Advertisement · 728 × 90

Posts by James P. Collins

Preview
The Sky is the Limit (NCAR) A documentary about a crown jewel of American science, the National Center for Atmospheric Research (NCAR) and the peril it faces.

Crowdfunding for "The Sky is the Limit," a documentary about a crown jewel of American science, the National Center for Atmospheric Research (NCAR) and the peril it faces.

3 weeks ago 0 0 0 0

A total of 13 states have tied or set all-time March heat records over the period Mar. 18-21, and one or more stations in Arizona and California have tied or broken the previous all-time U.S. heat record of 108°F each of those four days. This heatwave is in a class of its own historically!

4 weeks ago 129 63 4 7
"Gloria" shared this pic of herself picking strawberries in CAs Central Coast in a very watery row

"Gloria" shared this pic of herself picking strawberries in CAs Central Coast in a very watery row

"Gloria" shared this pic of herself picking strawberries in CAs Central Coast. "The water was up to my knees making it very hard to push my cart. My feet were so wet and cold, I couldn’t feel them. Growers have us harvest despite the flooding to avoid losing product." #WeFeedYou

1 month ago 463 147 3 8
A photo of UNC Chapel Hill's campus.

A photo of UNC Chapel Hill's campus.

UNC Chancellor Scraps Secret Recording Policy https://bit.ly/4aMlew6

1 month ago 4 1 0 1

EPA claims repeal will be "de minimis." Based on our forthcoming work on wildfire smoke (below) and a back-of-the-envelope calculation from Table 1 in the final rule, I estimate this action will kill over *200,000 Americans* over the 21st century. De minimis indeed!

www.nber.org/papers/w33829

2 months ago 18 14 0 1
Preview
Project Drawdown launches Climate Science Serving America Fellowship to support scientists working in the public good Fellowship seeks early- to mid-career researchers based in America who are committed to science and public expertise

While others are stepping back on climate, Project Drawdown is stepping up!

We’re proud to announce the Climate Science Serving America Fellowship to support U.S.-based Ph.D.-level scientists & engineers focused on climate solutions for the public good.

drawdown.org/news/project...

2 months ago 108 56 1 3
Preview
A dashboard for tracking FEMA major disaster declarations My last post described how presidential disaster declarations are an important indicator of how the 2nd Trump Administration intends to govern disasters.

Want to know how many disaster declarations are being made under the 2nd Trump Administration, and how long they are taking? I made a simple dashboard to help: andrewrumbach.substack.com/p/a-dashboar...

2 months ago 8 3 0 0
It must be very hard to publish null results
Publication practices in the social sciences act as a filter that favors statistically significant results over null findings. While the problem of selection on significance (SoS) is well-known in theory, it has been difficult to measure its scope empirically, and it has been challenging to determine how selection varies across contexts. In this article, we use large language models to extract granular and validated data on about 100,000 articles published in over 150 political science journals from 2010 to 2024. We show that fewer than 2% of articles that rely on statistical methods report null-only findings in their abstracts, while over 90% of papers highlight significant results. To put these findings in perspective, we develop and calibrate a simple model of publication bias. Across a range of plausible assumptions, we find that statistically significant results are estimated to be one to two orders of magnitude more likely to enter the published record than null results. Leveraging metadata extracted from individual articles, we show that the pattern of strong SoS holds across subfields, journals, methods, and time periods. However, a few factors such as pre-registration and randomized experiments correlate with greater acceptance of null results. We conclude by discussing implications for the field and the potential of our new dataset for investigating other questions about political science.

It must be very hard to publish null results Publication practices in the social sciences act as a filter that favors statistically significant results over null findings. While the problem of selection on significance (SoS) is well-known in theory, it has been difficult to measure its scope empirically, and it has been challenging to determine how selection varies across contexts. In this article, we use large language models to extract granular and validated data on about 100,000 articles published in over 150 political science journals from 2010 to 2024. We show that fewer than 2% of articles that rely on statistical methods report null-only findings in their abstracts, while over 90% of papers highlight significant results. To put these findings in perspective, we develop and calibrate a simple model of publication bias. Across a range of plausible assumptions, we find that statistically significant results are estimated to be one to two orders of magnitude more likely to enter the published record than null results. Leveraging metadata extracted from individual articles, we show that the pattern of strong SoS holds across subfields, journals, methods, and time periods. However, a few factors such as pre-registration and randomized experiments correlate with greater acceptance of null results. We conclude by discussing implications for the field and the potential of our new dataset for investigating other questions about political science.

I have a new paper. We look at ~all stats articles in political science post-2010 & show that 94% have abstracts that claim to reject a null. Only 2% present only null results. This is hard to explain unless the research process has a filter that only lets rejections through.

2 months ago 644 222 30 52
Advertisement

SUPER excited to see this live in the world. In particular hugely admire this team's dedication to openness and transparency about methodology, code, assumptions.

Every tool has strengths and weaknesses, but sharing openly is how the whole community can improve. Big kudos @carbonplan.org

2 months ago 75 18 2 1
Preview
A registered report megastudy on the persuasiveness of the most-cited climate messages - Nature Climate Change How to effectively communicate climate change to the public has long been studied and debated. Through a registered report megastudy, researchers tested the ten most-cited climate change messaging str...

"Our main conclusion ... is that several well-cited messaging strategies reliably influence pro-environmental attitudes and behavioural intentions in the United States ... We found little evidence of heterogeneous treatment effects between Democrats and Republicans..."

2 months ago 1 0 0 0

"The study’s findings are also in line with previous research: A 2021 study suggested that PM2.5 from wildfires is up to 10 times more harmful than PM2.5 from other sources, such as ambient pollution."

2 months ago 1 0 0 1
Post image

New data show that business concentration has increased persistently in many countries over the past century by sales, net income, and capital, but not by employment, from Yueran Ma, Mengdi Zhang, and Kaspar Zimmermann www.nber.org/papers/w34711

2 months ago 5 2 0 0
Preview
NSF Intent to Restructure Critical Weather Infrastructure

NSF just posted their Dear Colleague Letter regarding the restructuring of NCAR. www.nsf.gov/funding/info...

2 months ago 18 27 1 4
NWS Probabilistic Precipitation Portal

@nws.noaa.gov's experimental probabilistic precip forecast shows the chance of snow, ice, and total accumulations through Monday

2 months ago 0 0 0 0

any primers on this in utilities you can recommend?

3 months ago 1 0 1 0
Map of water table depth across the continental United States at 30-meter resolution, with inset panels showing detail at 100km, 10km, and 1km scales in both eastern and western US locations. Color scale shows depth from shallow (blue) to deep (yellow) on a log scale.

Map of water table depth across the continental United States at 30-meter resolution, with inset panels showing detail at 100km, 10km, and 1km scales in both eastern and western US locations. Color scale shows depth from shallow (blue) to deep (yellow) on a log scale.

New work from our team: we mapped water table depth at 30-meter resolution across the entire continental US.
That's ~8 billion grid cells, trained on over 1 million well observations. Highest-resolution estimate to date.

3 months ago 21 4 1 1
Preview
Horsification: Embodied gentrification in rural landscapes In this paper, I consider the role of animals in gentrification processes, developing a conceptualisation of ‘horsification’: the proliferation of hor…

rabbit hole OTD "I consider the role of animals in gentrification processes ... how the materialities of horse bodies are embodied in specific landscape changes: qualities of pasture, styles of fencing and stables ... and lead to the territorialisation and contestation of broader rural landscapes."

3 months ago 1 1 0 0
Preview
How One Company Is Pushing a Private Takeover of Flood Insurance

Neptune Flood (subsidiary of NYSE:NP) is actively working to privatize the NFIP. Company leadership has communicated their work and this goal, in line with Project 2025, to the White House.

Privatization would have big implications for flood-prone communities.

www.nytimes.com/2026/01/15/c...

3 months ago 4 4 0 0
Advertisement

Caveat here being that 5cm hail is listed as being where damage to buildings starts... but based on an EU study where much more of the building stock is tile roof.

Asphalt shingles, obviously much more common in US, take damage and need replacement at much smaller hail sizes.

3 months ago 26 5 1 0
Post image

Have you ever heard of the black-white mortality crossover?

Before about 82, black men have higher age-specific probabilities of dying than white men.

Between roughly 82–86, the gap closes.

After 86, the pattern actually reverses: white men become more likely to die at each age than black men.

3 months ago 13 5 1 0
A map of the US with each 2025 billion-plus dollar weather and climate disaster geo-located on it. Source: Climate Central

A map of the US with each 2025 billion-plus dollar weather and climate disaster geo-located on it. Source: Climate Central

After the US admin cancelled the $B Climate + Weather Disaster dataset, @climatecentral.org hired the scientists who ran it and set it back up.

Now the 2025 numbers are in: it's 3rd highest year on record and highest year w/o land-falling hurricanes.

More: www.climatecentral.org/climate-serv...

3 months ago 1047 510 18 22
Preview
Marin County Looked Like ‘a Lagoon’ After King Tides, Heavy Rain | KQED Flooding in Marin County this weekend foreshadows a far wetter future due to human-caused climate change.

County officials told KQED on Monday the exact damage estimates aren’t yet known, but that hundreds of structures were impacted by the flooding brought on by stronger-than-expected rainfall and king tides, the highest tides of the year. (via @kqednews.kqed.org)

3 months ago 117 24 4 3
Preview
Extreme Temperatures Promote High-Fat Diets Founded in 1920, the NBER is a private, non-profit, non-partisan organization dedicated to conducting economic research and to disseminating research findings among academics, public policy makers, an...

"Nutrition is a central determinant of human health, yet the direct impacts of climate on dietary intake remain poorly understood ... [we] show that both extreme heat and cold trigger a shift toward energy-dense diets, adding a previously overlooked behavioral channel to the climate–health nexus."

3 months ago 1 1 1 0
Post image Post image

Researchers recently asked Americans what income level other people need in order to live a good life.

A whopping 86% of Americans reported income levels well below what others say they themselves need.

Fascinating!

3 months ago 80 23 2 4
Preview
Polarization in Flood Risk Management? Sensitivity of Norm Perception and Responsibility Attribution to Frequent Flood Experience Abstract. In this study, we examine the relationship between frequent flood experience (FFE), norm perception, and responsibility attribution. Given that floods are assumed to occur more often in the ...

'respondents with multiple flood experience are more likely to perceive social norms supporting individual protective behavior, ascribe more responsibility to public authorities and less to their community'

nhess.copernicus.org/articles/25/...

3 months ago 4 3 0 0
Post image Post image Post image Post image

What would it cost to end extreme poverty?

"We estimate that reducing the poverty rate to 1% ... would cost $170B nominal per year."

"The results correspond to a cost of (approximately) ending extreme poverty of roughly 0.3% of global GDP."

3 months ago 113 50 4 4
Video

Where rivers have flooded in 2025, using data from USGS and @nws.noaa.gov

#StateOfFlood

4 months ago 16 11 4 1
Preview
Protect the National Center for Atmospheric Research Write today and urge your members of Congress to oppose the Trump administration's plans to dismantle the National Center for Atmospheric Research (NCAR.) Your message to Congress will help protect NC...

@ucs.org has released a letter you can write to your Congressperson to save @ncar-ucar.bsky.social

Share your voice on the significant value of this bedrock climate and weather insitution. Your story matters!

secure.ucs.org/a/2025-prote...

4 months ago 95 62 3 1
Advertisement
Preview
UNC cuts all six area studies research centers, effective 2026 As part of the University's plan to make $70 million in budget cuts across the institution, Vice Chancellor for Finance and Operations Nate Knuffman projected that cutting 14 centers and institutes in...

America First, at America’s first public university.

4 months ago 35 26 1 14