Animation forms a key component of many, if not most, AAPG and SEG oral presentations. The classic animation is of an uninterpreted well log or seismic data followed by one labeled with formation tops or seismic picks, faults, and lithology types. Other common animations include an animation through a seismic data volume or the evolution of a basin. While such animations usually are simple to generate, they can rapidly and unambiguously support the presenter’s message.
Although Interpretation has provided a format to include animations in the online (vs. print) version of the journal since its inception, few examples have been published, and most readers do not know the possibility of animations exists. While many readers access the print edition of Interpretation, the majority of our memberships prefer to access the journal electronically, particularly those who do so through university, institutional, or corporate library subscriptions.
Even though the first author is the editor and the second author is the SEG digital publications manager, we know that the Interpretation “Instructions to Authors” is not on everyone’s must-read list. Indeed, we suspect most authors read these instructions at the final stages of their work, making sure they have checked all the publication requirements, such as font size, citation of figures, equation format, and reference style. If they encounter the sections on animations, it is actually too late because such animations need to go through the same peer-review process as a figure.
We also think most authors look to recent publications, perhaps on a topic similar to their own, rather than to the more abstract “Instructions to Authors” as a go-by, particularly for matters of style. If animations are rare in the online version of a published journal, how would these folks know this format is available? For this reason, we will use this column to add some animations provided by current and former students and colleagues that they “wish they had submitted” as part of recently published papers. Given the editor’s background, these animations will be biased toward reflection seismology. However, we can easily envision animations of tectonic plate movement, palinspastic reconstruction, multiple geostatistical realizations, fluid migration through porous media, and continuous monitoring of hydraulic fracture microseismic events.
The simplest example of animation in an oral presentation is to successfully add layers of interpretation. In an online publication with more time to examine the images, the reader can more closely examine and validate (or invalidate!) the authors’ interpretation. Figure 1 is an animated version of vertical images published by Khatiwada et al. (2013) as two figures. We think an animation is more effective.
The equivalent oral presentation by a seismic processor will show slices through the seismic data before and after application of particular filter. Figure 2 is an image that Machado et al. (2016) had hoped to put in their paper but mistakenly thought that they did not need to go through the review process (which it does).
Don Herron provided the animation shown in Figure 3 that was originally provided as still images, one above the other, in Herron (2013). He often works with multiple migrated images of the same data, using different algorithms provided by the same company, each of which has their interpretation advantages and disadvantages. Animation allows the reader to validate the author’s observations.
The next simplest example and one used routinely by seismic interpreters is to animate through a suite of vertical, time, or horizon slices through a seismic-amplitude, seismic-attribute, impedance-inversion, or classification volume. Figure 4 shows an animation through peak spectral frequency modulated by peak spectral magnitude corendered with coherence that Tao Zhao wishes he had taken the time to publish in his (Zhao et al., 2016) paper.
One objective of interpretation is to extract objects, be they faults or geobodies. Figure 5 shows a an animation loop that Machado et al. (2016) wishes they had published that shows enhanced faults that are color-coded by their dip magnitude and dip azimuth.
The next two images come from presentations from the editor’s colleagues at the University of Oklahoma (OU) that they hope will become part of future publications. Figure 7 is made from a suite of time-lapse photographs of a sand model showing the surface deformation of a thrust fault. When combined with vertical slices through the sand model (only available at the last stage), one better visualizes the history of deformation.
Figure 8 shows the changes in pressure in a reservoir simulator. Each snapshot took almost one day to compute.
The mechanics of generating an animation file
Converting PowerPoint 2010 and 2013 animation file to an mp4 file
Because most presentations at the SEG, AAPG, and URTeC meetings use Microsoft PowerPoint, this is the simplest place to start. Fortunately, the 2010 and 2013 versions of PowerPoint make conversions of a PowerPoint slide using animation to an mp4 video format simple. This method was used to create Figure 1. With the desired slide in your viewer, select “File,” then “Export” rather than “Save” or “Save As.” In the next window, select “Create a Video.” In the right side of this window, the default “Presentation Quality” will be “Largest file size and highest quality” which appears as on the laptop used. Instead, click the dropdown menu and select “Internet Quality, Medium file size and moderate quality,” which on the laptop used appears as . You can use your previous PowerPoint timings and narrations. We will address voiceover issues in the next “From the Editor” column where we will promote audio abstracts. Instead, select the time you wish to have between each animation frame. In Figure 1, we chose 3 seconds. Finally, after having selected the time increment, click “Create Video,” define a file directory and name, and save it. Click on the file you have saved to make sure it plays the way you intended.
Because we are pushing the concept of video, we provide the same instructions as an mp4 video in Figure 9 and a supplemental file s10.mp4. The supplemental file s10.mp4 includes instructions on generating an mp4-format animation file shown in Figure 2 from a suite of images in PowerPoint 2003 or 2007 format. These images are first saved as a suite of jpeg-format images and then linked into an animation using Microsoft Moviemaker.
Figure 2 and 8 were generated from a suite of screen captures, while Figure 7 was generated from a suite of photographs. Following the simplest oral presentation style, we inserted one figure per slide. To generate an mp4 video from Figure 2, we again selected the “Export” to “Make a Video” option, set the timing between images to be 5 seconds, and saved the results. Details of this process are found at the end of the animation shown in Figure 9.
Capture of animations in commercial interpretation and processing software
Almost all 3D seismic interpretation software packages have animation capabilities that allow the interpreter to progress through successive vertical or time slices at a user-specified increment. While some software allows direct export of animation files, the one the first author uses does not. There are many free animation capture programs available, including aTube Catcher, Jing, and ezVid. We used “Snagit” to make the images in Figures 4–6. The key step is to slow the animation down so that you do not capture the computer image as it is “refreshing.” In Petrel’s animation “player” one can explicitly control the number of seconds used to display each frame, with the default being 100 ms, or . The Snagit default capture speed is . Depending on the complexity of your image (increasing with the number of co-rendered volumes) and the memory of your computer, the player will flicker as it reads in a new block of data from disk. These flickers are captured by Snagit, corrupting your animation. An easy solution (used in Figure 4) is to slow both the display and capture rates to , with the captured video replay rate at a faster . In contrast, we captured Figure 4 by manually moving the boxprobe. Here, to avoid the refresh artifacts, we set the Snagit capture rate to be () and slowly moved the image, waiting 5 s before each move. This takes a little practice but is still an easy way to generate an mp4 video. To demonstrate the image refresh problem, Figure 10 shows the same data volume where we used the default capture rate. Although we used a large memory computer, note the images are blurred as we move them. A simpler way to generate a movie is to simply make a series of still images and link them together as described in Figure 9 and supplemental file s10.mp4. Supplemental file s11.avi is a 3D visualization of the same data volume shown in Figure 5 but using the Snagit default image capture rate of . Note that the images are blurred until the movement stops, generating an inferior animation.
We strongly encourage authors to include animation in their papers when animations could help to clarify or better demonstrate scientific arguments. If you do not submit an animation with your original submission, you will have an opportunity to include one following minor or major revision.
Costs of saving such digital files is minimal. However, Interpretation knows that formats will change in the future. For this reason, only those formats in the “Instructions to Authors” will be acceptable. The Interpretation staff will assure that these files are updated to a new format when mp4 and others become obsolete.
Let’s get animated!
The current issue
The May 2016 issue contains 34 papers, which includes 11 technical papers and one paper in the “Tools, techniques, and tutorials” section.
1) Advanced logs and interpretation; editors: Yao Peng, Vivek Anand, Burkay Donderici, Tie Sun, and Xiaogang Han; 12 papers published.
2) Unconventional exploration and production — Achievements and remaining challenges; editors: Huyen Bui, Thomas Klopf, Hongliu Zeng, Robert Wiener, Dario Grana, and Rodney Johnston; 6 papers published.
3) Seismic data conditioning; editors: Doug Cook, Paul Constance, Scott Singleton, and Peter Harris; 4 papers published.