C.E. DeForest, W.H. Matthaeus, T.A. Howard, & D.R. Rice, Astrophys. J. 812, 108.

Encke was one of the first objects observed by HI-1, back in 2007.  The ion tail is not very feathery, so we could track its clumps in the HI-1 L2 movies, quantifying results that go all the way back to Biermann (1951, 1967). Raouafi (2015) independently wrote a really neat paper explaining why some comets' tails are less feathery.

C.E. DeForest & T.A. Howard, Astrophys. J. 804, 126.

We'd been taking a lot of flak for advocating that heliospheric imaging is possible from LEO -- after all, SMEI suffered from  background effects.  To show that they aren't a problem, Tim Howard degraded a bunch of STEREO images with simulations of worst-case scenarios, and I reconstructed the data.  Worked like a champ.

C.E. DeForest, T.A. Howard, & D.M. McComas, Astrophys. J. 787, 124.

This paper was Tim's idea.  He realized that we should be able to actually see waves moving in from the outer corona. Using Fourier analysis we showed there are inbound features throughout the STEREO/COR2 field of view -- out to 15 Rs, much farther than most prior models.  The next year, Anna Tenerani (2016; A. Tenerani, M. Velli, & C.E. DeForest, Astrophys. J. 825, L3) explained that what we saw is almost certainly inbound jets rather than waves.  That doesn't affect the primary result, but is itself a very interesting piece of science.

C.E. DeForest, R.A. Howard, M. Velli, N. Viall, & A. Vourlidas, Astrophys. J. 862, 18.

In 2014 Vourlidas ran a special deep-exposure campaign for us, on the STEREO-A COR2 instrument, just before STEREO-A passed behind the Sun.  All kinds of interesting stuff popped out of the data.  The most exciting result, for me, are the presence of an Alfvén “zone” and the overwhelming amount of temporal structure.


C.E. DeForest, T.A. Howard, & D.M. McComas, Astrophys. J. 745, 36.

One of the first things we saw in the reduced HI-2 data was a peculiar "V" shaped feature zooming outward around 1 A.U.  We found that it was an "anti-CME": a U-loop departing the Alfvén surface, reducing the sun's open magnetic flux.  We used kinematic effects to measure its flux via the tension force -- a pretty fun analysis.

Craig DeForest

Research articles

I've been publishing research articles since the early 1990s.  For a definitive list and publication metrics, you can use either the ADS Abstract Service or my Google Scholar profile.  

Here, I've grouped some of the more important-to-me ones, organized by research topic.  On most I am first author; on some I am not.  The common theme here is that I've continued the particular line of research for more than just a one-off study.  The articles themselves are linked here also for reference.  I include a little personal context with each listing.

Topics are:

     Solar Wind Imaging ("heliospheric imaging" and outer coronal imaging)

     Analysis Techniques & Scientific Computing

     EUV Imaging and Plasma Diagnostics

     Solar Polar Plumes

     Waves on the Sun

     MHD Modeling

     • Magnetic Tracking and the Solar Dynamo

Solar Wind Imaging ("heliospheric imaging" and outer coronal imaging)

This is the field that has become my research bread-and-butter over the last few years.  What began as a small side-track to an MHD modeling project has rapidly become the most dynamic and enervating part of my research "wheelhouse".  I am still amazed that we can simply see the solar wind directly, in visible light, against the background of a starfield 1000 times brighter.  

C.E. DeForest, W.H. Matthaeus, N. Viall, & S. Cranmer, Astrophys. J. 828, 66.

As early as 2011, Tim Howard and I noticed that the solar wind looks different than the corona does.  We thought it was pretty obvious and didn't think to publish it -- until Viall mentioned it at a Solar Probe meeting.  We were able to show that the difference is most likely from hydrodynamic turbulence above the corona.  Exciting times!

Analysis Techniques & Scientific Computing

Much of my work involves developing and appying new techniques.  Where possible, I include descriptions of those techniques in scientific articles that are focused on other things (the Detailed Structure paper above and the Magnetic Tracking I paper below are two examples).  But some papers really do cover just a technique or analysis that we developed for later application.  

C.E. DeForest, Astrophys. J. 838, 155.

Noise-gating arose from a pair of projects:  EUV loops, led by Terry Kucera, and EUV jets, led by Judy Karpen.  Both projects needed better SNR than SDO/AIA could offer.  It reduces image noise by 10x, post facto, by exploiting coherence to distinguish features from noise.  (It's also patented  -- but free to use for research).

C.E. DeForest & K. Glazebrook, ArXiV 1702.07753

This mini-book is incorporated into the Perl Data Language (PDL) documentation, and now uploaded to ArXiV.  PDL uses a metalanguage (PP) to compile quasi-C routines into C operators -- which you can incorporate into the runtime language directly in your scripts.  This functionality is a big part of why PDL is so darned useful.

T.A. Howard & C.E. DeForest, Astrophys. J. 752, 130;

C.E. DeForest, T.A. Howard, & S.J. Tappin, Astrophys. J. 765, 44;

T.A. Howard, S.J. Tappin, D. Odstrcil, & C.E. DeForest, Astrophys. J. 765, 45.

When we started the disconnection paper (above), I was surprised at how disorganized the literature was on Thomson imaging in general -- and also at how many misconceptions had crept into the zeitgeist.  We covered the basic theory and demonstrated something that should be obvious: heliospheric imagers can detect features over a huge range of out-of-plane angles. During the development of the initial paper, it became clear we needed a trilogy to cover polarization and its application to CMEs, to establish a method to analyze polarized images.

C.E. DeForest, Sol. Phys. 219, 3.

This is the paper that broke a two year early-career writing drought for me.  I'd been working on a bunch of techniques and on PDL for a while.  During a Thanksgiving visit to my in-laws' ranch, I got inspired and hacked out the first draft of PDL::Transform, which I have used daily for 15 years. This paper describes the technique.

EUV imaging and plasma diagnostics

I got my start in solar physics doing EUV imaging.  My graduate advisor, Dr. A.B.C. Walker, was one of two pioneers (Leon Golub was the other) who developed multilayer optics into a useful tool for solar imaging.  EUV imaging today is one of the most valuable probes we have into the structure and physics of the solar corona.  These articles are mostly about EUV imaging itself.

B. Poduval, C.E. DeForest, J.T. Schmelz, & S. Pathak, Astrophys. J. 765, 2.

This work was a follow-up to the TRACE stray light paper (below).  Bala Poduval painstakingly measured both the sharp and diffuse components of scattered light in each AIA telescope.  We also published both the PSFs and their inverses as FITS files.  Simply convolving AIA data with the inverse PSF will improve EUV photometry.

C.E. DeForest, P.C.H. Martens, & M.J. Wills-Davey, Astrophys. J. 690, 1264.

Around 2007 Wills-Davey and I looked at the "diffuse corona" in TRACE, which appeared inconsistent with  Alfvénic motions in active regions (Wills-Davey's dissertation).  We discovered that TRACE scattered ~30% of incident EUV, invalidating some prior studies and casting doubt on the idea that there is a diffuse corona at all.

C.E. DeForest, Astrophys. J. 661, 532.

Constant-width loops have been a conundrum since the 1970s.  After a lecture by Alan Title (speculating on why), I realized that the constant width might be an artifact. The paper is really about the limits of image interpretation in solar physics.  I've followed this up with several other projects, but this was my opening shot.

C.E. DeForest, C.C. Kankelborg, M.J. Allen, et al., Opt. Eng. 30, 1125.

This was my first refereed journal article.  The analysis was to figure how to determine plasma properties of solar features using an array of EUV telescopes.  It turns out, you can't -- which is why people now use DEM analysis, which allows for mixed-temperature plasma.  The 3-D spectral plots seemed so new and exciting back then.

Solar Polar Plumes

Working at the SOHO control room, I got really interested in polar plumes, since they are the quietest of quiet features in the corona -- if you can't explain plumes, you can't explain practically anything in the corona.  As it turns out, we couldn't explain plumes.  Truly new work on plumes (mostly not mine -- other very talented pepole have picked it up) continues to this day.

C.E. DeForest, P. Lamy, & A. Llebaria, Astrophys. J. 560, 490.

Phillippe Lamy and I got in an argument in 1998 over the lifetime of polar plumes.  Hours, or weeks?  It turns out both points of view were right, in a way.  In the meantime we managed to get an independent measure of the superradial expansion in the coronal hole, which agreed nicely with the white-light results.

C.E. DeForest, S.P. Plunkett, & M. Andrews, Astrophys. J. 546, 569.

Just how high up do polar plumes go?  Several groups were doing interesting, but conflicting and/or inconclusive, work on whether plumes make an imprint on the solar wind.  We put together a deep-exposure campaign with LASCO C-3 and showed that plumes extend at least 40 Rs from the Sun (in three dimensions).

DeForest, Hoeksema, Gurman, Thompson, Plunkett, R. Howard, Harrison, & Hassler, Sol. Phys. 175, 393.

The very first coordinated science campaign SOHO executed was JOP-39. In cruise phase on the way to L-1, Rick Bogart suggested we should off-point to look at the South pole.  I dove in to organize a big JOP to get joint plume data. Miraculously, everything went great, nobody's instrument broke, and we got a ton of good science.

A.B.C. Walker, C.E. DeForest, R.B. Hoover, & T. Barbee, Sol. Phys. 25, 1203.

I was a graduate student.  Art encouraged me to use the MSSTA film calibration on the 1989 rocket data.  I went a little farther:  I had just read Parker 1958 and fitted stationary Parker-like models to the data.  Of course, Parker flows are hydrostatic at those altitudes, so I could've saved some work -- but it was good learning.  

Wave motion on the Sun

A high point of my early career was spotting compressive waves in polar plumes in the EIT data.  Wave motion is universal and reasonably simple to measure with modern instruments -- but SOHO was the first platform sensitive enough to pick it up.  Nowadays it's used for all manner of plasma diagnosis as well as to explain energy and momentum transfer.

M.J. Wills-Davey, C.E. DeForest, & J.O. Stenflo, Astrophys. J. 664, 556.

Stenflo, Wills-Davey, and I had a very productive lunch meeting one afternoon, talking about EIT waves.  One outcome was a very nice theory of EIT waves as nonlinear fast-mode waves (solitary waves using the corona as a waveguide and MHD as the nonlinear element). I mostly kibbitzed, but I am still quite proud of the outcome.

C.E. DeForest, Astrophys. J. 617, 89.

I lifted some very high speed FUV data that the TRACE team collected, and drilled as deep as I could.  I found a very faint ridge-like spectrum in the k-𝜔 diagram.  The result is marginal and there are are reasons to doubt it result (as Mats Carlsson has pointed out).  We really need to get another imager looking in that frequency range.


L. Ofman, V. Nakariakov, & C.E. DeForest, Astrophys. J. 514, 441.

Leon cooked up a very nice 1-D MHD model and simulated the polar plume wave data from 1998.  He was the one who showed that these must really be slow magnetosonic waves: nothing else behaves quite that way.  Both Leon and Valery are terrific guys to work with.  Lots of fun, very sharp.


C.E. DeForest & J.B. Gurman, Astrophys. J. 501, L217.

This is the first direct imaging of waves in the corona.  Gurman and I both saw motion in the JOP-39 data.  We eventually decided it was waves because the speed was so uniform.  So much was going on at GSFC at that time – it was very exciting.  (E.g., Ofman may have priority on wave detection, with 2-altitude UVCS studies.)  


MHD Modeling

I've run off more than a few models of different kinds -- usually as chintzy as possible:  just enough to capture the kinematics or physics I'm looking for, and no more.  Charles Kankelborg and I developed the concept of fluxons more or less independently.  It turned out to be challenging to implement. Lately we've been using it to model solar wind flow with a little more rigor than WSA.  

L. Rachmeler, C.E. DeForest, & C.C. Kankelborg, Astrophys. J. 693, 1431.

This paper was a long time coming, partly because we had to vet the fluxon model, and partly because student papers always take time.  Rachmeler did a terrific job of demonstrating a new type of MHD instability, driving a stake through the Aly-Sturrock Conjecture (that magnetic forces can't drive CMEs without reconnection).


C.E. DeForest & C.C. Kankelborg, JASTP 69, 116.

This is the description paper for fluxon modeling, a concept both Charles and I were working on.  We merged our efforts, and by 2004 we had a working code that could solve toy problems.  We submitted this paper in 2005, and it took over two years to be published!  The fluxon code was dormant for a while but in 2019 we're using it again.


C.E. DeForest, C.A. de Koning, & H.A. Elliott, Astrophys. J. 850, 130.

For a long time, people have been trying to use polarization for 3D location of CMEs.  They've met varied success.  The key was SNR.  We found a  well-presented CME viewed by STEREO-A/COR2, and with noise gating the images were finally clean enough. We located the CME and verified its chiral sense with ACE.


C.E. DeForest, T.A. Howard, D.F. Webb, & J.A. Davies, Sp. Wx 14, 32.

As we were beginning to write the PUNCH proposal, we realized that there was still a lot of skepticism in the space weather community about whether polarized imaging would even be useful, even if technically possible.  We wrote this article to make the case in the Literature.  We did, of course, refer to it in the proposal.

T.A. Howard & C.E. DeForest, Astrophys. J. 800, L25.

In 2012 or 2013, Jeff Newmark challenged us to process a random HI-2 date with the background-subtbraction algorithm.  He picked a date and we ran it.  The first thing we saw was a crazy unusual long straight line -- like an AU long.  We puzzled over it for nearly two years and then finally wrote it up.  It's real, but still unexplained.  

C.E. DeForest, T.A. Howard & D.M. McComas, Astrophys. J. 745, 36.

This is the paper for which Tim and I started our whole heliospheric analysis project, connecting the first geoeffective CME of the STEREO era with its origins and release mechanism on the Sun.  This paper is broken out from Howard & DeForest 2012 (below).  We nearly split it again because the EUV analysis was so surprising.

T.A. Howard & C.E. DeForest, Astrophys. J. 746, 64.

This is our first paper tracing the 2008 geoeffective CME; it unified the full SECCHI field of view and led to our famous movie (GSFC video release).  We demonstrated that the coronal pre-CME cavity was preserved intact (though distorted) as a magnetic cloud at 1 A.U.  We followed up with a quantitative treatment in 2013 (above).

C.E. DeForest, T.A. Howard, & S.J. Tappin, Astrophys. J. 738, 103.

This paper kicked off quantitative solar wind imaging.  It arose from a discussion in Fall 2010: we had funding to study how CME onset affects morphology of the resulting ICME.  The HI-2 data couldn't yet constrain those models.  We beat on the starfield problem and cracked it.  Suddenly whole worlds of understanding opened up.

Magnetic tracking and the solar dynamo

I got really interested in solar magnetic fields around 2002, after a conversation with Clare Parnell. That led to an international workshop on magnetic tracking, and a series of really interesting papers that lasted until I got distracted by heliospheric imaging.  There's still a lot of work to be done.  Although most of these papers have other first authors, I was involved enough to feel quite pleased with the overall thrust of the resarch.

Peter, Bingert, Klimchuk, DeForest, Cirtain, Golub, Winebarger, Kobayashi, & Korreck, A&A 556, A104.

Jim, Hardi, and I started discussing coronal loop widths around 2011 or 2012.  Jim and I were sort of bogged down in wrangling over little details of the science, and Hardi and his student took the initiative (yay) and wrote up a rather nice treatment with input from us and several others.  I'm convinced there's more work to be done...

C.E. DeForest, H.J. Hagenaar, T.A. Howard, D.A. Lamb, C.E. Parnell, & B.T. Welsch.

This series was based around the Magnetic Tracking Workshop that we started in 2004, and cemented the idea of a frothy small-scale dynamo in which cross-scale dynamics (convergence and dispersal) are just as important as emergence and cancellation of flux.  Tim Howard came late in the sequence and helped get IV out the door.  I'm rather proud of the story arc this work revealed, even though Derek did most of the heavy lifting for the series (and is first author on most of the papers).


C.E. Parnell, C.E. DeForest, H.J. Hagenaar, B.A. Johnston, D.A. Lamb, & B.T. Welsch, Astrophys. J. 698, 75.

Clare extended greatly the work we'd been doing in the Magnetic Tracking Workshop, revealing that in fact the power laws seen in each instrument we used were limited primarily by the instruments themselves.  The power law implies that either there is one dynamo (not two) or the cross-scale linkage is strong compared to dynamo action.


S. Keil, T. Rimmele, & C.E. DeForest, Astro2010 White Papers, #153.

This was really an advocacy paper on why the ATST (now DKIST) was important to solar physics.  It turned into a really nifty mini-review of the state of dynamo physics and magnetic measurement techniques.  In 2019, I'm really looking forward to seeing what DKIST turns out once first light enters the Coudé room.


C.E. Parnell, C.E. DeForest, H.J. Hagenaar, D.A. Lamb, & B.T. Welsch, ASP Conf. Ser. 397, 31.

This was the first intercalibration of MDI and Hinode/SOT (NFI) data.  It came out of calibration work in MTW-3, our third Magnetic Tracking Workshop meeting, as we tried to get to the bottom of the weird Appearances problem (SMT II, above).  As always, Clare took the MTW stuff one step beyond.


D.A. Lamb, T.A. Howard, & C.E. DeForest, Astrophys. J. 788, 7.

Derek and I looked for several years for a correlation between existing flux and new emergences (sign of a local dynamo).  We didn't expect to find one, since flux subduction goes "all the way down", but a null measurement is a valuable constraint on dynamo models.  Derek developed a rather elegant statistical technique and a solid result.