Hacker News Reader

Comment by wildmusings | Hacker News Reader

To preface, I am inclined to agree with the anthropogenic climate change hypothesis. But the claims in this article are no better than climate change skeptics pointing to a particularly cold winter or mild summer as evidence for their position.

How can you possibly conclude this is caused by climate change vs. being a 100, 200, or 1000 year storm? Freak weather events have been occurring for billions of years before the industrial revolution.

Climate change is a gradual process that may in fact make such storms more likely, but pinning any single storm on climate change is suspect.

View on HN

"The flooding on Kauai is consistent with an extreme rainfall that comes with a warmer atmosphere," said Chip Fletcher, a leading expert on the impact of climate change on Pacific island communities.

Because no one can know the entire body of knowledge, we divide our understanding of the world among us. Some of us study the earth and climate and accumulate expertise on the subject matter.

“Consistent with” does not establish much at all, other than that the event does not disprove climate change. An event tells us nothing about climate change if it is consistent both with and without climate change.

My withdrawing $40 from my bank account is consistent with me having no cash at all left in my wallet. It’s also consistent with me already having some cash but needing more. The event, the withdrawing of the cash, does not tell us whether my wallet was empty.

What is the goal of your comments? Are you trying to convince people that when there is an increase in anomalous weather patterns we shouldn’t say it’s due to climate change? At what point would you say these things are a result of climate change? Do you deny the reality of climate change? Do you agree that it is a reality but are fighting against people ignoring the fact that an individual event might have a different cause?

Due to humans releasing massive amounts of greenhouse gasses into the atmosphere there is more energy in the system. Clearly this increase in energy leads to anomalous behavior. It’s ok to point this out.

I think you can look at any single (weather) event and conclude that "extremes happen." They have been happening for a long long time. Attributing is, given the complexity of the system and our minuscule historic sample size, difficult. But also perhaps pointless, in an apples v oranges sense.

On the other hand, aggravated weather events become "climate." We zoom out. We see the arc. We see what looks like a delta. It's the arc and the delta that matter.

It certainly seems that climate is changing. But there are three camps, not two.

- Deniers - it ain't happening.

- Sympathizers - it's happening, but it's not man made.

- Believers - it's happening AND it's man made.

I think too often the Sympathizers get lumped with the deniers, and that wrongly clouds the discussion.

How do you determine it’s an “increase” in the first place for such a rare and random event? Are you familiar with Poisson clumping?

The increase I spoke of was the increase in energy into the weather system from humans spewing greenhouse gasses into the amtmosphere for the last 200 years or so. How do you know the weather event in question is a random event? Are you familiar with time series and trends? Thermodynamics? Galois theory?

Can you predict it? If not, it’s a random event.

If I was given all the inputs to the system then yes. If it's known that increasing the energy in the system will lead to such events then yes. I may not predict it will occur at this date/time but one can know that such events will occur when enough energy is put in.

Besides, it's not possible to prove any given event is truly random. There may always be underlying explanatory causes that are not known.

By that logic nothing is ever truly random. Just define the states of all atoms in the universe and you have full determinism. But the fact of the matter is, you can’t establish a causal link with such flimsy evidence unless your theory also _predicts the future_. And by that I mean predicts in the strong sense: you make a prediction and record it with a third party, then see if it held. Not the kind of “predictions” we often see in climate science where a hundred different models are retroactively matched to what actually happened and a handful happen to kinda sorta match the data, if you massage it well enough.

My motivation? A commitment to truthful and honest reporting.

The claims in this article are no better than climate change skeptics pointing to a particularly cold winter or mild summer as evidence for their position.

Your last sentence is false. The difference is one of probability. There’s a much greater chance being correct with the climate change attribution over the other one.

We have to make conclusions and policies based on evidence and what is likely the cause rather than waiting for a formal proof. The most likely cause of the increase in anomalous weather events is due to the increase in energy in the system. It’s worth pointing this out when these events occur.

There’s a much greater chance being correct with the climate change attribution over the other one.

Isn't the null hypothesis that there is nothing specific to attribute the rain? Regardless, what's your statistical test and confidence interval?

Do you create a statistical test and confidence interval for all of the things you believe to be true? I doubt it. Do you have expertise in all areas? I don’t.

I rely on experts. Experts tell me that there is climate change. They tell me that there has been a massive increase in energy in weather system due to human activity. They tell me that the number of anomalies is increasing. Now I encounter a story about a very strange weather occurrence and your question to me is what statistical test I’ve run?

I don’t know you and I can only guess as to the purpose of your question. I’m very skeptical that you can answer your own question when it comes to making conclusions about things outside your expertise. So it seems your purpose is lacking in genuineness.

If it is news to you that people make conclusions based on the evidence they know about and weigh the possibilities and rough probabilities without conducting a formal statistical analysis then I highly recommend that you do some soul searching.

Isn't the null hypothesis that there is nothing specific to attribute the rain?

The person I responded to said that attributing this anomalous weather event to climate change was no different than claims that a cool summer disprove climate change. My claim is that attribution of anomalous weather events to climate change have a higher probability of being true than claims that a cool summer disproves climate change.

Neither claim is provable but one is more likely to be true than the other. The claims are different in terms of probability of being true.

We would have to wait for a few decades to come to such conclusions with scientific methods. It's a political decision to market the event as such.

Suppose we have two hypotheses:

1) The likelihood of extreme rain events hasn't changed over the last hundred years. This recent extreme rain event was just a four sigma event. Such outliers sometimes happen.

2) The likelihood of extreme rain events has shifted over the past hundred years. Here's a plausible mechanism (omitted). The recent rain event was only a two sigma event.

We may not have enough data yet to definitively select one hypothesis and reject the other, but which is more likely?

Where did you get that this is a "four sigma event"? I don't even understand what it means here. Is the calculation using the distribution of the daily rainfall in any day of the year? In a day that it is raining? During the rain season?

Reply on news.ycombinator.com