fema flood map for data journeys - #991
Conversation
Combined two fema raw cases into one. Facts checked and sources validated.
| @@ -0,0 +1,56 @@ | |||
| # FEMA Flood Maps and the Stationarity Assumption | |||
There was a problem hiding this comment.
Claude sees a few small errors about Camp Mystic. We don't need to give hyper-precise numbers:
https://claude.ai/share/82a413cc-9c1b-44f3-8af2-ae0e9335cbc8
ChatGPT has interesting points about various legal angles:
https://chatgpt.com/share/69d6ea50-41a0-83e8-ba5b-1c98ca354cf9
There was a problem hiding this comment.
Thank you Simon! I have reviewed and addressed the minor changes!
There was a problem hiding this comment.
A few more small issues remain:
https://chatgpt.com/share/69e008c3-f3d8-83e8-aa9a-e4e7708bf274
https://claude.ai/share/76d1ade1-7d11-460a-a24d-3fc18e3b0c73
To avoid arguing about very specific numbers, it's okay to use more approximate wording: eg, say the Camp Mystic tragedy resulted in over 25 casualties
| @@ -0,0 +1,56 @@ | |||
| # FEMA Flood Maps and the Stationarity Assumption | |||
There was a problem hiding this comment.
A few more small issues remain:
https://chatgpt.com/share/69e008c3-f3d8-83e8-aa9a-e4e7708bf274
https://claude.ai/share/76d1ade1-7d11-460a-a24d-3fc18e3b0c73
To avoid arguing about very specific numbers, it's okay to use more approximate wording: eg, say the Camp Mystic tragedy resulted in over 25 casualties
| @@ -0,0 +1,71 @@ | |||
| # Indonesia's Peat Map Conflict and the Governance of Fire Risk | |||
There was a problem hiding this comment.
comments:
https://claude.ai/share/1970c6fe-6ac8-4553-93b0-0e0783d32416
https://chatgpt.com/share/69e00abe-4adc-83e8-a174-9272e46a6e5e
I added to the prompt a request to comment on style, but we don't have to follow all the advice there
There was a problem hiding this comment.
Hello Simon, I have addressed the comments for fema and peat map. Please recheck!
There was a problem hiding this comment.
Hello @simonff
I have also PRed the flint_zip_mask case. Please check!. Thank you!
There was a problem hiding this comment.
some more LLM comments on the peat map:
https://chatgpt.com/share/6a00b1df-2dc0-83e8-8bb4-e09128d9eb8f
https://chatgpt.com/share/6a00b1df-2dc0-83e8-8bb4-e09128d9eb8f
|
|
||
| The Camp Mystic case in Kerr County, Texas, shows how FEMA's formal map-revision process interacts with regulatory exposure. According to [reporting by the Associated Press via the Texas Tribune](https://www.texastribune.org/2025/07/12/camp-mystic-flood-plain-FEMA/) and [NPR](https://www.npr.org/2025/07/12/nx-s1-5465564/fema-removed-camp-mystic), FEMA placed Camp Mystic within a Special Flood Hazard Area in its 2011 Kerr County flood map. In response to appeals — Letter of Map Amendment (LOMA) and related requests that property owners may file to correct apparent mapping errors, typically by submitting elevation evidence — FEMA amended the map in 2013 to remove 15 buildings at the original Camp Mystic Guadalupe site from the hazard area. Following further appeals, FEMA removed about 15 additional structures in 2019 and 2020, these at the nearby Camp Mystic Cypress Lake sister site, which opened to campers in 2020 and suffered less damage in the 2025 flood. | ||
|
|
||
| On July 4, 2025, a flash flood on the Guadalupe River killed at least 27 people at Camp Mystic — 25 campers and two counselors, by most accounts — along with longtime owner and executive director [Dick Eastland](https://fortune.com/2025/07/07/richard-dick-eastland-camp-mystic-owner-director-death-texas-flooding/), who with his wife Tweety had owned and operated the camp since 1974 (day-to-day direction of the Guadalupe River campus by 2025 rested with his son Edward and Edward's wife Mary Liz). Eastland was reported to have loaded campers into his SUV to drive them to higher ground and was swept away with them; his body was recovered near the vehicle. |
There was a problem hiding this comment.
let's reduce this whole paragraph to "killed 28 people at Camp Mystic: 25 campers, two counselors, and the owner Dick Eastland" and link to https://en.wikipedia.org/wiki/Camp_Mystic#2025_flooding
|
|
||
| FIRMs draw on historical precipitation and streamflow records and, in their conventional form, do not adequately reflect heavy rainfall trends or the best available climate science. FEMA Administrator Deanne Criswell [acknowledged in 2022](https://www.smithsonianmag.com/smart-news/federal-flood-maps-are-outdated-because-of-climate-change-fema-director-says-180980725/) that federal maps are "really focused on riverine flooding and coastal flooding" and often fail to capture record rainfall events driven by climate change. Analytically, this amounts to a working assumption of **stationarity** — that past patterns remain reliable predictors of future flood frequency — which is increasingly difficult to defend as extreme precipitation intensifies. | ||
|
|
||
| The Houston case illustrates how far the underlying climatology can drift from the maps in use. Until NOAA's 2018 update, Texas flood mapping relied largely on [Technical Paper 40 (TP-40)](https://www.weather.gov/media/owp/oh/hdsc/docs/TechnicalPaper_No40.pdf), a rainfall frequency atlas authored by D.M. Hershfield and published by the U.S. Weather Bureau in 1961, though some jurisdictions had adopted interim estimates (such as USGS updates in the 2000s) before Atlas 14 arrived. When NOAA released [Atlas 14 Volume 11 for Texas in 2018](https://www.noaa.gov/media-release/noaa-updates-texas-rainfall-frequency-values) — one year after Harvey — the updated rainfall frequency estimates revised Houston-area 100-year, 24-hour rainfall totals upward by roughly a third — from about [13 inches to 18 inches in NOAA's rounded release figures](https://www.noaa.gov/media-release/noaa-updates-texas-rainfall-frequency-values), or about [12.5 inches to 17.9 inches at one Houston point location analyzed by climate researchers](https://climatediscovery.org/atlas-14-texas-the-100-year-storm-is-now-the-25-year-storm-already/) (an increase of about 43 percent at that point); exact values vary by location within the metro. As [Community Impact reported](https://communityimpact.com/houston/city-county/2018/09/27/noaa-rainfall-data-houston/), events previously classified as 100-year storms became, under the updated statistics, substantially more frequent. In Houston-area 24-hour rainfall terms specifically, the new 100-year depth approaches what the older statistics classified as a 500-year depth: the regulatory categories remain intact, but the underlying rainfall climatology they were calibrated against has shifted. |
There was a problem hiding this comment.
remove "or about 12.5 inches to 17.9 inches at one Houston point location analyzed by climate researchers (an increase of about 43 percent at that point);" - this details doesn't add much
| @@ -0,0 +1,73 @@ | |||
| # Indonesia's Peat Map Conflict and the Governance of Fire Risk | |||
|
|
|||
| Indonesia's peatland fires have produced three major transboundary haze crises in the past three decades. The 1997–1998 fires burned on the order of 5 to 10 million hectares during an intense El Niño — estimates vary with methodology, and a [CIFOR review](https://www.cifor-icraf.org/knowledge/publication/32565/) cites figures at the upper end of that range. The 2015 fire season burned about [2.6 million hectares](https://openknowledge.worldbank.org/entities/publication/391e19f4-18f7-56c2-8c28-bb89cc9e7d6f) and produced economic damages that the World Bank estimated at approximately US$16.1 billion, equivalent to 1.9 percent of Indonesia's 2015 GDP. A [Harvard–Columbia modeling study](https://seas.harvard.edu/news/smoke-2015-indonesian-fires-may-have-caused-100000-premature-deaths) led by Shannon Koplitz and published in *Environmental Research Letters* estimated that PM2.5 from the 2015 smoke caused "upwards of 100,000" excess adult deaths across Indonesia, Malaysia, and Singapore, with mortality from exposure roughly doubling relative to the 2006 event (from about 38,000 to about 100,000); the Indonesian government [rejected that estimate](https://www.theguardian.com/environment/2016/sep/21/indonesia-dismisses-study-showing-forest-fire-haze-killed-more-than-100000-people) and attributed 24 deaths to the haze. According to [Global Forest Watch analysis by the World Resources Institute](https://www.wri.org/insights/intense-forest-fires-threaten-derail-indonesias-progress-reducing-deforestation), 40 percent of 2015 fire alerts and 42 percent of 2019 fire alerts in Indonesia were detected on peatland, even though peat covers a much smaller share of the national land area. | |||
There was a problem hiding this comment.
link to http://doi.org/10.1088/1748-9326/11/9/094023 for the Koplitz paper, not to the seas.harvard.edu site
| @@ -0,0 +1,66 @@ | |||
| # Flint, ZIP Codes, and the Modifiable Areal Unit Problem | |||
There was a problem hiding this comment.
could you create a new PR for this file? otherwise, if we iterate on this file while the rest of the files are ready, the PR review will drag on
There was a problem hiding this comment.
please remove the flint and peat map files from this PR
| @@ -0,0 +1,71 @@ | |||
| # Indonesia's Peat Map Conflict and the Governance of Fire Risk | |||
There was a problem hiding this comment.
some more LLM comments on the peat map:
https://chatgpt.com/share/6a00b1df-2dc0-83e8-8bb4-e09128d9eb8f
https://chatgpt.com/share/6a00b1df-2dc0-83e8-8bb4-e09128d9eb8f
| The Camp Mystic case in Kerr County, Texas, shows how FEMA's formal map-revision process interacts with regulatory exposure. According to [reporting by the Associated Press via the Texas Tribune](https://www.texastribune.org/2025/07/12/camp-mystic-flood-plain-FEMA/) and [NPR](https://www.npr.org/2025/07/12/nx-s1-5465564/fema-removed-camp-mystic), FEMA placed Camp Mystic within a Special Flood Hazard Area in its 2011 Kerr County flood map. In response to appeals — Letter of Map Amendment (LOMA) and related requests that property owners may file to correct apparent mapping errors, typically by submitting elevation evidence — FEMA amended the map in 2013 to remove 15 buildings at the original Camp Mystic Guadalupe site from the hazard area. Following further appeals, FEMA removed about 15 additional structures in 2019 and 2020, these at the nearby Camp Mystic Cypress Lake sister site, which opened to campers in 2020 and suffered less damage in the 2025 flood. | ||
|
|
||
| On July 4, 2025, a flash flood on the Guadalupe River [killed 28 people at Camp Mystic: 25 campers, two counselors, and the owner Dick Eastland](https://en.wikipedia.org/wiki/Camp_Mystic#2025_flooding). | ||
|
|
There was a problem hiding this comment.
join with the previous paragraph
| @@ -0,0 +1,66 @@ | |||
| # Flint, ZIP Codes, and the Modifiable Areal Unit Problem | |||
There was a problem hiding this comment.
please remove the flint and peat map files from this PR
|
|
||
| ### The Correction Lag | ||
|
|
||
| While improved datasets — CoastalDEM (2019), GLL_DTM ([2020](https://www.mdpi.com/2072-4292/12/17/2827)/[2023](https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2022EF002880)), DeltaDTM ([2024](https://www.nature.com/articles/s41597-024-03091-9)) — now exist, the installed base of SRTM-derived products remains large. Because many earlier exposure assessments were built on SRTM-based elevations, their conclusions should be revisited as improved terrain models become available. No newline at end of file |
There was a problem hiding this comment.
is there a diff here? if not, please remove from the PR
2 participants
combined two raw fema cases to one. facts checked and sources verified. @simonff