Presentation for Harnessing Massive Data Across Geophysical Domains and Applications: Committee on Solid Earth Geophysics Spring Meeting 2026

I enjoyed the opportunity to present in the spring meeting of the National Academy of Sciences Committee on Solid Earth Geophysics Spring Meeting 2026. The meeting was entitled: Harnessing Massive Data Across Geophysical Domains and Applications.

--Photo by Wendy Bohon, PhD

PRESENTATION

I tried to combine some ideas building from our OpenTopography project with nascent ML collaborations with Dr. Zhiang Chen. Thanks for their contributions.

Opportunities associated with AI/ML are really exciting, but there is a lot to think about for infrastructure, research, and education.

San Andreas Fault in the Carrizo Plain Field trips

Recently, I pulled together some notes for the SCEC UNREST Field trip in the Carrizo Plain. It was a great conversation to share with the esteemed colleagues.

So I can find it, here are the notes that produced as a handout: LINK

Here are a few other items/guides:

• Field trip along the San Andreas Fault in the Carrizo Plain--Guide I wrote some time ago
• Akçiz, S.O., and Arrowsmith, J.R., 2013, New views on the evolution of the San Andreas fault zone in central California and the Carrizo Plain, in Geologic Excursions from Fresno, California, and the Central Valley: A Tour of California’s Iconic Geology, Geological Society of America, p. 1–12, doi:10.1130/2013.0032(01).
• Poster of B4 lidar data derived DEM of the Wallace Creek-Phelan Creeks area

M4-M5 earthquakes in the Awash area of Ethiopia (late 2024)

There have been numerous M4-M5 earthquakes in the Awash area of Ethiopia over the last few months. This is within the Main Ethiopia Rift at its NE end where it begins to open into the Afar. There is a young volcano Fentale in that area which had an eruption last in 1820.

Nice post fromn the Euro-Mediterranean Seismological Centre (EMSC): SUMMARY

Here is a live link of that area: LINK--last 24 hours
Here is the same area: LINK--7 days

That would also mean that there have been many smaller events (Gutenberg-Richter law suggests 10x for each integer decrease in magnitude). The global seismic networks the USGS uses have an M4 cutoff outside the US. These are shallow events (10km default depth). I would bet they are even shallower but that is all the network location can provide.

No doubt the Main Ethiopian Rift is active and extending NW-SE a few mm/yr (hence the NE-SW trending faults in the area). So, we should not be surprised to see the events. The focal mechanisms (directionality of first motions) are consistent with motion along those faults.

The location accuracy is probably only good to 5-10 km so these events might actually be more well aligned with themselves and faults in the area instead of the cloud in the map above.

What is intriguing to me is that it is a swarm. I would not be surprised if there were also some interaction with the magmatic system of the Fentale to Mt. Dofan. I note the InSAR data in this LINK that seems consistent with intrusions along the rift NE of Fantale. Here is footage of an eruption on January 3, 2025 in the area NE of Fantale (Mt. Dofan) where the more recent events have occured: LINK. Thanks to Amy Rector for the link.

It is interesting and certainly in the near field, these events could damage unreinforced masonry structures as well as frighten people and animals. These events have been felt in Addis Ababa (~200 km distance) as moderate to weak shaking.

Links:

• Earthquake Insights post
• Volcano Cafe blog entry: LINK
• Mud volcano and phreatic behavior: LINK
• VOA article: LINK
• Newspaper report after recent larger earthquakes: LINK
• Newspaper report with some nice field pictures with fracturing: LINK
• Useful local commentary (in Amharic): LINK
• I did find an interview with Prof Ataley Ayale who runs the seismic station at Addis Ababa University. He does not have a lot more to say: https://addisstandard.com/although-earthquakes-measuring-5-0-or-higher-are-relatively-uncommon-in-ethiopia-they-still-pose-a-significant-risk-to-infrastructure-and-human-safety-atalay-ayele-prof-a-geoph/
• Earthquake history of Ethiopia by Pierre Gouin
• Interferogram by Carolina Pagli team indicating intrusions likely: LINK

Remembering Thomas C. Hanks

I attended the Tom Hanks, a Remembrance symposium at the USGS in Moffett Field Oct. 17, 2024. It was a great chance to remember Tom, reflect on his vast scientific and personal impacts, and to catch up with old friends and colleagues. It was also nice to meet Tom's daughters. I was fortunate to be able to make a presentation. Here is my talk: LINK.

Here is the program from the event: LINK

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I heard that Thomas C. Hanks passed away recently. He was a mentor to me. He worked for his career with the US Geological Survey. The memorials of him from his colleagues will be many and deep. I wanted to capture some of my memories of him. Tom was very supportive of young scientists and very broad in his scientific thinking. While he was most well known as a seismologist, his work in geomorphology and fault scarps and fragile geologic features was transformative.

This was a sticky on a manuscript draft he once gave me after a discussion on uncertainties in morphologic datting. Look at the nice handwriting (usually from a well sharpened #2 pencil). And the signature THanks.

Tom was on my Ph.D. supervisory committee. I was at Stanford and Tom was in Menlo Park at the USGS. Like many of his colleagues there, he was very generous with his time with the Stanford students. We talked a lot about fault scarps and diffusion, but also about the San Andreas Fault and I was able to drive for him on a few field trips to the SAF in the southern Bay Area into the Creeping Section. With Professor Gordon Brown's support (chair of our department at the time), Tom helped to lead an active tectonics seminar one quarter.

Tom's work on the age of scarplike landforms from diffusion-equation analysis (title of one of his latter papers on the subject) was very influential. He teamed up with Robert Wallace and others to take something simple about how fault scarps apparently change shape over time and quantify it in a realistic way. There are numerous important papers on the topic with Tom as an author but two seminal ones are:
Hanks, T. C., Bucknam, R. C., Lajoie, K. R., & Wallace, R. E. (1984). Modification of wave-cut and faulting-controlled landforms. Journal of Geophysical Research. https://doi.org/10.1029/JB089iB07p05771
and
Hanks, T. C. (2000). The Age of Scarplike Landforms From Diffusion-Equation Analysis. https://doi.org/10.1029/rf004p0313 in Quaternary Geochronology: Methods and Applications. In AGU Reference Shelf 4 (Vol. 4).

Among many other contributions on the age of scarplike landforms, Tom introduced a simple morphological dating approach: reduced slope-offset. He argued for a measure of the scarp midpoint slope (reduced by the far field slope) versus the vertical offset and he developed a nice calibration along with his colleagues for the rate constant k. He favored analytical solutions (tolerating my numerical approach).
One small anecdote that I always appreciated on the geomorphology side was his desire to name a unit for GK Gilbert (1m²/kyr = 1GKG). See the seminal 1984 Hanks et al JGR paper. It did not catch on but was a fun idea.

In 2007, David Haddad and I went with Tom to Northern Arizona University to see the collection of his father's photographs that he had endowed: Repeat Photography Site for The James J. Hanks Photographs, 1927-1928. Tom, like always, was deeply engaged/obsessed with the topic at hand. He worked hard to relocate and repeat his father's photographs, as well as to tell their story.

Whilst on the trip to Flagstaff, Tom, David, and I stopped to see and discuss the Granite Dells (near Prescott, AZ). Tom had been leading parts of the seismic hazard analysis for the Yucca Mountain possible nuclear repository. The problem they were coming up with was the age of the landscape was great (million year old landforms) and there were fragile geologic features and precarious rocks that may have been there fragile for a large fraction of that time. However, the extrapolation of the ground motion predictions would be to extreme, possibly unrealistic levels. Tom was interested in these million-year-old landscapes of fragile geologic features and recognized their value as an observational constraint for seismic hazard analysis. This is an impressive product of their work:
Hanks, T. C., Abrahamson, N. A., Baker, J. W., Boore, D. M., Board, M., Brune, J. N., Cornell, C. A., & Whitney, J. W. (2012). Extreme Ground Motions And Yucca Mountain. Extreme Ground Motions and Yucca Mountain Open-File Report 2013–1245, US Geological Survey.

Tom was interested in precariously balanced rocks given their use as a part of seismic hazard analysis. He thought it might be helpful for new people to get involved. So, he pulled David and I into it. He was supportive and helped generate some funds for us. That lead to a couple of nice papers lead by David. I regret that we did not have Tom as a coauthor:
Haddad, D. E., Akciz, S. O., Arrowsmith, J. R., Rhodes, D. D., Oldow, J. S., Zielke, O., Toke, N. A., Haddad, A. G., Mauer, J., & Shilpakar, P. (2012). Applications of airborne and terrestrial laser scanning to paleoseismology. Geosphere, 8(4). https://doi.org/10.1130/GES00701.1
Haddad, D. E., Zielke, O., Arrowsmith, J. R., Purvance, M. D., Haddad, A. G., & Landgraf, A. (2012). Estimating two-dimensional static stabilities and geomorphic settings of precariously balanced rocks from unconstrained digital photographs. Geosphere, 8(5). https://doi.org/10.1130/GES00788.1

A final lesson from Tom is that senior scientists should be generous and use their privilege to do good. Tom was a widely appreciated mentor of younger scientists--men and women. He was also a leader who did not shy away from trying to do the right thing. Just one example relates to another senior scientist who recently passed away: Paul Tapponier. Professor Tapponier led a transformation of our understanding of continental tectonics. He favored results with relatively high slip rates and thus the inference that the deformation even in plate interiors was more plate-like. Tom supported his colleague Wayne Thatcher who had come up with a result based on geodesy for the deformation of the Asian continental interior (Thatcher W. 2007. Microplate model for the present-day deformation of Tibet. J. Geophys. Res. 112:B01401) (and that did not sit well with Paul). Zack Washburn and I had written a paper based on paleoseismology in which we could not support enough earthquakes to support a high slip rate). Tom stepped in to mediate between Wayne and Paul and consulted me as part of his preparations. Tom had the stature, the intelligence, maturity and deserved respect so that he was able to set the tone for what I gather was a productive meeting.

I ended up with a copy of Tom's USGS bio and I note the following which is a nice example of his writing and matter-of-fact approach:

Open Science recognition prize at AGU 2023

Our OpenTopography project was honored at this year's AGU with the Open Science Recognition prize: "For outstanding contributions in cyberinfrastructure, data management, training, and outreach associated with open-access high-resolution topography." It is a great honor and nice reognition for more than 15 years of work by our team. Huge thanks to Roman DiBiase for leading the nomination and for the letter writers Mike Oskin, Paola Passalacqua, and Josh Roering.

Chelsea Scott and Chris Crosby made a nice presentation summarizing our efforts. The recording is here: link.

It was also nice to meet the other winner, Tasha Snow and appreciate her efforts. In particular, I really appreciated her articulation of Open Science values (image from her presentation):

The full award ceremony is recorded here: link.

Here are a couple of pictures of our team:

Other links:
OpenTopography news release: link

And we celebrated with the other ASU / SESE awardees (Vernon Morris and Everett Shock):

Determining the cell size of a Digital Elevation Model from point density

We regulary refer to high resolution topography (HRT) as having Digital Elevation Model (DEM) pixel sizes with their edges less than 1 m. Many raster DEMs derived from airborne laser swath mapping available for example from OpenTopography are delivered at 1m/pix. In many cases, the point density might be high enough to support even higher resolution or smaller pixels. This seems like a fairly simple question: what is the recommended cell size of a DEM for a given point density?

The Langridge, et al., 2013 paper cites Hu, 2003 and suggests the following:

where S is the estimated cell size (typically in m), n is the number of sample points and A is the area containing the sample points.

Here is a little spreadsheet to solve this equation given A and n: link.

In the trivial case, that equation can be rearranged and show that 1 point per sq. m is consistent with a 1 m DEM. That seems ok in the sense that then on average there is one point for each pixel which would be the logic of this equation. It assumes then that the points are well distributed and that whatever average for point density that we might use to estimate a cell size represents the data well. A complication is the method of DEM computation: local function or linear interpolation. An example of the local implementation is the Points2Grid tool (see prototype here and here). Tinning or the use of triangular irregular networks linearly interpolates between (selected) points to estimate DEM pixel elevations (see blast2dem for example).

Here are some illustrations from the Jemez River Basin dataset cited below. These data have a stated point density of 9.68 points/sq. m and were computed within the OpenTopography portal using the TIN approach.
Here is an example from a site called Sulphur Creek (1 m/pix for this Digital Surface Model, DSM):

Here is a zoom to that site with the 1m/pix DSM hillshade and 14% of the points displayed from ArcMap. We can see the 1 m pixels and that there are a decent number (about 9.7) of points for each one. This has all of the points (all classes):

Here is the same view but with a 0.32 m/pix (recommended resolution from equation above), but applied to the Digital Terrain Model (DTM). Note this is not really correct resolution estimate because the number of points classified as ground is only about 1/5 of the total. And, this shows the ground returns only.

Zooming in even more, we can see the big triangular facets where the TINNING algorithm spanned the data gaps in the ground classified points for this 0.32 m/pix DTM:

Finally, zooming back out with the 1 m/pix DTM displayed, we can see that at this scale, the DTM landscape is well represented in general. We do need the interpolation by tinning across the gaps in the ground points:

References Hu, Y., 2003. Automated Extraction of Digital Terrain Models, Roads and Buildings Using Airborne LiDAR Data (PhD thesis). Department of Geomatics Engineering, The University of Calgary, Calgary, Alberta, Canada.

Jemez River Basin Snow-off LiDAR Survey. Distributed by OpenTopography. https://doi.org/10.5069/G9RB72JV . Accessed: 2023-05-14

Langridge, R.M., et al., Developing sub 5-m LiDAR DEMs for forested sections of the Alpine and Hope faults, South Island, New Zealand: Implications for structural interpretations, Journal of Structural Geology (2013), http://dx.doi.org/10.1016/j.jsg.2013.11.007

A simple evolutionary model for fragile geologic features

The fragility of geologic features, such as precariously balanced rocks (PBRs), can be measured by a simple parameter like α. For the case of a PBR is the smallest of the angles between the vertical from the center of mass and its rocking points. In a landscape, each object will have a fragility and so the ensemble will be a fragility distribution. The controls on the initial distribution and its long term evolution will be from the environment and its history (material properties, landscape evolution (lowering), weathering, shaking, etc.). The distribution can be disturbed by an earthquake (or other loading like windstorm, human impacts, etc.) which abruptly removes features with fragility below a threshold α.

Example of PBRs in Granite Dells, Arizona

I started to think an analogy with a fruit tree. I am not sure this is an original idea maybe I heard it somewhere. The progressive ripening of the fruit can be interrupted by a shake which will remove completely a subset of the most ripe fruit. For the case of the rocks, the ripening is a gradual decrease in fragility over time and then a fragility reset of a subset rather than removal after they are toppled in a shaking event. Ripening continues and the processes repeat over time. This simple model does not account for changing ripening rates or much variation in shaking effects other than threshold α. The basic idea then for seismic hazard is that the fragility distribution at a location reflects the history of the long term ripening and episodic shaking and reset. Therefore, if we can produce fragility distributions for landscapes that otherwise comparible (ripening and threshold α), we might be able to say which has seen more recent shaking and of what severity. This does not directly address the age control for the history and this remains a significant problem.

I wrote a simple code to explore this problem with the hope that it helps us explain and isolate the basic controls on fragility distributions. See the figure below which presents the fragility distributions through the experiment.The algorithm is simple:

• Set up: Specify number of objects, the initial α distribution (assume normal), the ripening rate distribution (also normal; this is the loss of fragility per year), the timing of the earthquake, the threshold α, and the max time
• Interseismic period 1: ripen α until the earthquake (begin and end are the upper two plots below)
• Earthquake: remove α < threshold α (third plot below) and reset those α drawing from the same initial distribution (fourth plot below)
• Interseismic period 2: ripen α until the end of the model time (final plot below)

Fragility distributions through the experiment. Upper plot is initial α. 2nd plot is α and the end of interseismic period 1. 3rd plot shows removed objects with α < threshold α. 4th plot is reset distribution of < threshold α Note that some α results below threshold α. Lowest plot shows continued ripening until the end of the model.

The additional two figures show the evolution of fragility with time of a subset of the objects and a spatial view of the randomly positioned features, their fragility at the time of the earthquake, and the circled objects that failed. The evolutionary diagram helps illustrate the interruption of the more fragile features by the event while the others do not notice. The map provides an idea of the search challenge that this scenario presents.

Evolution of fragility of a subset of the objects. The red star indicates the time of the earthquake and the threshold α.

Map view randomly positioned features, their fragility at the time of the earthquake, and the circled objects that failed.

Maybe more realistic with a more sparse set of features and a lower threshold α (0.3 as opposed to 0.4 above).

This is just a sketch of the problem, but it is a toy model in which we can explore the importance of the distribution widths, timing of earthquake, etc.

The MATLAB code is in this repository: https://github.com/jrarrowsmith/MATLAB-Geomorphology; make sure to get the script PBRevolution.m and functions ripenPBR.m and shakePBRs.