The COVID-AM blog is a partnership between the UMI 3157 iGLOBES and the Institut des Amériques, coordinated by François-Michel Le Tourneau, Deputy Director and Marion Magnan, researcher at the Institute. About the blog.

Sentinels and tipping points: two concepts to understand the pandemic better

September 15, 2020

by David Blanchon, Professor in geography at Université de Paris Nanterre and researcher at iGLOBES at the University of Arizona. 



On March 24, Mario Cuomo, governor of the State of New York (19 million people), stated that "New York is the canary in the coal mine. New York is going first. We have the highest and the fastest rate of infection. What happens to New York is going to wind up happening to California, and Washington state, and Illinois, it's just a matter of time. We're just getting there first"

He was hoping that in his position as sentinel (comparable to that of the canari which the miners used to detect the presence of toxic gas), and with the experience he had gained, the State of New York would enable other states to take measures that would reduce the spread of the epidemic when it arrived. That was not the case.


Despite the 32 000 confirmed deaths in New York, the epidemic flare-up of Covid-19 the following summer in the three most populated States in the United States (California, Texas and Florida with respectively 39, 28 and 21 million people) shows that they did not learn a lesson from the New York experience. And to this day, over 190 000 Americans have died from Covid-19, a number that is increasing by about 1 000 deaths per day.


On a larger scale, the manner in which the Covid-19 epidemic spread between the US federated States, showing a much greater timing interval than between the European nations, should lead us to think about the role of "sentinel" territories first, which could help us be better prepared, and secondly the existence of "tipping points", both health and political, which seem to be elements of this epidemic.

t to a spatial spr
Figure 1 The spread of the virus in the United States on 3 different dates (04/1, 07/1 & 09/1).



Faced with the risk of pandemics, the WHO put in place a monitoring network, long before Covid-19, in territories where the inter-species transmission of viruses could take place. Their role is to alert by enabling others to prepare. As F. Keck wrote (Dictionnaire Critique de l’Anthropocène, 2020 - pandemic article), "the pandemic rests on sentinel reasoning and early warning signals. Instead of being a radical change that has already taken place and whose traces are visible on earth, the pandemic warns of a disaster to come, but whose environmental conditions are already present. Being prepared for future pandemics becomes an essential element in community living in the Anthropocene era: significant environmental disruptions are revealed by the rise of new viruses". 



In Figure 1, the "canaris" in April - essentially New York and its region - then the spread to Florida, Texas, Southern California and Arizona three months later. The "restart" of the epidemic in the US is not due to a "second wave", but to a spatial spread.


Yet, faced with COVID-19, all the sentinels failed. The town of Hong-Kong, for example, had had this role and had sent a warning about SARS COV-1 and the virus H5N1 flu. But it was far from the main spread of SARS-CoV-2 outside of China. The other laboratories specialized in detecting emerging diseases, dispersed globally, also did not notice the weak signals of this new virus' appearance, that was probably in circulation before its first epidemic breakout in Wuhan[1].


At a more local level, territories where the virus was widely circulating were not seen as having had early signs of what could happen in the rest of impacted countries. In large federated states, such as the United States and Brazil, some areas particularly hit at the beginning of the epidemic were not considered as "sentinel-territories", signaling a future threat for the rest of the country, but as insignificant unique cases. If these territories had been identified as the "mine's canari", it would probably have led to preventative measures to limit the epidemic. The failure of the sentinels accounts for the breadth of the epidemic in Brazil and in the United States.



Figure 2. Navajo Nation curfew on Memorial weekend.
Figure 2. Navajo Nation curfew on Memorial weekend.

The first reason sentinels failed could be linked to the virus' "political environment", which reaches marginalized communities in a disproportionate fashion, in all sense of the word: by their geographic location in the spatial margins of the States (Native American reservations), by their racial profile (Afro-Americans), by their social status (poor, emigrants...), even by their age (residents of retirement homes). The signals sent by these sentinels were not recognized as pertinent - or at least not until very late - by a large portion of the "majority" community. One simple example, while the virus was actively circulating in Northern Arizona and the Navajo reservation was on a strict lockdown due to the increase in mortality, governor D. Ducey lifted the restrictions in the rest of the State as early as May in favor of reboosting the economy in the more populated counties (and specifically in the cities of Tucson and Phoenix). The three day Memorial weekend (May 25-27), which marks the end of the school year in Arizona and the start of summer vacation, took place as usual, with many traveling to tourist spots and large family gatherings. This "return to normal" seems to have played a role in the epidemic's rise in the months of June and July. The Native Americans' situation, which could have played a sentinel role, was not taken into account in the rest of the State.

Figure 3 Covid-19 in Arizona (Source: AZ Dept of Health). The worst hit counties in total numbers are Pima (21 000 cases) and Maricopa (134 000 cases). But in percentage of population, it's the Northeast counties where Native Americans live.
Figure 3 Covid-19 in Arizona (Source: AZ Dept of Health). The worst hit counties in total numbers are Pima (21 000 cases) and Maricopa (134 000 cases). But in percentage of population, it's the Northeast counties where Native Americans live.



A second hypothesis to explain the failure of the sentinels where COVID-19 is concerned seems to be the nature itself of the epidemic, which articulates around two antagonistic states of equilibrium separated by tipping points that are difficult to determine. This process, well documented by environmental sciences, rests on the idea that a system can have two very different states of equilibrium and that the shift from one to the other is not gradual, but abrupt. These "tipping points" are defined as: "Point or threshold at which small quantitative changes in the system trigger a non-linear change process that is driven by system-internal feedback mechanisms and inevitably leads to a qualitatively different state of the system, which is often irreversible. This new state can be distinguished from the original by its fundamentally altered (positive and negative) state-stabilizing feedbacks." (Milkoreit et al. 2018 )[2]

Figure 3 Tipping points
Figure 3 Tipping points

According to Sheffer (& al.): “Early-warning signals for critical transitions”, Nature, Vol 461j3 Sept 2009jdoi:10.1038/nature08227. The figure shows the difference between linear (a) transitions, gradual with turning point (b), catastrophic with tipping point between the 2 states of equilibrium (c) & (d).

In the case of the COVID-19 epidemic, the two states in question would be a weak circulation on one end, with a few localized clusters - Arizona's situation for example in May 2020, and on the other end a generalized epidemic leading to a high mortality rate requiring drastic measures of confinement to be contained. Between those two states, there is no gradual transition, but tipping points, with few early warning signs.


A team of researchers, with the support of the University of Arizona, studied the origin and the spread of the virus when the first infections happened in Europe and the United States. It showed that "the value of detecting cases early, before they have bloomed into an outbreak, cannot be overstated in a pandemic situation". In their opinion, anticipating the tipping point, even by a few days or weeks, coupled with a monitoring system and preventative measures, would really help confront a new epidemic outburst.


The role of sentinels, essential in this type of situation, is then extremely difficult because the system's behavior is very difficult to modelize, and the risk of a "false alarm" is very high. This is very evident in the arguments on the coming of a "second wave" in European countries.




So the failure of the sentinels does not come from them, but rather from the numerous political leaders' inability to perceive the signals they are sending. Some deliberately chose to ignore the experts' advice at the beginning of the epidemic, such as D. Trump and J. Bolsonaro. Others simply were unable to see them. We can certainly recognize how incredibly difficult it is to manage such an epidemic. With tipping points that are hard to predict in systems with two unstable equilibrium states, making the right decision at the right time is a formidable challenge. 


In this, the COVID-19 crisis is much closer to climate issues for which it is also possible that there are two very different equilibrium states, separated by tipping points - extremely hard to determine and to predict -, and not a gradual progression. In this type of situation, it is extremely difficult to determine the position of the tipping point and consequently to predict the transition into the other state.


And just as in the current pandemic, the messages from the sentinels can seem too far off (for example the fires in oriental Siberia and the collapse of the ice cap), or touching populations that are too different (for example the livestock farmers on the Horn of Africa and the inhabitants of Bengladesh and the Maldives), to be taken seriously before it is too late.


Political courage, as governor M. Cuomo saw, consists in anticipating the tipping into a new irreversable state and reacting before it is too late by observing the sentinels' warning signals.

[1] According to Lucy van Dorp and al., the virus would have already appeared in October 2019 and could have been in the United States in December 2019. Lucy van Dorp, Mislav Acman, Damien Richard, Liam P. Shaw, Charlotte E. Ford, Louise Ormond, Christopher J. Owen, Juanita Pang, Cedric C.S. Tan, Florencia A.T. Boshier, Arturo Torres Ortiz, François Balloux (2020) : “Emergence of genomic diversity and recurrent mutations in SARS-CoV-2, Infection”, Genetics and Evolution, Volume 83, 2020.

[2] Milkoreit M., Hodbod J., Baggio J., Benessaiah K., R Calderón-Contreras R. Donges J., Mathias J.-D.,Rocha J.-C., Schoon M.,Werners S. (2018): “Defining tipping points for social-ecological systems scholarship—an interdisciplinary literature review”, Environmental Research Letters, 13 (3), 033005.

David Blanchon is a Professor in geography at Université de Paris Nanterre and researcher at iGLOBES at the University of Arizona. He recently published a book on the geopolitics of water, Géopolitique de l’eau : entre conflits et coopérations, Editions le Cavalier bleu, Paris, 168p., 2019 and was one of the authors of Dictionnaire critique de l'anthropocène, CNRS Editions, June 4 2020. For more information on the topic, see article Sentinel Territories: A New Concept for Looking at Environmental Change, Metropolitics, May 8 2020.