What do Manaus and Guayaquil tell us?

In Manaus (Brazil) and Guayaquil (Ecuador) we see large corona outbreaks. These are locations where it was/is warm and there is often high humidity. These two examples are regularly used to show that what I (and many others) say about the effect of high humidity on the spread of the virus cannot be true. But precisely those statements (unfortunately not only made by amateurs, but also by virologists and epidemiologists) show that one doesn’t have a clue about how the virus actually spreads, so one doesn’t know how to fight it in an intelligent way and suffices with repeating mantras like “keep a 1,5 meter distance” and “wash your hands well” and “mouth guards only give false safety” and “beware of the second wave”.

  • A pattern in which the flu breaks out once a year (in winter). We see this above 30 degrees north latitude, where the areas with temperate weather (i.e. also the area where we live) are locate
  • The spread pattern of COVID-19 shows a clear resemblance to that of influenza, and a lot of study has been done on this. When you look at the scientific studies of the spread of influenza around the world, striking patterns can be identified:

In tropical countries we don’t see that pattern. There are sometimes two peaks (but not in winter) or a more even distribution. A recent article from November 2019 in Popular Science describes this well.  And also this Australian article describes that. Close to the equator we see a very different pattern with regard to influenza than we know in our region.

In all those years scientists have not found a good explanation why that pattern is so different in those tropical regions. I do have a suspicion, but that is very speculative and not relevant at the moment.

Both Manaus and Guayaquil are very close to the equator (2 and 3 degrees south latitude).  In both cities there has been a super spread event, which I will come to in a moment. So it seems that COVID-19 also shows a different spreading pattern close to the equator than in temperate weather areas. For the same – seemingly unknown – reasons that influenza does!

And so the developments around the equator cannot be used to apply those one-on-one to the patterns in our areas. In the Netherlands we have to focus on the patterns we see in areas above 30 degrees north latitude, and not on areas with other patterns.

Having said that, I like to address the big outbreak in those two cities near the equator. Because even there it is clear, that super spread events triggered an exponential increase in infections.

If we look at Manaus and Guayaquil, we see large-scale carnival events at the end of February. This is a picture of Manaus on the 22nd of February. More or less similar to the pictures from other places.


  1. that the first case was discovered on February 29th in Guayaquil (!), in a woman who had come from Spain.

In our northern regions, it seems unlikely to me that the large outbreak occurred during carnival parades outside. Or the City-Pier-City run on March 8th in the Netherlands, or the Feyenoord-Willem II soccer match caused those. It is much more likely that if there were large-scale outbreaks, they happened indoors, with many people in confined spaces like clubs, party halls, bars, etc. (as has been the case on winter sports in the big clubs in Sankt Anton. Here’s a report from an Amsterdam group that was there). 


But now one can object that there have been carnival celebrations all over South America, without any major outbreaks there. And that exactly hits the core of what you can learn from these two outbreaks in South America.

  1. First of all, the condition of the infection is of course that there was at least one person present who was infected and didn’t know it himself. Just at the end of February, the virus hadn’t spread that massively yet, so you had a very good chance in South America to be present at major events, without getting infected there. In Brazil it is indicated that the first infected persons were rich Brazilians who had been on winter sports in Northern Italy. In Ecuador it is said  that the first case was discovered on February 29th in Guayaquil (!), in a woman who had come from Spain.
  2. In our northern regions, it seems unlikely to me that the large outbreak occurred during carnival parades outside. Or the City-Pier-City run on March 8th in the Netherlands, or the Feyenoord-Willem II soccer match caused those. It is much more likely that if there were large-scale outbreaks, they happened indoors, with many people in confined spaces like clubs, party halls, bars, etc. (as has been the case on winter sports in the big clubs in Sankt Anton. Here’s a report from an Amsterdam group that was there).If you see the development of the number of infected people shortly after the outbreak, it seems to have started during the carnival there. But take a look at this map of that city and the many clubs in the center, including some large ones. People are partying in there and I wouldn’t be surprised if the airco’s are working very hard. I don’t think that’s much different from the party hall in Kessel with 500 people on March 5th. (Except of course the music and the clothes they wore). But it could be that the unique situation close to the equator (near the 37 degrees and high humidity) in the evening, the aerosols stayed airborne for a long time.

If you look at the development in the figures of Manaus you see that there was no outbreak during the Carnival in February. The super spread event must have taken place later, between 10 and 15 March, because that’s when the outbreaks started. During parades, church services or feasts. And also in Manaus there are a number of very large clubs.

It’s a pity that in the decades of research into influenza, so few hard explanations have been given for the different patterns of spread. If we had, we would have been better off fighting the spread of COVID-19.

But there is something interesting about studying COVID-19. Because nobody was infected yet, the spread is much easier to investigate. The pattern of influenza spread was (strongly) obscured by the fact that a not insignificant proportion of people were already immune to the “flavor of the year” of the influenza virus. This is not the case with the coronavirus.

You would think (and hope) that this is an ideal opportunity for good scientists to study and explain the spread patterns. This will also help us with future new influenza outbreaks.

And if that’s going to happen, I’m sure much of what I’ve written in my blogs over the past 6 weeks will be confirmed.

We can do so much better, it has to be so much better

This will be a short blog. Maybe that will help the media and the general public to understand my findings better. I’m not a virologist and I don’t pretend to be one. However, I am academically educated and able to assess scientific research.  My background as a social geographer and my experience with data and statistics help me to arrive at well-founded insights into the COVID-19 crisis.

I have long argued that aerosols indoors play an important role in the spread of the virus and that the risk of infection is significantly lower outdoors. This means that 1.5 meters distance indoors is not sufficient under certain circumstances, but also that 1.5 meters outdoors need not always be necessary (effective if possible, but not necessarily).

I have also long argued that worldwide evidence shows, that so called super spread events played a very important role in the exponential rise of the virus at the beginning of the outbreak. Because there are hardly any such events in the world anymore, we see a strong deflection of the exponential growth everywhere and there are no large second outbreak waves. Unfortunately, it is not yet understood that aerosols also play a role in the spread within healthcare institutions, as I communicated to responsible authorities 5 weeks ago. As a result, situations have sometimes arisen in which virtually everyone within these institutions became infected.

Instead of believing me on my word, I advise you to read this article dit artikel te lezen by a professor from Colorado, who gives the best description of aerosols I have read so far. Very accessible and very balanced. See the end of this article.

Also read this summery of a recent Japanese study:



RIVM and adjoining experts, who regularly appear on TV, think that the role of aerosols is minimal. They ignore my input from relevant scientific sources. What’s more, it leads to my findings being marginalized in the media. Time and again, what I have said appears to be confirmed by new studies and new examples. (Like the risks I described of the virus spreading among workers in the meat processing industry, that have now materialized in the Dutch city of Velp).

I find this interview last Saturday with the Dutch governments Corona fighters Van Dissel and Wallinga staggering. After two months we only hear that the is still not enough information to design an exit strategy. Even the most basic information about how many percent of the population is infected appears to be lacking. Also the calculation of the R0 is not based on hard information and is partly based on assumptions that have not been made public. Nevertheless, much more is already clear. It is only high time that all relevant scientific insights are included in the considerations for policy and measures.

As long as the information about the aerosols isn’t taken seriously, RIVM and virologists are unnecessarily pushing us into the “new normal” of the so called “1.5 meters society”. In addition, they ignore a source of contamination that may be relevant when schools reopen soon, with all its possible consequences.


Aerosols, the crucial missing link

This is really the most crucial question in the fight against the spread of the virus. The answer will have a huge impact on developments, not only in the coming months, but also in the years to come. Are aerosols playing an important role in the infection with the virus, or not? Lees meer

Meat processing industry: superspreading hot spots

It struck me for the first time a month ago: in the US, workers in the meat processing industry went on strike because COVID-19 had broken out among a number of colleagues. That turned out to be only the beginning, as we now see large numbers of infected workers in the meat processing industry outside the US as well. They form the base for outbreaks in the area around those factories.

There are two components that make these kinds of factories superspreading hotspots. First of all: if you look at photos of those factories, you can see that people are working very closely together.

But I don’t think that’s the only reason that makes COVID-19 strike in meat processing factories. The second reason seems to be that in those factories the temperature is usually kept a bit lower. Not only in storage areas, but also in the rest of the factory. And that creates favorable conditions for micro drops to stay airborne longer and thus contaminate more employees. It is also known that in a colder environment the virus can be found for a longer period of time on, for example, metal. Whichever of the three causes, in combination with each other, every meat processing plant has the risk of becoming a superspreading hotspot.

As a result, in areas of the US where COVID-19 had not yet really broken out and the weather conditions were actually unfavorable for a major outbreak of the virus, it still happened because of the infected workers in those factories.

With the help of Google I came across these outbreaks in the vicinity of factories like this in North America.

If you read the messages, there also seems to be a connection with the type of employees that work in those factories and the way some employers deal with sick people. (Let them come to work anyway).

But also in other countries we see in the media that there are outbreaks at slaughterhouses and meat processing factories:

  • South-Brazil
  • Ireland, waar in het parlement is gemeld dat er verschillende onbekende hotspots in vleesverwerkende fabrieken zouden zijn
  • Birkenfeld, Germany waar veel Roemeense gastarbeiders besmet zijn


In the rest of Europe and in the Netherlands I haven’t really been able to find the kind of outbreaks like in the US. But I don’t know if that means it’s less the case in Europe, or if it’s because it hasn’t been recognized yet that these kind of factories could be hotspots of COVID-19 contamination. Especially in areas where there are already a lot of infestations anyway, it might not be noticed.

This is a website where you can find the locations of those factories in the Netherlands.

It wouldn’t surprise me if this is also the case in Europe. Maybe you know of examples of outbreaks in and around those factories?

What do the numbers say. Are we doing well or not?


In the past few days we heard in the various talk shows on TV, that the R0 (reproduction factor) is on the rise again towards 1. That gives you the feeling that “we” are going in the wrong direction again, despite all the strict measures,.

But if you delve a bit deeper into the actual data, you see on the one hand how shaky that observation is, and on the other hand, that this increase, if it really would be there, doesn’t mean much at the moment.

So let’s look at the figures in the right perspective.

  1. It strikes me that the RIVM does not mention the Reproduction Number (R0) in its daily reports. It only appears in their weekly briefing.
  2. R0 is an indication of how many people are infecting other people. So if a total of 100 people are infected and they infect 100 others within a certain period of time, then R0 equals 1. The R0 of Covid-19 in the Netherlands was well above 2 in mid-March, but has been decreasing since March 21st (I think mainly because meetings with many participants are forbidden, so the so called super spread events didn’t take place anymore).
  3. It is difficult to determine the exact R0. In the first place because we do not know the real number of infections (Based on international research, such as this one, we estimate that the real number is a factor of 50 higher than the number of registered patients). It is therefore a derivative of other figures. Which one the RIVM uses I would not know (I assume hospitalizations or deaths).
  4. The big problem with the known figures of our hospital admissions/deaths is that we don’t properly register where those victims come from. Let’s assume that half of them come from healthcare institutions. What happens there is actually independent of what happens outside those institutions. There may therefore be an increase in hospital admissions/deaths from care institutions, while the R0 in the rest of society continues to fall. So if the inflow of sick people from institutions and fatalities are also included in the calculation of R0, the R0 doesn’t really say anything about the situation in the rest of society anymore.
  5. Last but not least. Look at the development of the hospitalizations below.  At their peak (end of March) more than 500 a day.  That has now dropped to less than 100 per day. The number of registered deaths peaked about 1 week later, with about 170 deaths per day. The current percentage of deaths compared to the peak is 30%. Considering the usual delay, we have to look at the hospitalizations from a week ago. That number was less than 20% of the peak a week ago.  So based on the registration of the deaths, you’d say it’s going slower than the hospital admissions show. Which of the two gives the right picture?  And which of these does the RIVM use?

Because there are clear problems with both figures. It can go up or down without actually doing so.

Suppose, for example, that more and more people do not want to go to hospital with Corona complaints. That could explain the stronger decrease in the number of hospitalizations. Or the other way around.

With regard to the mortality rate, we already know for sure, that something important has happened.  Statistics Netherlands calculated that more than twice as many people have died from Corona than the RIVM has registered. Similar differences can be seen in many countries (This is an overview from the New York Times).

Since April 10th, people have been trying to do their best to better register the correct number of Corona deaths. Maybe this will increase the number of deaths a bit. But in short: the development of the death rate is not an undisputed number either. So there are still some uncertainties in the hardness of the figures themselves.

But even if the figure is very hard, it is important to know which part comes from healthcare institutions and which part does not. Perhaps by now the number of deaths has mainly come from healthcare institutions, whereas at first that contribution was less?

But even if the R0 would have really risen back to around 1, that would mean that we at worst would end up in a situation, where the number of newly infected people would remain stable and would therefore by no means create a situation with strong growth again. That risk will probably only arise again in the autumn.

Finally. On average people spend 3 weeks in an ICU.  Three weeks ago about 70 persons a day were placed in Dutch ICUs.  These people will now flow out on average per day. The influx at this moment is around 20.

Based on this, it can be assumed that the decrease of patients in ICU’s per day will be 40 to 50 per day in the coming days. So on May 1, 2, or 3, we will drop to a figure of less than 700 persons in our ICU’s. And in three weeks that number will be close to 500 if the number of new ICU admissions stays below 20 per day.

Even if the R0 will be around 1 again in the whole of the Netherlands, there will be no significant increase in the number of people in the ICU’s, at least until the autumn.  Not even if we expand the measures considerably.

It wouldn’t surprise me if in the next 2 months, the number of patients in the ICU’s will still be well below 500.

When are we going to start working on that other important patient who is also in serious condition in the ICU: the Dutch economy and society? The situation has only got worse in the last three weeks.

The government is pointing in the wrong direction

In the advice of RIVM and the explanations by specialists in the talk shows, it strikes me how little attention is paid to important new scientific research. That is why I made this short video. No extra words are needed.

Swedish approach costs 30 more deaths per day

For everyone in the world who is involved in the search for the right COVID-19 virus strategy, Sweden is super interesting. There they have not taken as hard measures as in many other countries (e.g. The Netherlands). Lees meer