Dear mr Towe,

I completly agree with you about the climate change fashion.

I didn't mean to look fashionable, but you are right, I threw it in the list as it came up on my mind, without too much thinking.

Anyway, the good thing of this method is that it allows the extraction of georeferenced lines and polygones (thus, distances and areas) from single ground-based historical phorographs. It has been done with historical postcards for example. As photography came earlier than aerial photography in many places of the world (much earlier than satellite imagery), the exploitation of such photographs might provide extensions to the analysis time baseline. Challenges are hard (approx locations where images were taken, DEM), but the potentials are high, in my view.

Ideally mr Towe, let's assume we have historical photographs taken by tourists from a famous lookout towering over a coastal embayment (sandy beach). These photographs date early 1920. What could be "vectorized" from these photographs to link with climate change?

Thanks for your observations mr Towe, I always learn something new.

best regards

Nic

Thanks for your points mr Towe. I am planning to do some tests later this month.

The location is Stockton Bight, New South Wales, Australia. It is a trasngressivre sand dune barrier of 34 km with massive transverse dunes.

I am planning to take pictures from lookouts on high promotories that enclose this embayment. I will take pictures with two different cameras (low-budget, medium-budget), track in-situ foredune positions via differential GPS, trace the foredune from the oblique photos from the lookout and compare its accuracy.

Might be the first step towards the validation of the WSL Monoplotting tool for coastal studies, before actually incorporating the historical-climate change related research with vintage photos.

We'll see how it goes.

Thx again,

Nic

Hi Nicolas,

this looks very intereseting for me, since I made my PhD on mapping sea level rise in the coast of my region (Andalusia, Spain), and I have been supervising a PhD about rephotography, which is the technique that I think that you are planning to use.

I have found huge problems in measuring coastal changes by means of oblique photos. Coastal areas are too changeable to quatify values: it is almost impossible to find a couple of photos in which tidal range, aire pressure or wind conditions are exactely the same (sea level depends on these three variables). Most of my work with my PhD student has been very qualitative, and we hve been phocused in checking the evolution of sand dunes, dunes vegetation, etc.

As SLR and coastal erosion researcher, I can tell you that you have to take into account that sediments are moving along the coast. It makes evern more complicated to take any reference os sea level rise, since you neer know the volumen of sand coming into your area. There have been some approaches using coastal profiles (since Per Bruun in 1956), but I wouldn't try to do that with common pics.

Nevertheless it is great to find someone working on this topics, so please contact me if you think you can improve our methods or if you have more questions.

Regards,

Pablo

Wow!

thank you so much Pablo for you enthusiasm in this topic, it fuels my curiosity knowing that others before me have tried this bumpy road!

I see excactly your points when you say that the coast is extremely tricky to quantify changes in a consistent and robust way. That is actually why I am studying the coast, it is challenging but rewarding at the same time.

Pablo, I will definetly contact you as soon as this idea becomes alive, that's for sure. At the moment I am stuck with a geospatial analysis of a coastal transgressive sand barrier, but in a few weeks, I hope I will get to the field for some fieldwork, and there I will test the oblique methodology.

Thank you so much for having replied to this question and for showing you support and ethusiasm.

Best Regards,

Nic

Great!

This is a very little conference paper that we made last year in our first approach to this topic. It is in spanish. Soon we are going to publish a chapter in english ina  handbook about sand dunes conservation. 

https://www.researchgate.net/publication/303544114_El_uso_de_la_fotografia_repetida_para_la_monitorizacion_de_cambios_morfologicos_El_caso_de_la_costa_de_Huelva_1986_-_2001_-_2015

Conference Paper El uso de la fotografía repetida para la monitorización de c...

Hi Nicolas, it is good to hear about people doing monoplotting with oblique photography!  Thanks to you and to the other specialists giving feedback.

I have no experience extracting features from this kind of material, but I have a bit of experience in doing monoplotting (with ILWIS-ITC, there is a free downloadable academic version) and working with traditional photogrammetric photos (close to vertical).  Also, if you have the opportunity, test the PHOTOMOD software (a russian product), that is very powerful (it includes UAVs processing); you can get freely  the full version but limited to some amount of photos (no more than 15 I think, or no more that some amount of megabytes of data, DEMs, etc). I can´t remember now if it has a monoplotting option,but it is very precise and efficient for traditional photogrammetric processing, also includes satellite imagery processing capabilities. It computes authomatically the tie points using correlation and performs the relative orientation of the photos fastly.

Regarding your question, if you need to test the methodology for oblique photos, I think that the coastal locations are probably the worst scenario (what  is precisely your need). You can test the methodology over cities or places in which you can  accurately measure the coordinates.

I have had good experience processing isolated photos (again, close to vertical) extracted from videos over coastal zones and getting precise planimetric coordinates from high resolution imagery (some GIS commercial software give you access to hi-res satellite imagery when you purchase their software). For the altimetric info I have had good luck with the DEM that I got.

I am speaking of  digital photos with resolution between 50 cm. and 1 meter. Perhaps your coastal oblique photo has 20 cm. or better, for which this kind of satellite sources will be not useful. In any case you need good GCPs...

Hope this helps a bit...

Best regards

Gabriel

Pablo:

Thx for your paper, I read it carefully, very interesting! I found the first part, when you write about the historical photography missions and related archives extremely useful. The only difference from your method is that I want to try to extract quantitative changes from the same kind of photography you guys used for change (erosion) detection. I will try to use monoplotting for it. But as you, Gabriel and Kenenth highlighted, coastal areas are the worst scenario. That is why I need suggestions on testing monoplotting accuracy in this complicated and extremely dynamic landscape.

Gabriel:

Thx for sharing your experiences and suggestions with us Gabriel, it does help, a lot.

I am not familiar with ILWIS-ITC because so far, as a student I only had the possibility to work with commercial GIS and image processing softwares (ArcGIS, ENVI mainly).

I will give a look at PHOTOMOD lite version, if I can stay below their limitations.

Yes, I will not work with near-vertical aerial photographs but with oblique ground-based ones. As I will be able to go into the field (the scene I will capture with the camera) and record highly accurate GPS points, I will not need satellite imagery for georeferencing.

Also, I will use highly accurate dGPS for acquiring GCPs and UAV DSM to correct and place the oblique photos into real world coordinate.

 But what you did is very intresting to me, because with high resolution satellite imagery and a video over the surveyed area you actually didn't need to go into the field and acquire GCPs (as you extracted from satellite, am I right?).

Very interesting for remote locations.

But how did you get the DEM for monoplotting in these remote areas? SRTM or ASTER dems?

Thx for your points guys!

Best regards

nic

Nicolas:

I worked several years with ARCGIS and ERDAS Imagine, and now with ERDAS 2014 (Hexagon). I have less experience with ENVI, although my colleagues at the university have a strong experience with ENVI and IDL programming, then I learn a bit about ENVI with them.  

But I mentioned ILWIS because I like this software with which I started doing something in GIS on early 1987. And I was doing monoplotting, that was not provided at that time by ERDAS or ARCVIEW (that was ARCINFO really on 1987...), using a DEM that I´ve done by contour interpolation in ILWIS, the unique source for a DEM that I had at that time. 

Talking about the present, I used SRTM-1 (30 meters resolution). And the GCPs were extracted from high resolution satellite imagery, you are right. 

Regarding PHOTOMOD Lite, there are five or six complete datasets(with all the working files) for testing and learning (including one for UAVs) that I downloaded from the site. I can´t remember if this software has the capability for doing monoplotting , but it will be useful for you as a photogrammetric tool if you need it.

Best regards

Gabriel