iiinova — where your digital self keeps evolving

Robert Simmon @rsimmon.bsky.social · 1h

The things I do for science.

Temperature map of CONUS with a day-glow rainbow palette.

Temperature map of CONUS with a day-glow Jet rainbow palette with two areas of exceptionally strong contrast.

Temperature map of CONUS with a day-glow Turbo rainbow palette with perceptual linearity but reduced contrast and it’s really, really ugly.

Temperature map of CONUS with a lightened and linearized rainbow palette purportedly developed for Tableau.

3170

Robert Simmon @rsimmon.bsky.social · 09/10/2026

Map friends, do either of these maps give you a stronger sense of “less” and “more” than the other? I’m trying to decide if this is a strong enough example for #NACIS2026.

Map of South America with a dark blue to yellow palette on a light-blue background. According to dark-is-more bias, the dark strip along the Andes should read as “more”.

Map of South America with a dark blue to yellow palette on a black background. Even with dark-is-more bias, the bright yellow Amazon should read as “more” because of the saturation and contrast against the background.

321

Robert Simmon @rsimmon.bsky.social · 09/10/2026

So let me get this straight — Google won’t link to sci-hub, but will harvest the entirety of the Internet to build its artificial^h^h^h^h^h^h^h^h^h^h super intelligence models. Help me make it make sense.

Text of a court order preventing Google linking to search results to an archive of scientific journals.

181

Robert Simmon @rsimmon.bsky.social · 08/10/2026

Semi-serious #datavis research project idea: Which makes a more intuitive sea surface height anomaly palette: Blue > Red for higher water usually being warmer, or Red > Blue, with blue indicating deeper water? Data from the Copernicus Marine Data Store: data.marine.copernicus.eu/products

Map of the eastern Pacific showing a long, red tongue of positive sea surface temperature anomaly extending along the equator from the coast of Peru into the mid Pacific.

Map of the eastern Pacific showing a long, blue tongue of positive sea surface temperature anomaly extending along the equator from the coast of Peru into the mid Pacific.

480

Robert Simmon @rsimmon.bsky.social · 02/10/2026

TIL: Eduard Imhof made thematic maps. #cartography www.davidrumsey.com/luna/servlet...

Two thematic maps of Switzerland: temperature in January (top) and July (bottom). Temps in January are cool or cold, and the colors are mostly blue and light green. In summer it is warmer, and colors are light green, yellow, and red.

1295

Robert Simmon @rsimmon.bsky.social · 02/10/2026

Is there any chance someone has a scan of Bertin’s color palettes from one of the original editions? + plate 5 from a German edition of Imhof? I’d like to get the colors as accurate as possible for a presentation. cc @kennethfield.bsky.social #color #dataviz #NACIS2026

Page 84 from the ESRI edition of Bertin’s Semiology of Graphics. It shows a variety of color palettes.

183

Robert Simmon @rsimmon.bsky.social · 02/10/2026

With both Lake Powell and Lake Mead at record-lows, here’s a look at how their water levels have varied since Powell was filled in 1980. Neither has recovered from a severe drought in the early 2020s, while Colorado River flows continue to drop. #dataviz #climate #globalwarming

Two graphs of water stored in Lake Powell (left) and lake Mead (right). The data are measured over the water year, which runs from October 1st to October 1st. Years when the lake level increased are marked in blue, years when they decreased are marked in red. Levels in Lake Powell maxed out at about 24-million-acre–feet in the early 1980s, and are currently just over 5-million-acre-feet. Lake Mead was just over 25-million acre-feet in the early 1980s and late 1990s, but is now just under 7-million acre-feet.

12012

Robert Simmon @rsimmon.bsky.social · 25/09/2026

Must we still use psychedelic color schemes for hurricane satellite imagery? Here’s two quick alternates to the standard NOAA product (Band 14 — 11.2µm thermal infrared).

Satellite imagery of Hurricane NOLO shown black > blue > green > white. The structure is clear with some contrast in the coldest cloud tops.

Satellite imagery of Hurricane NOLO shown black > purple > blue > green > white. The structure is clear with more contrast in the coldest cloud tops.

Satellite imagery of Hurricane NOLO shown with grayscale augmented by a saturated rainbow palette. The core of the storm is emphasized over the outer bands. Extreme contrast in the transition between green and red gives the impression of a sharp transition that is not in the data.

051

Robert Simmon @rsimmon.bsky.social · 25/09/2026

Re-created versions of Moses Harris’s color wheels (published in 1766!) by the always-delightful @rougeux.bsky.social www.c82.net/natural-colo... #color #dataviz

vintage color wheel of prismatic colors (red, yellow, blue).

Vintage color wheel of compound colors (orange, purple, green).

1193

Robert Simmon @rsimmon.bsky.social · 24/09/2026

New color tool for #dataviz just dropped: Phosphor by @jscarto.bsky.social This is my first crack at a palette, rendered in QGIS with compositing in Illustrator and Photoshop. You can download or modify the color sequence here: www.joshuastevens.net/phosphor/#c=...

Equal area map of Earth centered on the Americas. NDVI (vegetation) is shown in shades of green and beige. Darkest greens are the most heavily vegetated. Very light green and beige is barren rock, sand, or soil. Open ocean is blue, ice is gray, and there are dark gray graticules spaced every 15˚.

1326

Robert Simmon @rsimmon.bsky.social · 24/09/2026

It’s been a quiet hurricane season in the Atlantic, while the Pacific has been unusually busy — with Hurricane Polo currently raging off Mexico’s west coast. What explains this unusual pattern? El Niño. Read more on @climate.us www.climate.us/news-feature... #climate #weather #hurricane

0114

Robert Simmon @rsimmon.bsky.social · 24/09/2026

With multiple satellites, the FireSat constellation will be able to detect fires shortly after ignition, and eventually monitor their progress over intervals as short as twenty minutes. This series of images shows the Big Grass fire over 48 hours, as it swept over the grassland of southwestern Idaho

Time series of satellite observations of the Big Grass fire in midwave infrared (top, burgundy) and thermal infrared (bottom, orange). The fire smouldered over a wide area on the first night, burned intensely on a narrow front 12 hours later, and slowly subsided over the course of the next 3 observations.

1124

Robert Simmon @rsimmon.bsky.social · 24/09/2026

The heart of FireSat is a pair of midwave infrared bands — one extremely sensitive, designed to monitor small and cool fires. The other attenuated, like a pair of eclipse glasses, for precisely measuring the intensity of hot fires.

Satellite image with a grayscale thermal infrared background, and highlighted fires in midwave infrared, colored dark burgundy to light yellow.

Attenuated midwave infrared satellite image of the Big Grass fire along the Oregon/Idaho border. Small clustered bright spots indicate active fire, mostly along the eastern (right) edge of the fire. Cool areas are dark burgundy, hot fires light yellow. The fire perimeter is outlined in white, and accompanied by an inset locator map.

Four satellite images, showing a shortwave infrared, near infrared, red composite, midwave infrared, attenuated midwave infrared and thermal infrared. Different wavelengths reveal different aspects of the fire. From smoke, to fire intensity, to the temperature of the surrounding landscape.

Color shaded relief map with detailed insets showing bright midwave infrared and thermal infrared images of a woodland fire in Portugal. The fire was in the hills above a steep river valley.

151

Robert Simmon @rsimmon.bsky.social · 24/09/2026

Last week, Earth Fire Alliance released first light images from the initial handful of FireSats. A constellation designed to monitor fire, globally, in near real time. earthfirealliance.org/news-article... The data are stunning, with 5 bands measuring 4 different wavelengths.

Three satellite images showing Dubbo, Australia in different wavelengths. The leftmost image is mostly green, revealing vegetation in a shortwave infrared, near infrared, and red light.

The second image is midwife infrared, colored dark red to light yellow. At this wavelength the image is a combination of emitted and reflected light, so the image reveals information about landscape composition. Urban areas and bare earth tend to be bright, dense vegetation and water dark.

The third image is thermal infrared, showing temperature from dark orange to light yellow. Vegetation tends to be cool, fields and sparse vegetation warmer.

23717

Robert Simmon @rsimmon.bsky.social · 24/09/2026

I love the detail on some of these, even shadows!

Identification card for a Nachi-class cruiser. It features detailed drawings of the top and side of the ship, plus information on its guns and armor.

090

Robert Simmon @rsimmon.bsky.social · 23/09/2026

Happy fall (or spring, depending)!

26728

Robert Simmon @rsimmon.bsky.social · 18/09/2026

Here’s the best view of an aurora I’ve ever seen (and that includes watching a solar storm from a Fairbanks snow bank).

24316

Robert Simmon @rsimmon.bsky.social · 15/09/2026

Awesome to see methane maps from Carbon Mapper making a difference.

Colorful satellite map of a methane plume overlaid on true-color imagery.

0113

Robert Simmon @rsimmon.bsky.social · 14/09/2026

Maybe I’m just frustrated my work was on the cover of both AR3 and AR4 and I’ve never been asked to contribute to the inside.

Cover of IPCC AR3, featuring a picture of the Earth At Night over Africa, Europe, and South Asia.

Cover of the IPCC AR4, featuring two globes with exaggerated color.

110

Robert Simmon @rsimmon.bsky.social · 10/09/2026

These are fabulous. 😍 I especially like this one, with the person in the back who just wouldn’t stop moving.

Early color photograph showing a stack of yellow watermelons in a small square. In the background of the long exposure is a blur, suggesting someone moving through the frame.

0246

Robert Simmon @rsimmon.bsky.social · 09/09/2026

Any QGIS gurus know if it’s possible to change the raster resampling method for layout exports to PDF? Data that looks fine in QGIS has artifacts in the export. First image is how it should look, second image is an exported PDF.

Screenshot showing smoothly rasterized pixels in a temperature anomaly dataset of California and Nevada. Pixels range from light to dark red.

Screenshot showing incorrectly rasterized pixels in a temperature anomaly dataset of California and Nevada. Pixels range from light to dark red, but have a subtle artifacts that are likely the artifact of the QGIS resampling method.

310

Robert Simmon @rsimmon.bsky.social · 06/09/2026

Anak Krakatau’s (Krakatoa) eruption plume rose above a sea of ash yesterday, captured by Sentinel-2. Data & image from browser.dataspace.copernicus.eu 10 meters per pixel, rotated 13˚ counter-clockwise. #volcano #eruption 🌋🛰️

A roiling white eruption plume rises above a brown background of low-altitude ash and stream left (west) in a satellite image collected on September 5, 2026.

0359

Robert Simmon @rsimmon.bsky.social · 03/09/2026

Yeah (at least until the phone battery drained), the Alltrails map is at least as detailed as the 1:24000 USGS topo. The problem was 3 almost-kids from sea-level trying to summit a 14,000 foot peak and getting "advice" probably intended for experienced and acclimated hikers.

Detail of USGS 1:24,000-scale map of the summit of Mount Shasta. No trails or routes are shown.

Zoomed in Alltrails map of the summit of Mount Shasta. A route is shown o the map, that does not directly follow the indicated trail (there are no maintained trails at the summit).

230

Robert Simmon @rsimmon.bsky.social · 02/09/2026

There's a better way! Use hue to separate two scales with the same graduation in lightness and saturation. The same data, with the same emphasis on water warm enough for hurricane formation, but now perceptual above & below the breakpoint. science.nasa.gov/blogs/earth-...

Equal area map of the world centered on the tropical Pacific. Water too cool to support cyclone formation, colored blue, covers all high and mid-latitudes as well as the eastern Pacific. Water warm enough to support hurricane formation, colored red, covers the western Tropical Pacific and Tropical Atlantic.

081

Robert Simmon @rsimmon.bsky.social · 02/09/2026

I am having a particularly frustrating week, so here’s one of the best things I’ve ever done. The eruption of Sarychev Peak in the Aleutian Islands, seen by the crew of the space station. www.nasa.gov/image-articl...

10860851505

Robert Simmon @rsimmon.bsky.social · 02/09/2026

Neat pseudo-3D effect from loading gridded point data into QGIS. Anyone have any advice to properly re-rasterize this? It’s a raster converted to ASCII with columns for lat, lon, and value. #gischat www.ncei.noaa.gov/monitoring-c...

Map of Alaska with seam artifacts showing.

250

Robert Simmon @rsimmon.bsky.social · 01/09/2026

On the subject of volcanic AI-abominations. Original from: www.esa.int/Applications...

AI-generated image of a volcanic eruption with misleading features labeled. There are a lot of them, and they are egregious.

150

Robert Simmon @rsimmon.bsky.social · 01/09/2026

If I’m reading this graph correctly, this weekend’s deadly flash flooding in the Grand Canyon didn’t even bring the Colorado River close to its *average* flow above Diamond Creek (at the western edge of the park). waterdata.usgs.gov/monitoring-l...

Graph showing one week of continuous water flow in the Colorado River above Diamond Creek, Arizona. During the period from August 25th to August 30th the flow ranged from a daily low of about 8,000 cubic feet per second to a high from 9,500 to just under 10,000 cubic feet per second. The daily pattern resembles a sine wave. On the 31st, flow rose to a max of about 10,800 cubic feet per second — well below the average (mean) of about 15,900 cubic feet per second for this time of year.

050

Robert Simmon @rsimmon.bsky.social · 28/08/2026

I guess we didn’t need to be worried about a full parking lot. (Skyline Watershed Ridge Trail.) www.sfpuc.gov/learning/com...

One car in an otherwise unoccupied pre-dawn parking lot.

140

Robert Simmon @rsimmon.bsky.social · 28/08/2026

Landsat appears to have captured the flooding in Nepal yesterday *in progress*. Compare Landsat 9 from Aug. 26th with Sentinel 2 from Aug. 27. h/t @wxnb.bsky.social who found the exact location of the glacial collapse in the same overpass.

Satellite image of flooding along the Trishuli River in Nepal just northwest of Kathmandu. Brown water fills the entire river channel, surrounded by green vegetation. There are white clouds along the bottom of the image.

Satellite image of flooding along the Trishuli River in Nepal just northwest of Kathmandu. There is a relatively narrow strip of water in the river channel, with gray scoured banks on either side. Outside the river channel is mostly green vegetation, with a few clouds upper left and lower right.

2207

Robert Simmon @rsimmon.bsky.social · 27/08/2026

Philanthropy idea: buy a high-res digital elevation model of the Himalaya and release it into the public domain. There’s an 8m/px version from NSIDC, but it’s got lots of gaps. 8m data with holes compared to 8m + 30m ALOS nsidc.org/data/hma_dem...

Shaded relief map of the area between the site of the glacier collapse and border station along the Tibet/Nepal border. There are many gaps in the high resolution elevation data.

Shaded relief map of the area between the site of the glacier collapse and border station along the Tibet/Nepal border. Gaps in the high resolution elevation data are filled with lower resolution data. A curved river valley links the site of the glacier to the Gyirong Border Station.

194

Robert Simmon @rsimmon.bsky.social · 26/08/2026

Probably not. cdn.star.nesdis.noaa.gov/GOES18/ABI/C...

Cropped GOES satellite image of the Bay Area showing a plume of smoke near Monterey, that does not appear to be getting close to Oakland.

120

Robert Simmon @rsimmon.bsky.social · 26/08/2026

Me, for 19+ years at NASA: “You should reconsider using the rainbow palette to show ozone data.” Client, this morning: “I see this and think red is bad.” (Red is good, purple is bad.) From NASA Ozone Watch: ozonewatch.gsfc.nasa.gov/monthly/mont... 📊🛰️🗺️

Map of the South Pole ozone hole on September 29, 2025, rendered as if seeing the Earth from space. The blob of blue and purple colors over Antarctica represent the ozone hole. It is surrounded by a ring of higher ozone levels, indicated by red. The colors are counter-intuitive and don’t map well to human perception.

2201

Robert Simmon @rsimmon.bsky.social · 26/08/2026

Banger of a Space Shuttle photo, from 2003. www.nasa.gov/image-articl... 🛰️🖼️

On Nov. 26, 2003, astronauts on board the International Space Station took advantage of their unique vantage point to photograph the Himalayas, looking south from over the Tibetan Plateau. The perspective is illustrated by the summits of Makalu [left (8,462 meters; 27,765 feet)] and Everest [right (8,850 meters; 29,035 feet)] – at the heights typically flown by commercial aircraft.

0141

Robert Simmon @rsimmon.bsky.social · 22/08/2026

Each clump of trees on this hillside used to be a single giant redwood. Thinking about that as the administration threatens to remove wilderness protections.

Photo of a hillside with dense stands of second-growth redwoods. There is a fern in the foreground.

Photo of a hillside with dense stands of second-growth redwoods. There are two fallen logs and green ground cover, mostly ferns.

082

Robert Simmon @rsimmon.bsky.social · 21/08/2026

The process of scientific discovery is almost always more complicated (and more interesting) than the common narrative. www.realclimate.org/index.php/ar...

Typewritten letter with the text:

"Our base at Halley Bay, Antarctica is currently reporting rather low values of ozone. Value are around 200 dobson units, which is consistently lower than our 1957 - 72 average. We would be interested to know if this is confirmed by satellite data. If so, is it possibly connected with the El Chichon eruption - there’s some evidence that an increased aerosol load has been detected by turbidity measurements with an angstrom pyroheliometer."

0112

Robert Simmon @rsimmon.bsky.social · 20/08/2026

Journals requiring a plain-language summary in addition to an abstract is a very, very good thing. Compare the abstract to the summary for this article: agupubs.onlinelibrary.wiley.com/doi/10.1029/... via @weatherprof.bsky.social

Abstract from a journal article on Global Warming and Earth’s energy imbalance, and a finding that warming is likely a bit higher than some predictions based on past temperature changes predicted. It’s really dense and filled with jargon.

Plain language summary for a journal article on Global Warming and Earth’s energy imbalance, and a finding that warming is likely a bit higher than some predictions based on past temperature changes predicted. It’s quite readable and doesn’t lose much compared to the more technical abstract.

151

Robert Simmon @rsimmon.bsky.social · 11/08/2026

Lots of stories about water in the Colorado, but most focus on Lake Mead or Lake Powell. I wanted t look at the system as a whole. Assuming I did my math right, as of August 8 the reservoirs along the Colorado River were at 32.7% of total capacity.

Two pie charts showing current water storage in the reservoirs in the Colorado River watershed vs. the total system capacity. Mead and Powell are the largest two reservoirs by far, and both are far smaller than normal. Some of the smaller reservoirs, Flaming Gorge in particular, have a larger percentage of the total than normal. Current storage is about 32.7 percent of the maximum possible.

1156

Robert Simmon @rsimmon.bsky.social · 07/08/2026

Three things: 1. This is only helpful if the data is in Zarr, and it’s implemented correctly. 2. It underestimates how hard it is to set up a scientific programming environment. How many cartographers, designers, and journalists are you excluding? 3. Obligatory: imgs.xkcd.com/comics/stand...

Three panel comic of stick figures talking to each other.

SITUATION:
THERE ARE 14 COMPETING
STANDARDS.

14?! RIDICULOUS!
SOON: WE NEED TO DEVELOP ONE UNIVERSAL STANDARD THAT COVERS EVERYONE'S USE CASES.
… YEAH!

SITUATION:
THERE ARE 15 COMPETING
STANDARDS.

100

Robert Simmon @rsimmon.bsky.social · 06/08/2026

An “entirely new kind of view”.

Color print of the view from an 18th-Century balloon over the English countryside that resembles a fisheye perspective.

0115

Robert Simmon @rsimmon.bsky.social · 05/08/2026

I’ve been thinking recently about specialty map projections (and how to explain them) and came across this gem from the Rumsey Map Collection. 🗺️ www.davidrumsey.com/luna/servlet...

Map with an equidistant projection centered on South America to show air travel distances by great circle routes.

1241

Robert Simmon @rsimmon.bsky.social · 02/08/2026

Nighttime satellite imagery taken about 2 a.m. August 1st (left) & 2nd (right) over Spokane, Washington. The bright spots that appear on the 2nd are fires that ignited on the 1st. NOAA/NASA VIIRS Day/Night Band from Worldview go.nasa.gov/4wGuvy8 🛰️🖼️🌖

Grayscale nighttime satellite imagery showing the city lights of Spokane (center right) and a wildfire (upper left). The terrain is visible by moonlight.

Grayscale nighttime satellite imagery showing the city lights of Spokane (center right). New wildfires have appeared to the northwest (upper left), including one or two adjacent tot he city. Smoke and terrain is visible by moonlight.

072

Robert Simmon @rsimmon.bsky.social · 01/08/2026

Neat — GDAL can convert USGS Topo PDFs to TIFFs!

USGS quad 7.5' 1:24000 map for Hayward, California. Bright red Interstate 580 runs left-right through the center, the upper right is hills and green-colored woodland, the lower left is mostly gray surface streets.

3253