Sunday, 3 November 2019

Return to Sonoma

Earlier this year I was part of a bid led by Lisa Bentley, from Sonoma State University, to explore new methods to estimate fuel loads and fire impacts in forest plots in Northern California. Lisa's bid for CALFire funding was successful (yay!) and as part of it, we are helping Lisa's team deploy TLS in these plots to see how well it can be used to estimate standing fuel, deadwood, brash etc. by looking at plots that were affected in the fires of 2016. The aim is to see how new tech (lidar, drones, mobile phones etc) might augment or improve existing fire risk assessments. 

The urgency of this was brought home very starkly by the huge fires that broke out right in the area we were working, and much more widely, and are still burning as I write this at the start of November 2019. The Kincade Fire in particular has burned right through the Pepperwood Preserve, as well as causing mass evacuation, rolling power blackouts etc. It's hard to imagine how life must feel for people living in close proximity, knowing that a period of dry weather followed by Santa Ana winds, and a spark can lead to this. Lisa and her kids were evacuated late October - all safely back now thankfully. But the conditions leading to these kinds of fires are seemingly only going to get worse i.e. prolonged drought periods followed by high winds.

Phil and I flew out in late September, and met up with Lisa and her team at Pepperwood. Pepperwood itself is an amazing 3200 acre site, in the Mayacamas Mountains between Sonoma and Napa counties, and is home to the non-profit Dwight Centre for Conservation Science, which carries out all kinds of public outreach activities as well as facilitating cutting-edge science.

The field team in one of the plots affected by the 2016 fires - all these trees (pines) are dead from the previous fires and so at some point will fall, becoming fuel for future fires.
The pine plot shown above, but looking up to the crown. These are large trees (probably > 40m) and so you can get an idea of how severe the fires were in 2016.
The TLS scans were being collected in about 25 plots spread across Pepperwood, each 50 x 50 m i.e. 0.25 ha. The sites were a mix of deciduous (oak and other), evergreen (pine), and more savanna-like plots in the rolling hills. 

Panorama of an early morning walk to a lower stature oak-dominated plot, with mist lying over Santa Rosa in the distance.  The site is a working ranch so cattle grazing plays a strong role in the nature of the plots.

One of the plots reached from the walk above, showing the steep terrain in places down to creek beds as well as the noticeable impact of fire.

Working in the Dwight house, or accommodation for most of the time we were there. Great views from the balcony!
We were lucky to be able to stay for much of our trip in accommodation actually on the Preserve, in the Bechtel House, which adjoins the Dwight research station. The view from the lounge out over the hills was pretty spectacular, and the sunsets even more so!

Sunset / moonrise looking West from the porch of the Bechtel House, with Hairy Mountain in the foreground.
The weather was sunny but cool - having seen forecasts of 30+ degrees (C that is) Phil and I were maybe a little underprepared for overnight temperatures down in the low single figures and frost on the windows. Phil and I were reduced to sharing a hat.

Lisa's team very quickly picked up the challenge of using the TLS and the software, and were soon zipping through the plots, processing data and starting to think how to use them (them, Paris). A key part of this is likely to be how to provide baseline estimates of fuel loads that can be compared to used to update, the more rapid but more qualitative line transect estimates of fuel that form the basis of routine assessment currently. The challenge is to find ways to improve on those estimates, using approaches that are rapid and easy-to-deploy if possible. Lisa's team also used our ZEB-REVO to scan the ground layer - can those data be used to estimate ground fuel and even duff for example? Meanwhile, Matt Clark was collecting UAV spectral data over the same sites, in order to explore the possibilities for using eg SfM to estimate fuel loads over wider areas. 
Lisa, hard at work, controlling the Riegl with her cellphone (in left hand) - master of all she surveys.
Phil noticed this cool oak growing out of a large rock so we scanned it will Paris and Sean were climbing it. Phil then generated a nifty Sketchfab model of it!
Rock-grown oak.

We'll be back with Lisa's team next year, to carry out measurements at other sites, as well as potentially back to some of the existing plots if they've been affected by fire again in the current season. 

Tuesday, 30 July 2019

Book alert!

The work of our group features in several chapters in a new book describing the new opportunities and challenges for measuring the biomass of forests from space. The book, "Forest properties and carbon cycle studies from Earth Observations", edited by Klaus Scipal, Anny Cazenave, Teodolina Lopez (ISSN: 0169-3298 (Print) 1573-0956 (Online)) is a Special Issue of Surveys in Geophysics (vol 40, issue 4) and is the outcome of a Workshop on “Space-Based Measurement of Forest Properties for Carbon Cycle Research” held at the International Space Science Institute (ISSI) in Bern, Switzerland, from 6 to 9 November 2017.

The ISSI workshop pulled together experts from various fields and space agencies to discuss key science questions, community needs and new technologies for estimating forest biomass from space. The book summarises the state-of-the-art in space-based observations of biomass including new space-based radar and lidar (laser) instruments being launched now and over the coming few years. In addition, there is an exploration of the various challenges to making best use of these new observations, both scientifically and in terms of policy aimed at reducing emissions due to deforestation and degradation.

I led a chapter on terrestrial laser scanning as a key tool for improved calibration and validation (cal/val) of satellite observations of biomass.
Figure from Disney et al. (2019): TLS of 3 contrasting forest types, with each individual tree coloured separately. Top to bottom: Wytham Woods, UK; Caxiuanã, Brazil; Lopé, Gabon.
I also contributed to a chapter by Laura Duncanson on new cal/val protocols for biomass, and to a chapter by Jerome Chave on the importance of more and better ground data to underpin these space missions.

Figure from Chave et al. (2019) illustrating the Super Site concept for ground-based data collection for cal/val of new EO missions providing biomass estimates.
The book is aimed at a general audience of those interested in carbon cycle science, but will be of particular interest to people at the interface of science and policy. The success of initiatives such as the UN REDD+ program to reduce deforestation and degradation will rely on the observations and methods described in the book.

Wednesday, 17 April 2019

Tallest trees, new publications and Hardy Ash revisited ...

We've been involved in some exciting work measuring what is likely to be the tallest tree ever measured in the tropics, in the rainforests of Sabah, Malaysian Borneo. The tree, nicknamed “Menara” or Malay for “tower”, may also be the tallest angiosperm (flowering plant) in the world, a title currently held by "Centurion", a Eucalyptus regnansin Tasmania, Australia, measured at 99.67 m in 2016. Menara is a Shorea faguetiana (common name Yellow Meranti), of the Dipterocarpaceae family that dominates the humid lowland rainforests of SE Asia, and is 100.8 m from top to ground. It was originally identified in airborne lidar by a team from Nottingham University and partners including the South East Asia Rainforest Research Partnership) in 2014. The tree was then located on the ground by a Malaysian team and climbed and measured by Unding Jami. Phil then helped train local researchers to operate our laser scanner, in conjunction with a team from Oxford University led by Allie Shenkin. 
The view from the bottom of the tallest tropical tree in the world
'Menara', picture by Unding Jami. See
Allie also produced some amazing visualisations of the tree using the TLS and drone data. More pictures soon when the paper on this comes out.

We've been busy at the start of spring preparing for new campaigns in Peru, Gabon, the UK and in Germany. This means piecing together the jigsaw of travel plans and logistics, which Andy and Phil are leading. The German work will be in support of an ESA campaign in Kermeter, Germany later in the year, where airborne P-band radar and lidar data are being collected as part the lead up to the BIOMASS mission in 2021.

Matheus has defended his PhD, and passed with minor corrections - of course! Examined by Prof. Mark Danson from Salford, and Prof. Peter Muller from MSSL here at UCL, they were both impressed by his work. Matheus has already published 5 papers before submitting his PhD, including the most recent one in MEE summarising his leaf-wood separation work. This is a really excellent paper, providing a significant advance in what we're able to do with TLS data and opening up lots of new areas of exploring tree structure and function.

We've had a few other new papers out since the last post, perhaps most notably one of a collection by various authors on forest biomass, in an edited volume that is coming out in Surveys in Geophysics: Forest Biomass and Structure from Space. This book arises out of the ISSI-organised meeting in Bern in late 2017 and will be an excellent resource, as it will contain a really useful summary of the state-of-the-art in the field. Our first contribution is the chapter "Innovations in ground and airborne technologies as reference and for training and validation: Terrestrial Laser Scanning (TLS)".

Figure 1 from Disney et al. (2019) showing various examples of TLS point clouds.
Meanwhile I've been out and about doing various talks, including an invitation to speak as part of the lunch-time lecture series to staff at Kew Gardens (including a live feed to the Kew Millenium Seedbank at Wakehurst Place, Sussex). This was a real honour and slightly nerve-wracking as talking about trees to the Kew experts seems a little like teaching your grandmother to suck eggs as the expression goes. I also recently appeared via video link to speak as part of the Climate Conversations lecture series at Lawrence Berkeley National Laboratory, SF - another fun and interesting experience. Last but by no means least I was invited to my daughter's school in Hackney to talk about their current topics, trees and the Amazon - it's always fun to try out new material on a tough crowd - if it's not snappy they lose interest pretty quickly.

News from the GEDI mission - things are looking very good so far and the first release of data is pretty impressive. Glad to see they're continuing to rinse the Star Wars link for all it's worth :-) Can't wait to start seeing some of these data over our sites - this is going to be fascinating.

GEDI returns over US, from: 

Lastly, Phil went back to scan the Hardy Ash in Old Pancras Church yard following some remedial tree pruning work. See the details here, but also Phil's Sketchfab model below.

Monday, 25 February 2019

Vegetable monsters

We've been keeping an eye out for unusual urban trees we can potentially scan. This is partly to build up a collection of the 3D tree structures in more extreme environments, extreme in this case being the level of pressure and management they are likely to have undergone. More generally, this is part of understanding the wider issue of urban tree cover, and how and why some trees survive and thrive. Mostly this is down to management and preservation but of course that reflects what has been planted in the first place and why.

Phil was alerted to these amazing redwoods on Canons Drive, Stanmore via twitter - the go-to research tool for urban forests, obvs! This private drive has an avenue of sequoias (or Wellingtonia - see below for some history of that name) planted by the Duke of Chandos. How we ended up with sequoias in the UK is an interesting bit of history as well; clearly the Duke of Chandos wanted to build an estate that was right on the cutting edge of the latest horticultural fad.

Phil, KC and I spent a morning admiring the trees and scanning up the drive. The tallest is around 33 m and you can see from the figure below the range of shapes. This is another great illustration of the variety of 3D structure we see from particular species in the same climate, but under different other external pressures.

Extracted point clouds of the Canons Drive sequoias (P. Wilkes)

Phil has also produced a rather nice flythrough down the drive based on these data.
Canon St Giant redwoods from Phil Wilkes on Vimeo.

Thursday, 6 December 2018

Night at the museum ....

Last Friday, we were at the Natural History Museum Lates event, the monthly NHM public 'open house' events showcasing new science: people drop in, have a drink, chat to scientists, talks, interactive sessions etc. This one was showcasing NERC science, and in particular the 2018 Impact Award nominees. My NCEO colleague Shaun (Quegan) was up for the Economic Impact award for leading the ESA BIOMASS mission, and I was helping out on the stand, showing the TLS stuff we're doing for BIOMASS cal/val. BIOMASS is being built by Airbus in the UK under a 180M contract from ESA. So, definitely impact, and Shaun went on to win the award on the Monday - yay! We had a stand with a film produced by Sheffield University, a model of the sensor from the Airbus team, some tree discs, my 3D movies and the ZEB-REVO, and movies of deforestation impacts from Joao Carreiras (also from Univ of Sheffield).

My son Rudi came along for a mooch about and I got him to make himself useful and collect some ZEB data in the NHM main hall. He did a great job.

Section through the NHM main hall, whale in view, with people milling about below. Scan R. Disney.

Looking down from the balcony. The NCEO/BIOMASS stand is on the opposite side, just above the whale's head. The giant globe is the Earth installation by visual artist Luke Jerrans - which was also spectacular! Scan by R. Disney.

Walk-through of the ZEB-REVO data collection. Animation by Phil Wilkes.

The sketchfab models are pretty nifty too (as always).

We had a lot of visitors to the stand - Shaun and I literally didn't stop talking from 6-9:30. Not *that* unusual for me, but definitely for Shaun! We had lots of great discussions and questions from visitors and I really enjoyed it (which is the main thing). The event seemed really successful, with about 3500 visitors on the night, and capped with Shaun's win at the award dinner on the following Monday.

And meanwhile, congratulations to the NASA GEDI team - successfully launched to the ISS on 5/12/18 - let the lasering of forests commence! Great news.

Friday, 30 November 2018

You can't make an omelette without .....

Destructive harvesting in Brazil

A few updates on some of our activities over the past month or two. Most excitingly, this involved a destructive harvest experiment in Caxuianã, where we first scanned in 2014. Andy and Matheus led the experiment to scan, cut down and weigh four large tropical canopy trees. The aim of this work is to provide the best TLS estimates of tree volume we can, and to compare these with the harvest values. We also wanted to make measurements of wood density, as a function of tree diameter and with height. Wood density is critical in going from volume to mass of course. But it can vary a lot within and between species and even within a single tree. In the hyperdiverse tropics, the relative lack of wood density measurements may be one of the big uncertainties in forest biomass estimates based on allometry - size-to-weight relationships - that convert tree diameter (or height from satellites) to biomass. That's ALL the estimates then.

Here is a 46 m tall, 20 ton tree coming down. The team did an amazing job of clearing around the tree to give a fantastic, clear view with the TLS - some of the best data we've ever been able to get for a tall tree like this. This is a highly-skilled job as you can see.

I very much have mixed feelings about this - it's hard to watch a tree like this coming down and knowing you're responsible for it. The flipside is that if we can show how TLS can work in weighing trees like this, we can massively improve our estimates of global forest carbon stocks, particularly in the tropics where few trees of this size have ever been weighed - and none have been done like this with TLS in the Amazon.

The tree weighed over 20 tons wet, probably half that dry, of which half again (~5 t) is carbon. Early results show the wood density is a lot more variable than we'd ever expected - a problem for converting volume to biomass with a single value, but really interesting in scientific terms. Regardless of the mechanical properties, it's easy to see why these trees are so commercially valuable just on the basis of the beauty of the wood.

Prof. Lola da Costa (far right) and the chainsaw team weighing a large. and beautiful, piece of hardwood trunk.

Matheus (front left) and the team with the results of cutting up the tree above. Giant heavy beermats.
An interesting aside - when we came back, Feng Yin pulled down the Sentinel 2 satellite data covering the period the harvesting was going on - a few days before and over ensuing couple of weeks. Amazingly, the gaps due to felling 3 trees can be seen from space at 10 m resolution! The NDVI images highlight this even more. A sobering thought - 3 trees and some clearing leaves a hole that can be seen from space.
Sentinel 2 images showing the site and the tree harvest area circled: dates are 26/09, 1/10, 6/10, 26/10.

NDVI generated from the images above, highlighting the holes in the canopy.

Early analysis by Andy and Matheus already shows some amazingly clean point clouds from the TLS, which will allow for high quality reconstructions and estimates of volume from the TLS. The wood density is also showing some very interesting variations!

Huge thanks to Lola da Costa who organised the local teams - without him this wouldn't have worked at all. Also huge thanks to Patrick, Lucy, Ingrid and the local crew.

New papers

Our leaf angle distribution (LAD) paper came out in AFM, which shows a new way to get at LAD from TLS and comparing this to a photographic method. LAD is an important parameter in understanding canopy photosynthesis. But it's very hard to measure so it's often assumed to be simply distributed to make life easy (and because we don't have any other info). This work was largely led by Matheus and Jan Pisek, applying Matheus's leaf-wood separation to TLS data from Kew, and comparing to Jan's photographic method. It shows that the TLS method does surprisingly well even for small leaves and odd morphologies.
Figure 8 from Boni Vicari et al. (2019) showing the TLS point cloud, separated leaves and extracted leaf angles.
Here is a 46 m, 30 ton tree coming down. The chainsaw team did an amazing job of clearing around the tree to give us a fantastic clear view with the TLS - some of the best data we've ever had of the upper part of a tree like this. This is a highly-skilled job as you can see.
Two other papers accepted for publication are Andy's paper on treeseg, his tree extraction tool for TLS data; and Matheus's paper on leaf-wood separation, as part of his lidartf tool development. These are both potentially really useful tools for anyone working with TLS in forests. More on both those soon. 

Lastly, my Tansley Insight piece in the New Phytologist, on the potential for TLS in plant science, has come out. I've had some very nice feedback from people on it, mostly to do with figure 1, the montage of trees sketched by my daughter's primary school class! People really seem to find it relatable - which was the original idea, but you never know if these things will work or fall flat.

Monday, 17 September 2018

Summer sun, rain, wind and trees

**UPDATE 18/9/18** Ollie Baines won the RSPSoc President’s Prize for the best oral presentation of his MSc Remote Sensing dissertation project, estimating London’s urban forest cover, at the NCEO annual conference, Birmingham, 5-7 September. See below for more details.

**UPDATE 2**
See Phil's beautiful Sketchfab model of the Curtain Fig at Yungaburra, scanned by he and Allie during the recent work in Aus. That is an *amazing* tree.

It's been a busy summer for the various teams we have out there measuring trees. It started in July with Andy and Allie off to Malaysia, with nary a hiccup. Oh, other than our kit going missing for 2.5 weeks in transit. It got there in the end, but that wasn't in the plan - Andy and Allie just got started than had to come back, to be spelled by Phil. This is part of our NERC project measuring large tree architecture to understand the relationships between form and function, with Yadvinder Malhi, Lisa Bentley and Brian Enquist.
Allie and Andy, trying out the new tree QR codes for automatic tree recognition from the lidar data. 
Allie with our collaborators in Malaysia.
So in this case, the aim is the structure of a smallish number of specific trees, for which there are traits measurements, rather than 1 ha plots. This involves making sure we can ID the specific trees, so Phil developed some fancy new QR codes. We've always thought that being able to ID the trees automatically from the lidar *must* be the way to go, but Phil's worked out a good way to do it - QR codes that are big enough to be recognised automatically in image data form, even as point data.

Once again, *who* chose this plot? Some "challenging" terrain shall we say. 
Allie and the team in the field. 

Allie and Phil then moved on to Australia, northern Queensland, to some of the plots where Kim Calders and the Ghent team are also working. We're working with them to double up on TLS measurements, along with Harm Bartholomeus and the Wageningen Ricopter team. Although again, some shipping issues have led to a few delays. What is it with UAVs, batteries and paperwork??

The methods we're using here involve destructive branch sampling. So first of all some bowskills are required - Phil "Katniss" in action here, so end up a line and snag a branch for climbing and cutting.

Then the branches are placed in a lab, or the lounge of your AirBnB, to get no wind & stable lighting conditions, to capture high resolution lidar scans as well as 360 photography for SfM point cloud reconstruction. Allie prototyped this method back in the UK and it seems to work really well thus far.

Meanwhile, all these measurements have led to us thinking about 'extreme' trees - what the drivers of morphology are. So, while soil type, terrain, competition, herbivory, snow, rain etc. all have their place - wind can do funny things. Here's a 'tree' from a Pembrokeshire clifftop that kind of illustrates just how plastic this morphology is. A few 100 m away, in a more sheltered part, these hawthorn trees grow quite straight, up to 10 m or so in height.
Any way the wind blows, doesn't really matter.
Andy and Matheus are preparing to go off to Caxuianã to do some destructive tree harvests. We have axes, tarps, straps, logging tools. And this. Can hardly even lift the weighing scale it's so heavy!

Meanwhile, we have had various MSc students working with us over the summer on a variety of lidar-related topics (not all TLS for once!). Wanxin Yang did some very nice work using our PiCUS sonic tomography instrument on some trees in and around UCL to try and understand if and how it can be used to make more quantitative measurements of wood density. This approach has been used for quite a while by forestry professionals to indicate the status of trees (damage, hollows etc). But if we can use relatively quick, non-invasive measurement like this quantitatively, this would be very useful for biomass and structure studies more generally. Wanxin's work has shown some very interesting results - more on that soon.
Measuring wood density of Plane trees (L, R) and lime (Centre) in and around UCL. Picture: Wanxin Yang, UCL 2018.
Quantifying how hard the nail is being hit with the sonic hammer (yes, really). Picture: Wanxin Yang, UCL 2018.
Meanwhile, Oliver Baines was doing some excellent work mapping London's trees in terms of number, height and % cover, by combining the EA lidar data with Landsat, and some up-scaling. More on that later too. Olivers' headline figure is just over 9 M trees, which is slightly higher than current best estimates but very much in line with our recent work in Camden on this. 
Trees per unit area, at 30m resolution across the Greater London Area. Picture: Oliver Baines, UCL, 2018.
Oliver's work is something we want to take further, with other lidar and EO data and comparing to ground-based TLS estimates particularly for height and crown-size validation. The numbers on tree height and canopy cover have some quite important implications for valuing and managing (& hopefully expanding!) urban green space. More on that soon as well.

Both Wanxin and Oliver presented their work to the NCEO National Conference in Birmingham early in September, and generated lots of interest.
Wanxin Yang presenting her work on the PiCUS and wood density.

Oliver Baines talking about London's tree cover.
Phil's poster on the ZEB-REVO work in Peru.