later part.
I anticipated a rise in the number of articles that refer to multiple basins over the decades as global datasets and numerical models became available. The number of such articles has in fact doubled from around 6% to 12% of the total in recent years (though it is still less than I expected). However, the number of multi-basin studies is almost certainly undercounted here as studies that look at globally-integrated quantities (e.g. “A New Climatology of Air–Sea Density Fluxes and Surface Water Mass Transformation Rates Constrained by WOCE”) are harder to capture using string-matching methods.
A new ocean
In addition to the classical oceans a new ocean basin has developed in JPO over the decades whose rise to prominence has been just as meteoric as that of the Southern Ocean. This new ocean is the Idealized Ocean.
The word “idealized” did not appear in any articles during the first four years. While usage was bumpy in the first decade, from the late 1980s onwards the Idealized Ocean became more and more studied until almost 20% of articles referred to it in 2017. The Idealized Ocean differs from the traditional oceans in a number of ways: an f-place assumption is typically valid and the wind is typically quite steady. It is surpising that the Idealized Ocean tends to feature primarily in numerical studies - the absence of surface waves and dangerous boundaries in the Idealized Ocean would seem to make it ideal for observational studies.
We can break the analysis above down into sub-basins e.g. North Atlantic and South Atlantic. For the Southern Ocean we divide it into the open ocean Southern Ocean and the Antarctic marginal seas. Unsurprisingly we find that the North Atlantic and North Pacific dominate the contribution of the Atlantic and Pacific respectively.
More interestingly, we find that the rise in articles on the Southern Ocean is divided relatively equally between the open ocean Southern Ocean and the Antarctic marginal seas though with the former tending to have 1-2% more articles per year. No clear trends are apprent in the other sub-basins.
Many of these geographic studies focus on the major current systems. The Gulf Stream is the most studied current, with references to the Gulf Stream in 6% of JPO articles. This is followed by the Kuroshio and Antarctic Circumpolar Currents with about 3% of articles referring to each of these current systems. The Agulhas is referred to in 1% of articles.
The other major group of currents referenced in JPO are the various equatorial currents. Equatorial currents and undercurrents are found in the Pacific, Atlantic and Indian Oceans. However, it is challengiing to determine exactly which one is being referred to with simple string-matching, so I have lumped them all together here. Inspection of the articles suggests that the majority of the articles refer to the equatorial currents in the Pacific.
The number of articles devoted to other currents is very small. The 14th most referenced current is the Somali current that is referred to in 0.2% of articles - this is just 15 articles over 50 years. As such estimating trends is only feasible in the most-referenced currents.
The Gulf Stream was the primary focus for the first half of the time series with 7-10% of articles each year. From 1994 until 2010 there was a steady decrease in the number of articles that refer to the Gulf Stream to less than 2% of articles in 2010. In recent years 3-5% of articles have referred to the Gulf Stream.
The progress in oceanography over the 50-year period is apparent from the titles of articles related to the Gulf Stream. Taking studies of the eddy activity around the Gulf Stream for example in the early years there are titles such as “Life Cycle of a Gulf Stream Anticyclonic Eddy Observed from Several Oceanographic Platforms” which present the first observations of eddies shed by the current. In the middle years we have studies such as “Eddy–Mean Flow Interaction in the Gulf Stream at 68°W” looking at the interaction of the eddies and the current while in recent years we have studies such as “Warm Spiral Streamers over Gulf Stream Warm-Core Rings” looking at the submesoscale dynamics through the eddies.
The Kuroshio has gained in interest through most of the period. In the early years it was referred to less than 2% of time - which corresponds to 1-2 articles per year in that period - to about 5% of articles more recently.
The most dramatic change in interest has been in the Antarctic Circumpolar Current (ACC). In the first six years of the journal a circumpolar current was only referred to twice in the context of global ocean simulations. From 1987 onwards the references to the ACC undergo almost continual increase until a peak of about 6% of articles in 2014. The range of topics expanded from the initial focus on hydrography (e.g. “Structure and Transport of the Antarctic Circumpolar Current at Drake Passage from Short-Term Measurements”) to the controls of the ACC (e.g. “Equilibration of the Antarctic Circumpolar Current by Standing Meanders”) and its role in ocean mixing (e.g. “Direct Estimate of Lateral Eddy Diffusivity Upstream of Drake Passage”).
In the next post I will look at how oceanographers have carried out their studies looking at the rise and fall of different obseravational technologies and the rise and rise of numerical simulations of the ocean.