Last year, Matt, Romain and I decided to unite our efforts in researching the disappearance case of Kris Kremers and Lisanne Froon, the two women that disappeared in Panama in 2014. Altogether, we’ve written numerous articles, discussed the case for dozens of hours, and even have a trip to Panama planned for later this year.
After a lot of collaboration and persistence, our team has finally unearthed new case data.
In this article, Matt breaks down the new photo data that’s never been seen publicly until now. The photos he’ll discuss are the daytime photos – the photos that Kris and Lisanne took while hiking the El Pianista trail prior to their disappearance. This is new information, directly from the original photo’s EXIF records. Further, he discusses the content of the missing “day photos” and lays out the entire timeline in graph and table form.
This new case information, in conjunction with our trip to Panama later this year, should yield some important conclusions.
New Photo Data Released
This article discusses the day photos, which were taken on April 1st, 2014. I like to thank Romain (firstname.lastname@example.org) for providing comparable photos of El Pianista Trail and information of image locations on the trail. Please also check out Romain’s article on the day photos (link here) where he wrote from the perspective of having hiked the trail twice.
The data provided in this article is based on the original images and their time stamps, which I was given access to for review. The data from those images is presented below. Any intentional or unintentional errors in the original data will be present in this article. I will refer to the timestamps as “correct”, which refers to the data in this article matching official data.
Please note that, many details around those images have been discussed exhaustively and I will not re-discuss those same points. I may only mention those points briefly. The scope of this article is to make the new data available, rather than discussing new and old theories.
The new data is:
- Correct image number for all images
- Correct time stamp for all images
- Description of content of unpublished images
The remaining 11 images or 33% have never been made available to the public.
Some of the existing public images had embedded meta data, while the majority did not. With new data available to me, I was able to assign times to all 33 images and place them on a timeline.
Please note, that I am not able to publish the actual missing 11 images. However, I will give a description of what they show and since all images were taken in groups of at least two images, it is very easy to visualize what the missing images show.
Table 2, shown at the bottom, shows a summary of all 33 images, the time they were taken and a description of the content. The table also shows the correct image numbers for the previously available images.
Firstly, I will review the images for possible signs of manipulation. For images to be manipulated there would have to be a motive. Since a lot of speculation has been around this case and the images in particular, I will not elaborate too much on it and can simply say that if someone had tried to cover up 3rd party involvement, they would have motivation to manipulate some or all images to obscure evidence or to mislead investigators.
If a motive exists, the ways discussed below of manipulating the images are possible. Evidence of image manipulation, depending on its degree, would be suspicious and may indicate 3 rd party involvement. No evidence of image manipulation could mean that it is either undetectable or was not done but on its own is not evidence against 3 rd party involvement.
One can crop out parts of an image that shows anything unintended. This can be detected in two ways. First if the cropped image has lost relevant content compared to the original, it is an obvious sign of manipulation, assuming an unedited original is available. In addition, the image pixel size of the image will reduce every time it is cropped. This is not easy to detect if the image is upsampled again and if only small portions are cropped off.
If the image is not cropped carefully, it is possible to change the aspect ratio of the image. An absence of change in aspect ratio is no evidence that the image was not cropped but a change in aspect ratio is evidence that the image was cropped.
The Canon SX270HS that was used to take all 33 images produces images of 4000 x 3000 pixel at highest quality and an aspect ratio of 1.33. Thus, any image with an aspect ratio other than 1.33 was obviously cropped.
Of the 22 previously available images the following images have an aspect ratio other than 1.33:
Images 476, 481: 485, 489 are screenshots of a TV screen and obviously cropped and no further investigation is necessary. The original that I have reviewed do not show signs of being cropped.
Image 497, 498: Those images were very slightly cropped. In comparison to the originals, nothing of interest was cropped and is likely that they were cropped for “artistic” purposes or to fit an image template better. The originals do not show signs of being cropped.
It can be concluded there are no signs of manipulative cropping and compared to the originals, no relevant image portions were cropped off the known images. However as mentioned above, I can only compare to the originals available to me and assume those were not slightly cropped. In summary, it is very unlikely that images were manipulated by cropping and what we see on the images is the complete image.
2. Image Deletion
One can delete entire images out of a series of images to make unwanted content disappear. For this to go unnoticed one has to either delete the first or last image or if deleting images within a series of images, one would have to renumber all images subsequent to the deleted image(s).
It is possible that this was done with image 509 which is missing and was not even recoverable from the memory card. It is unknown if this image was deleted by Kris or Lisanne after image 510 was taken or if it was secure deleted to make restoration impossible.
It does not appear as if Kris or Lisanne deleted images since no other image numbers are missing. In image 495 Kris has her eyes closed and appears to be moving her hair with her right hand. This image would be one she would have wanted to delete, especially since there are several like images taken right after that look acceptable. It is therefore unlikely that image 509 was deleted by Kris and Lisanne because they didn’t like it, simply because they did not seem to have deleted other images that seem unlikeable. Image 509 might have taken at a time where there was a problem and under those circumstances there would not be motivation to delete images.
The other option is that the image was deleted by a 3 rd party. We then must assume that the motive was to prevent anybody from seeing what was shown in this image. This on its own is not evidence of 3 rd party involvement since someone may simply not have wanted to be the target of an investigation, however it makes some 3rd party involvement a possibility.
Lastly, the image could be missing due to accidental deletion or technical defect, but those are both rather unlikely. Digital cameras are essentially computers and it is not likely they just miss count of images. Accidental deletion also seems unlikely since for that the Canon SX270HS requires 4 small and rather hard to locate buttons to press, including one to acknowledge a delete warning. For this to happen unintentionally is nearly impossible.
A further problematic detail is that this image was taken between the last day photo, image 508, the first 112 call and before the distressing Night photos. It is therefore entirely possible that image 509 showed something that someone did not want to be seen. In summary, it is possible that at least one image was deleted and possibly more (see 4).
3. Pixel Manipulation
It is possible to manipulate images on a pixel level (“photoshopping”), however this is very difficult to do in front of busy backgrounds and with hair showing in the image. Hair is notoriously difficult to photoshop with and if subjects are to be placed into images, it is the biggest issue to get the hair to look right. It is possible for someone with expert knowledge to make some manipulations, however I do not see any evidence for this to have happened.
So either it was done extremely well by an extremely good expert or it was not done. I believe the latter is true and there was no pixel level manipulation done.
Especially since one image was deleted, one could have deleted another one without raising any more suspicion. It is much easier to simply delete images with unwanted content rather than performing difficult to do and relatively easy to detect pixel manipulation. The only motivation for pixel level manipulation would be to place something into the image that wasn’t there originally. However there would be little motivation because there are enough natural images of Kris and Lisanne at the Mirador.
In summary, it is very unlikely that images were manipulated on a pixel level.
4. Meta Data Manipulation
All digital images contain meta data that the camera writes into the image. In this case the most interesting would be the date and time the image was taken based on the camera’s clock as well as “last edited” data (to me only the time the image was taken was available).
This data can easily be manipulated, and this may not be detectable by anybody. It is possible that the camera writes checksums or other identifiers into the meta data to allow detection of manipulation. But this knowledge would be privy only to Canon experts or possible government Intelligence Services.
Another possibility of detecting amateurish meta data manipulation would be if it is impossible to place the image timeline in the physical environment. For example, if the data of two images shows them being taken at two places that can not be reached within the time difference between them that is taken from the meta data.
Thus, if the original meta data was manipulated, it might not be possible for us to detect it.
One example where the time difference between images is difficult to align with physics is image 500 and 501. Only 12s lapsed between those two images were taken, which is not much time since Kris and Lisanne had to switch spots, trade the camera, focus and take another image. Which is pretty not much time to do all those steps.
I have conducted an experiment. The assumption is that, the subject is approximately 4m ±1m away from the camera. At normal pace but with effort taken to be quick, it took two test subjects 14s to take comparable images of each other. It can be concluded that the time of 12s that Kris and Lisanne took is possible but very fast and one wonders if there really was such reason for the haste to save seconds.
Another explanation could be that those images were taken by a 3rd party, which on its own is not evidence for 3rd party involvement as it could have simply been another hiker. Also noteworthy about images 498, 500 and 501 is that Kris and Lisanne are making, for them, untypical and probably for young people their age rather untypical “thumbs up” hand signals. Could this be a connection with a 3rd party who took those pictures? It can not be said and has been discussed before, so I will not discuss this theory further.
Meta data manipulation would also need a motive which could be that unrelated images are being inserted or removed from a series and the time and date and subsequent images are corrected to fit.
This however was not with image 509 that seems to have simply been deleted. If someone deleted one image without correcting the meta data of the remaining images why would they have put more effort into deleting other images?
It is also possible to manipulate image data and time data to make it appear as if they were taken at a certain date / time while in reality they were taken at a different date / time.
For this to be done realistically with only a few images, it would be required to match the locations, lighting conditions and looks of persons shown and it would be easier to manipulate an entire series of images to make it appear as if they were taken on a different date or time. While the former appears unlikely, the latter is not impossible but not likely.
One such example is image 501 (left) where it has been speculated that it does not fit with the other Mirador images because of the clouds not matching and the light being wrong since it appears that there was no harsh sunlight when images 501 was taken.
I do not believe those points to be a concern. The clouds do in my opinion match reasonably well as we can
seen in the following collage:
The images are 501 (left) and 499 (right), taken 18 seconds apart, so in regards to cloud movement, virtually at the same time. It appears as if one image was taken under cloud cover and the other under blue sky. However when looked at closely we see that the clouds in the image to the left end in three “fingers” and the image on the right has also three pieces of clouds that could connect together. In my opinion the clouds do match reasonably well.
Another problem is the light. It appears as if Kris is under diffuse light under clouds while Lisanne is lit by harsh sunlight from the right.
A possible reason for the discrepancy could be the “flat” colors in the left image caused by the low contrast in the image and I have manipulated the colors to make the image look more as if it was taken in sunlight.
If we look at this image it looks like similar lighting conditions as the right image. We also see harsh shadows on Kris’ face. Kris and Lisanne face in different directions which explains while Lisanne is in sunlight from her right side and Kris isn’t.
Although the lighting and clouds can be explained, I am not fully satisfied that everything is right in those two images, but it is not in the scope of this article to explore this further.
I would also encourage the reader to review Juan’s analysis (link) on some of the bushes where it appears as they are in different stages of rolling their branches. This is difficult to explain and I recommend reviewing Juan’s excellent analysis on this subject.
Another concern are images 493 and 494. Romain suggests those depict locations AFTER the Mirador, however on the timeline they are placed BEFORE the Mirador. Since those images were taken approx. 1.5h after start of the hike it would be nearly impossible that Kris and Lisanne passed the Mirador and then returned to the Mirador as the image time could not be placed in the physical environment due the fact that it is impossible they reached the Mirador so quickly. However, I do not have confirmed evidence about where those two images where taken and therefore do not want to speculate further until their location is established.
It is also noteworthy that Kris and Lisanne reached the Mirador exceptionally quickly. Romain who has hiked the Pianista twice, estimates that it takes between 1:40h and 3:00h to reach the Mirador (his fastest time was 2:00h).
Based on his experience, there is one location (see image 3) where it is possible to take a wrong turn and the wrong turn (red arrow) could be more intuitive to most hikers (it happened to him on his first hike) since it leads uphill, while the correct path (green arrow) lead downhill and looks like a creek rather than a trail.
It took Kris and Lisanne 1:52h, since they would have started at 11:08 and took the first photo on the mirador at exactly 13:00. This also assumes they reached the Mirador and took photos right away, while in reality they may have spent a few minutes first enjoying the view.
The time they took is very short and required that no wrong turn on the trial is taken. Although it is possible to have happened, there is some doubt and it can be wondered if they had someone with them that knew the trail and also paced them.
In summary, it can be concluded, that there are some areas of concern about the timing seen in those images. This alone is not evidence for 3rd party involvement though as there might be explanations for the issues I have pointed out.
5. Fake Image Insertion
In order to manipulate a series of images one can insert unrelated images to suggest that the photographer was at certain times and certain places. This would have to go along with point 4 (Meta Data Manipulation) and therefore does not need to be discussed further here.
Descriptions Of The Missing Images
All 33 images were taken in 8 groups of 2 – 10 images each and show the same subjects or scenery from slightly different angles within those groups. Therefore, if there is one image from a group available, it is very easy to visualize what the missing ones show.
Since I cannot publish those images, I am giving a description of what they show.
|Found Photo||Photo Description|
|478||This image shows scenery very similar to 481. Please refer to Table 1 to find the group of images 478 belongs. It shows the same landscape and features.|
|480||This image shows scenery very similar to 481. Please refer to Table 1 to find the group of images 478 belongs. It|
shows the same landscape and features.
|487||This image shows a small creek|
|488||This image shows a small creek.|
|490||This image shows the same location as 491.|
|491||This image shows the trail and the same stream as 491.|
|495||This is the first image taken on the mirador|
This image is a selfie like 498. Kris touches her hair/neck and has her eyes closed.
|496||This image is a selfie like 498.|
|503||This image shows the same scenery as 504|
None of the previously unpublished images shows anything that is particularly noteworthy or not visible in a
Image 5 shows the timeline of the events on 01 April 2014. Since all images are taken in groups, they are placed in groups on the time line. The green arrow shows the time window in which a person who started the hike at 11:08 should have reached the Mirador. Kris and Lisanne reached the Mirador very quickly.
I also indicated a possible time span in which image 509 could have been taken. Take in mind though, that this is pure speculation and it could have been taken much later. I assigned a general description of the general area the images were taken in. The location of images 493 and 494 is described as “unknown” until established on site. The timeline places them before the Mirador however Romain believes they could have been taken after the Mirador.
Table 1, below gives an overview of all images and all times.
|Image||Event||Location||Time of Day (Panama)||Lapse Time from Start |
|Time between Events |
|N/A||Start||Trail Head||11:08:00||0:00:00||N/A||Time estimated|
|Image 476||Creek||11:18:24||0:10:24||10:24||Image previously published. This mage is actually portrait orientation and cropped into landscape. The sky and mountain top is missing.|
|Image 477||Creek||11:18:34||0:10:34||00:10||Image previously published|
|Image not published||Image 478||Creek||11:18:38||0:10:38||00:04||This image shows scenery very similar to 481|
|Image 479||Creek||11:18:44||0:10:44||00:06||Image previously published|
|Image not published||Image 480||Fields||11:25:04||0:17:04||06:20||This image shows scenery very similar to 481|
|Image 481||Fields||11:25:10||0:17:10||00:06||Image previously published|
|Image not published||Image 482||Fields||11:25:26||0:17:26||00:16||This image shows scenery very similar to 481|
|Image 483||Fields||11:25:30||0:17:30||00:04||Image previously published|
|Image not published||Image 484||Fields||11:25:32||0:17:32||00:02||This image shows scenery very similar to 485|
|Image 485||Fields||11:25:34||0:17:34||00:02||Image previously published|
|Image 486||Fields||11:25:44||0:17:44||00:10||Image previously published|
|Image not published||Image 487||Open Trail||11:42:14||0:34:14||16:30||This image shows a small creek|
|Image not published||Image 488||Open Trail||11:42:18||0:34:18||00:04||This image shows a small creek|
|Image 489||Open Trail||11:42:26||0:34:26||00:08||Image previously published|
|Image not published||Image 490||Jungle Trail||12:02:40||0:54:40||20:14||This image shows the same location as 491|
|Image 491||Jungle Trail||12:03:08||0:55:08||00:28||Image previously published|
|Image not published||Image 492||Jungle Trail||12:03:16||0:55:16||00:08||This image shows the trail and the same stream as 491|
|Image 493||Unknown||12:42:24||1:34:24||39:08||Image previously published|
|Image 494||Unknown||12:42:32||1:34:32||00:08||Image previously published|
|Image not published||Image 495||Mirador||13:00:22||1:52:22||17:50||This image is a selfie similar to 498|
|Image not published||Image 496||Mirador||13:00:36||1:52:36||00:14||This image is a selfie similar to 498|
|Image 497||Mirador||13:01:06||1:53:06||00:30||Image previously published|
|Image 498||Mirador||13:01:14||1:53:14||00:08||Image previously published|
|Image 499||Mirador||13:01:38||1:53:38||00:24||Image previously published|
|Image 500||Mirador||13:01:44||1:53:44||00:06||Image previously published|
|Image 501||Mirador||13:01:56||1:53:56||00:12||Image previously published|
|Image 502||Mirador||13:02:00||1:54:00||00:04||Image previously published|
|Image not published||Image 503||Mirador||13:06:14||1:58:14||04:14||This image shows the same scenery as 504|
|Image 504||Mirador||13:06:20||1:58:20||00:06||Image previously published|
|Image not available||IMG_2125 iPhone4||Mirador||13:14||2:06:00||07:40||Based on timing similar to 20140401_131430 it is assumed the image was taken on the Mirador|
|Image not available||IMG_2127 iPhone4||Mirador||13:14:00||2:06:00||00:00||Based on timing similar to 20140401_131430 it is assumed the image was taken on the Mirador|
|Image not available||20140401_131420 Samsung S3||Mirador||13:14:20||2:06:20||00:20||Based on timing similar to 20140401_131430 it is assumed the image was taken on the Mirador|
|Image not available||20140401_131424 Samsung S3||Mirador||13:14:24||2:06:24||00:04||Based on timing similar to 20140401_131430 it is assumed the image was taken on the Mirador|
|Image not available||20140401_131430 Samsung S3||Mirador||13:14:30||2:06:30||00:06||This image shows similar clouds and scenery as Image 502|
|Image not available||20140401_131456 Samsung S3||Mirador||13:14:56||2:06:56||00:26||Based on timing similar to 20140401_131430 it is assumed the image was taken on the Mirador|
|Image not available||IMG_2128 iPhone4||Mirador||13:15:00||2:07:00||00:04||Based on timing similar to 20140401_131430 it is assumed the image was taken on the Mirador|
|Image not available||20140401_131513 Samsung S3||Mirador||13:15:13||2:07:13||00:13||Based on timing similar to 20140401_131430 it is assumed the image was taken on the Mirador|
|Image 505||Continental Divide||13:20:32||2:12:32||14:12||Image previously published|
|Image 506||Continental Divide||13:20:38||2:12:38||00:06||Image previously published|
|Image 507||Continental Divide||13:54:50||2:46:50||34:12||Image previously published|
|Image 508||Continental Divide||13:54:58||2:46:58||00:08||Image previously published|
|N/A||112 Call||Continental Divide|
|16:39:00||N/A||N/A||First 112 call.|
Note that the official time that this time is based on might be off by 1h. This is not relevant for this article.
I’d like to thank Matt and Romain for their excellent work on this research. We’ve all spent dozens of hours analyzing this information and preparing it for publication.
After photo 508, the last day-time photo shown above, photo 509 was deleted from the camera entirely, for unknown reasons. Beyond that point, the peculiar night photos were next in the camera. Matt discussed the night photos heavily in his other article: A Deep Analysis of The Night Photos.
Matt can be reached by emailing: email@example.com.
We have more information coming out in the coming weeks. A big thank you to all of our readers and supporters for helping us to keep this investigation alive. Feel free to post a comment or send us a message.