2016 - A year in review

Not a bad year overall and thanks to the readers of the blog.

A new team member, 3 peer reviewed journal papers accepted, awarded over €200k in funding, one project finished, one project approaching successful conclusion, a new project kicking off in Mar 2017, an ERC application submitted, multiple conference and workshop presentations, an MSc class graduated, 3 trade journal articles, an RPAS consumer guide and some big Copernicus news to be announced in the new year from Tim.

Roll on 2017 and lets see if we can get even more done.

Christmas Tree Revetment

What a great use of old Christmas trees. These are placed along coastal or river regions experiencing erosion and help the native grass (maram or otherwise) to take hold again and keep the bank together. The example in the link is for the River Dee and fisheries board - I have not come across many examples of this applied in coastal regions in Ireland although plenty online for the US. I wonder is their any harmful side introducing that many Christmas trees to a river area - like changing the pH of the soil. Possible that only happens with live Pine.

The Fall of Hyperion

Bad news on the Hyperion front. 

I posted in a burst of enthusiasm earlier in the year about free satellite hyperspectral datasets that I had found for Ireland. The Hyperion sensor on the EO-1 satellite was recording long strips over agricultural areas in Munster and Connaught. NASA had even agreed to let users request it to capture new sites of interest - which i did almost immediately for a number of our coastal projects around the country. Unfortunately due to a GPS malfunction on the satellite and also I am presuming problems with cloud cover, none of my coastal test sites were ever successfully mapped. Really disappointing as this would have been an invaluable source of hyperspectral imagery for Ireland. To make matters worse - we won't have a chance for a repeat request as the satellite is being decommissioned this month. 

No more free hyperspectral satellite tasking for geospatial researchers.

Propaganda Maps from the 20th Century

Some excellent WW1 and WW2 propaganda maps here compiled by the National Geographic - you really need a monitor to appreciate them - it won't load properly on a phone. I could look at these for hours, there is just so much going on. Would be nice prints for a wall.

What is in the hands of "Ireland" in this WW1 one? A bottle for sure - but what are the tweezers? 

Portugal looks quite content in this one - only country in Europe that is.

And is this rifle supposed to be a scale bar?

Drone Prosecutions

At any of the UAV/UAS/RPAS/SUA/Drone events one of the major talking points has been regulation and prosecution of unlicensed operators. This is a major concern, whether it was at the kick off meet of the UAAI, or the SCSi Working Group on RPAS, or at Survey Ireland, or Drone X. Each week videos pop up on youtube which have been captured in clearly unsafe conditions, by unlicensed operators. This understandably annoys the operators who are playing by the rules and the example is regularly given of Director A who needs 30 seconds of aerial footage for the next episode of their show but needs it this week. They ask Operator A who has a license but cannot get the required permissions in time. He regrettably turns it down,. Operator B who has no insurance, no regard for safety or proper operating procedure gets the 30 seconds of footage and so the business. The word on the street was that a prosecution was coming - but the news broke in the last few weeks that the DPP thought a drone prosecution was too heavy handed. This makes me concerned that we will not get on top of regulation any time soon and the people who play by the rules will be the losers. We have had a drone crash - total IMU failure - and others will too so these safety procedures are essential. As a colleague of mine regularly says - "if you have never crashed your drone you're not flying enough".

I remember coming up to Christmas last year the rush to get the new regs in place as the IAA were expecting a flurry of drone purchases over the holidays (I even saw one for sale in the pound shop). It seems over 6,000 drones have been registered in 2016 alone! 

Hydro International article online

Our article on 'Improving satellite derived bathymetry' is now online. This covers some of the work with GSI and TechWorks Marine but also the more recent INFOMAR shortcall 2015 results.

Kildare FM

Of course it wouldn't be Christmas without my annual call with Kildare FM about our role in 'mapping for santa'. Tune in tomorrow at 14:20 to hear all about it.

Flood Damage 2015

What a difference a year makes. Sun outside, hardly any rain in weeks - cold, with a little mist  - basically typically beautiful Christmas weather on the north campus (although I am aware that there are weather warnings in place for later in the week - so this does not count as jinxing us).

December 2015 saw me fleeing a Sci:Comm conference in Athlone at the head of some of the wettest weather for winter on record, resulting in the activation of the Copernicus Emergency Mapping Scheme for areas along the Shannon catchment and elsewhere on the 8th December 2015. Readers of this blog might have come across some of the work we did on flooding this year, demo-ing the capabilities of EO data and the archive of Copernicus Emergency Mapping in articles for the SCSI Surveyors Journal and RICS Land Journal, but also in three conference presentations and a Society of Chartered Surveyors CPD talk. I came across an interesting post yesterday on Irish Economy blog summarising the financial cost of the event quite well.

"Counting the cost of last winter’s flooding: Evidence from disruptions to the road network"

Some sobering statistics on the cost of that one event (albeit a prolonged one) and great use of maps and distance/travel metrics to display the financial cost of the flooding to commutes etc.

Galileo goes live

Great news for people standing in the rain doing fieldwork - the EU constellation of positioning satellites (Galielo) are now  offering initial services which adds to the existing GNSS constellations like GPS and GLONASS, etc. More satellites means better accuracy (no more 'poor PDOP' - an in-joke if you've ever used a Trimble rover), less chance of dropping below the minimum required to calculate 3D position and faster initialisation times too so you can get out of the rain even sooner. With eighteen satellites in orbit now they need 24 for the full constellation (2 failed in launch on a Soyuz rocket early in 2016) and will be adding to functionality over the coming years.

Anyone interested in more detail on GNSS satellite orbits - I recommend this neat webpage - stuff in space. If you search for Sentinel 2a, Meteosat and Galileo you will see examples of three main orbits, polar (low earth orbit), geostationary and the one that Galileo is in, a medium earth orbit.

Paper Accepted for Publication - Environment and Planning B

The outputs from this year really are clumped in the final two months - two papers and two funding awards - but I suppose that's down to chance with a paper review process of undefined duration and also two funding decision deadlines at the end of the year.

We've just been informed that after some minor revisions our ESPON paper, "Data Imputation in a Short-Run Space-Time Series: A Bayesian Approach" has been accepted for publication in "Environment and Planning B". I will post a link when I get it.

Change Detection - Funding Success

More good news on the funding front - although like the previous post not formally announced yet. Approx €150k for a 2 year study looking at change detection and change classification combining orthoimagery, image-derived point clouds and remotely sensed satellite imagery.

More to follow.

Port Management - Funding Success

I was successful in a recent funding call - approx €60k for applications of RS data/GIS/decision support tools for port management/marine spatial planning in Dublin bay. This will involve working with a Dublin-based planning company on the project and runs for two years.

More to follow.

New Paper Published

The final paper demonstrating the potential of MIMIC for optimising Mobile Laser Scanning System configurations,

"Improving MMS Performance during Infrastructure Surveys through Geometry Aided Design"

has just been published. We never used the VQ-250 for coastal surveys - but i would love to have had that dataset for validation against the drone surface models we made for the dune erosion in Portrane in 2012. 300,000 points per second! A coastal erosion study would have been the ideal test case for MIMIC - using the software to find the optimal FOV and scanner orientation to provide maximum point density on the dunes - concentrate all sensors on one side of the system for a single pass at the right tide level.

LiDAR as Art 2017

The final five LiDAR as Art finalists are up online for voting and you can back your favourite here. My favourite is the, 'inside looking out' entry. Considering that the theme is 'winter solstice - personally I think we missed a trick here in Ireland not entering a scan of Newgrange or one of the other passage tombs in Brú na Bóinne. These tombs are over 5000 years old and specifically designed to channel the light on the summer and winter solstices! We'd have been a shoe in.

Newgrange - Discovery Programme 3D Icons project

Passage and chamber at Newgrange designed to channel light into the chamber at the winter solstice.

3D scans of the whole site and passage chamber available here from the always excellent Discovery Programme.

Google Earth Timelapse

There is a great example doing the rounds on social media at present of the power of satellite imagery for mapping changes in the landscape. Google Earth Timelapse combines all of the available archive Landsat digital imagery (from approximately 1980 on so I'm guessing L4 is the earliest) to present almost 40 years of continuous imagery for looking at changes in the landscape. The really powerful thing is that this is available in a web broswer for anywhere in the world.  Just zoom, pan and move the slider. No processing or RS knowledge needed to use it.

See the excellent example below - changes in the course of the river in Tibet over the period.

40 years of Landsat data for river monitoring 

I looked for Irish examples but even changes in Dublin through the Celtic tiger era are a bit hard to make out, partly because we get so many cloudy days and there is therefore less data for a smooth animation. Line banding on the Landsat 5 sensors in Dublin bay also makes it look a bit messy. The construction of the islands in Dubai is well worth a look to get you started.

Poverty and Satellite Data

I love when I come across examples of data that I would usually discard being used by other scientists in their research. Optical imagery captured at night would be of limited use to me - but I can think of a few practical applications without too much bother if I could see the urban areas in it (urban sprawl, light pollution, thermal band for heat loss, etc). The value of optical imagery captured at night, with no urban areas in it at all was less obvious to me, but some clever folks at Oxford have come up with a collaboration between economists and remote sensers and used satellite imagery and GIS to develop a metric for counting poverty globally based on the number of people living in darkness at night. The famous photo of North Korea between South Korea and China taking from the space shuttle giving one example - the capital, Pyongyang visible as the bright hub in the centre of the darkness. 

Note - I have seen this research attributed to both Stanford and Oxford - I picked the Oxford link for this post.