Category Archives: Science

The Polar Plantscape Is Changing Rapidly

As the Anthropocene climate changes accelerate, we are seeing life forms (with the exception of humans) responding to the changing conditions.  Animals are moving uphill and northward, following the warming trends.

At the same time, plants are “migrating” and adapting.  Oak trees and other species are “moving” north, inhabiting areas that were previously too cold, and dying out in over heated southern ranges.

Even in polar regions, the inhabitants are adapting to the melting ice and warming air.  Penguins seem to be shifting nesting grounds, presumably following changing conditions.

This fall two studies report on how the plant (or at least non-animal) life is “moving” in the Arctic and Antarctic.

Until recently, some areas of Antarctica had large areas of moss that was covered by snow and ice in the winter.  In the summer, the cover melted these beds were exposed to the sun.  In this very wet, very sunny situation, the moss greened and thrived.  In fact, the beds seem to have persisted in the same location for years, probably centuries.

These are the “miniature old growth forests” of East Antarctica.  They are also home to many (small) animals and other species such as fungus.

A new report shows that some of these beds are rapidly dying out, apparently because the climate is drying out [4].  (Growing moss is all about water.)  Other more tolerant mosses are invading the area.

Tiny as they are, this rapid change to the “moss forest” is a huge ecological shift.

At the same time, another study reports that in northern regions tundra plants such as grasses have grown taller, and larger species are moving north [2].  In fact, getting on twice as tall.  There are no trees in these harsh locations, so these ankle high species are the “miniature old growth forests” of the tundra.  And they are bulking up in response to warmer, wetter conditions.

compare it to the ecosystems around your house like the forest nearby – if you imagined that forest getting twice as tall; that is a pretty dramatic change,” (Isla Myers-Smith, quoted in [1])

The presence of these taller plants might create a positive feedback, insulating the soil (e.g., by trapping blowing snow) and lowering the albedo (due to foliage sticking above the snow).  These effects could contribute to more melting of ice and permafrost, further warming the area.

Not only are the old growth forests shrinking and dying, these “miniature” old growths, mosses and grasses and such, are also changing rapidly.

It’s kind of a cool image, “miniature old growth forests”.

(For the record, the data is available from “Team Tundra” and the Austrailian Antarctic Data Center (AADC) )

  1. Jonathan Amos, Taller plants moving into warmer Arctic, in BBC News – Science & Environment. 2018.
  2. Anne D. Bjorkman, Isla H. Myers-Smith, Sarah C. Elmendorf, Signe Normand, Nadja Rüger, Pieter S. A. Beck, Anne Blach-Overgaard, Daan Blok, J. Hans C. Cornelissen, Bruce C. Forbes, Damien Georges, Scott J. Goetz, Kevin C. Guay, Gregory H. R. Henry, Janneke HilleRisLambers, Robert D. Hollister, Dirk N. Karger, Jens Kattge, Peter Manning, Janet S. Prevéy, Christian Rixen, Gabriela Schaepman-Strub, Haydn J. D. Thomas, Mark Vellend, Martin Wilmking, Sonja Wipf, Michele Carbognani, Luise Hermanutz, Esther Lévesque, Ulf Molau, Alessandro Petraglia, Nadejda A. Soudzilovskaia, Marko J. Spasojevic, Marcello Tomaselli, Tage Vowles, Juha M. Alatalo, Heather D. Alexander, Alba Anadon-Rosell, Sandra Angers-Blondin, Mariska te Beest, Logan Berner, Robert G. Björk, Agata Buchwal, Allan Buras, Katherine Christie, Elisabeth J. Cooper, Stefan Dullinger, Bo Elberling, Anu Eskelinen, Esther R. Frei, Oriol Grau, Paul Grogan, Martin Hallinger, Karen A. Harper, Monique M. P. D. Heijmans, James Hudson, Karl Hülber, Maitane Iturrate-Garcia, Colleen M. Iversen, Francesca Jaroszynska, Jill F. Johnstone, Rasmus Halfdan Jørgensen, Elina Kaarlejärvi, Rebecca Klady, Sara Kuleza, Aino Kulonen, Laurent J. Lamarque, Trevor Lantz, Chelsea J. Little, James D. M. Speed, Anders Michelsen, Ann Milbau, Jacob Nabe-Nielsen, Sigrid Schøler Nielsen, Josep M. Ninot, Steven F. Oberbauer, Johan Olofsson, Vladimir G. Onipchenko, Sabine B. Rumpf, Philipp Semenchuk, Rohan Shetti, Laura Siegwart Collier, Lorna E. Street, Katharine N. Suding, Ken D. Tape, Andrew Trant, Urs A. Treier, Jean-Pierre Tremblay, Maxime Tremblay, Susanna Venn, Stef Weijers, Tara Zamin, Noémie Boulanger-Lapointe, William A. Gould, David S. Hik, Annika Hofgaard, Ingibjörg S. Jónsdóttir, Janet Jorgenson, Julia Klein, Borgthor Magnusson, Craig Tweedie, Philip A. Wookey, Michael Bahn, Benjamin Blonder, Peter M. van Bodegom, Benjamin Bond-Lamberty, Giandiego Campetella, Bruno E. L. Cerabolini, F. Stuart Chapin, William K. Cornwell, Joseph Craine, Matteo Dainese, Franciska T. de Vries, Sandra Díaz, Brian J. Enquist, Walton Green, Ruben Milla, Ülo Niinemets, Yusuke Onoda, Jenny C. Ordoñez, Wim A. Ozinga, Josep Penuelas, Hendrik Poorter, Peter Poschlod, Peter B. Reich, Brody Sandel, Brandon Schamp, Serge Sheremetev and Evan Weiher, Plant functional trait change across a warming tundra biome. Nature, 562 (7725):57-62, 2018/10/01 2018.
  3. Victoria Gill, Climate change kills Antarctica’s ancient moss beds, in BBC News – Science & Environment. 2018.
  4. Sharon A. Robinson, Diana H. King, Jessica Bramley-Alves, Melinda J. Waterman, Michael B. Ashcroft, Jane Wasley, Johanna D. Turnbull, Rebecca E. Miller, Ellen Ryan-Colton, Taylor Benny, Kathryn Mullany, Laurence J. Clarke, Linda A. Barry, and Quan Hua, Rapid change in East Antarctic terrestrial vegetation in response to regional drying. Nature Climate Change, 8 (10):879-884, 2018/10/01 2018.


NASA Tensegrity Robot v2

NASA continues to work on their tensegrity robot.

This is so cool. Still.  Always will be.

The latest video illustrates two of the key assets of this design for planetary exploration: it is bouncy and it can be flat-packed.  Just the thing for dropping onto Mars or another planet or ice moon.

The video also illustrates one of the hard problems they are working on:  locomotion.

This robot gets around by scrunching its various legs in ways to make it “roll” in the desired direction.  Navigation amounts to a sequence of such moves, creating a slow, and somewhat drunken walk.  (In the video, apparently visiting some kind of alien Stonehenge.)

Programming these “gaits” is certainly different than other robotic systems, and is very complicated.  There are a lot of possibilities and permutations, and I don’t think there are many natural models or intuitions to work from.

Earlier efforts worked out some locomotion by “hand”, but the robot needs to be able to adapt to terrain and events, so it’s not enough to just be able to talk in one way.

Recent work is developing adaptive guidance using reinforcement learning.  The techniques also allow non-periodic “moves”, i.e., much more flexible sequences of steps.  It’s kind of cool, though I don’t really understand it very well.

Deformation-based locomotion of tensegrities generally relies upon the geometric relationship between the supporting base polygon and the center-of-mass” ([1], p. 1)

Specifically, the design is “algorithmic discovery of adaptive feedback behaviors”. The researchers describe this as “the middle of a hierarchy of considerations for useful deployment of mobile tensegrities”, above wired in hardware but lower level than a planning algorithm.  (OK, that’s a nerdy point, but it makes sense to me, ‘cause I been there, trying to figure out software architectures.)

It’s still early days on this approach, but I think it has a lot of promise.  Pretty soon now they’ll have to figure out how to paint the ‘NASA’ logo and US flag on it, which is going to be a challenge.

  1. David Surovik, Kun Wang, and Kostas E. Bekris, Adaptive Tensegrity Locomotion on Rough Terrain via Reinforcement Learning. arxiv, 2018.



Space and/or Robot Saturday

Giraffe Spots

Animal coloration has long fascinated us, and provided a profound question for materialistic biology.  Just how do complex strips and spots come to be?  They look designed, but appear to just happen.  And they are neither random nor precisely the same in all cases.

Charles Darwin and others have argued that many stipes, spots, and other coloration have been selected though value as camouflage, or perhaps species or intraspecies signaling, or other survival benefits (thermo regulation?).

In the twentieth century Alan Turing proposed that patterns such as stripes and spots developed via chemical processes [1], and since that time genetic and metagenetic mechanisms have been invoked.

Still, just how do animals get their cool colors?

Of course, the coolest animal of all is the Giraffe (Giraffa Camelopardalis), with their characteristic beautiful spots. Every giraffe has recognizably “giraffe-like” spots, but each is a little different.  Different populations and even individuals can be recognized by their distinct fingerprint of spots. (As far as I know, the spots of an individual animal do not change over their life, though aging giraffes may grow darker.)

A Giraffe in the Mikumi National Park, Tanzania (Credit: Muhammad Mahdi Karim) License: GPL 1.2

So how much of the spot pattern is inherited, and what else influences their growth?

A new study examined the spots of giraffes, mothers and calves.  (You had me at “giraffe”.) The research collected digital images of wild giraffes, Each sampled individual could be identified from the imagery.

(How cool is it to have a job studying giraffes???)

The images were processed to characterize the spots by 11 visual features. The research used public domain software to normalize the images and then analyzed the size and shape of the spots for each individual.  This data was used to algorithmically classify each pattern (individual).

The sample included 31 mother-calf pairs, which could be analyzed for possible inheritance of spots.  Specifically, the 11 traits were compared between mothers and their calves.

The study also tracked the apparent survival of all calves encountered related to their spots.  I.e., which traits might be adaptive, helping babies survive (presumably as camouflage).

With so many dimensions the statistics get complicated.

The basic finding is that some features are highly correlated between mother and offspring, and therefore likely passed genetically from the mother.  These features are also correlated with the survival of the babies.  This offers evidence that the coloration may be important for the survival of young giraffes.

It is important to note that these animals, mother and offspring, inhabit the same area with its characteristic vegetation and shading.  Thus, if these patterns are particularly favorable for camouflage in this locale, it would affect both mother and calf.  (I.e., the mothers must have survived childhood in the same environment.)

At the very least, other traits might be just as heritable, but would be more adaptive in different conditions.

There is still much to work out about what genetic and developmental processes determine an individual giraffe’s coat pattern.  The study indicates that giraffe spots are rather complicated, with at least 11 dimensions of variability, and they did not include color in the study.  This means there may well be multiple mechanisms combining to determine the specific spots of an individual giraffe.

“Coat pattern variation may reflect discrete polymorphisms potentially related to life-history strategies, a continuous signal related to maternal effects, or a combination of both.”  ([2], pp. 16-17)

The authors note that their techniques are a useful advance in the study of giraffes.  They utilize low cost, high quality digital imaging (which does not disturb the animals), along with public domain image analysis and classification software.  These techniques may “provide a new, robust dataset to address taxonomic and evolutionary hypotheses.” (p. 17)

I wonder if these techniques can be automated further, as a mostly automated pipeline of processing.  This would make it easier to collect a larger dataset, maintaining consistency.  It would also make it possible to reproduce the findings, and to create improved classifications.

  1. The chemical basis of morphogenesis. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 237 (641):37, 1952.
  2. Derek E. Lee, Douglas R. Cavener, and Monica L. Bond, Seeing spots: quantifying mother-offspring similarity and assessing fitness consequences of coat pattern traits in a wild population of giraffes (Giraffa camelopardalis). PeerJ, 6:e5690, 2018/10/02 2018.

Smart Phone Microscopy: Detecting Lead in Water

With the ever more ubiquitous presence of smartphones equipped with processors, network, and astonishingly good digital cameras, there is a growing trend to use these devices as medical sensors.  Indeed, with appropriate lenses, a smartphone can be used for many optical measurements, including microscopy and even spectroscopy (maybe).

The big idea, of course, is to put sophisticated sensing in the hands of everyone. The pro-social version of this vision is to make DIY kits, available to anyone.  (The anti-social version aims to use the technology for surveillance, marketing, and iffy self improvement.)

A nice example of how this should work comes from University of Houston reserchers, who have released an Open-source do-it-yourself multi-color fluorescence smartphone microscopy [3].

This particular effort implements fluorescent microscopic techniques, which are potentially useful for many kinds of analysis. The kit includes 3D printed clip on housing, battery powered LEDs, and a 3D printed stick on lens.  Plans for the whole rig are available, and can be fabricated for roughly $20.

Of course, there is some significant software, too, to process the images.  The initial version collects the data images and sends them to a desktop computer for viewing and further analysis.

One potentially interesting use of this technology would be detection of lead in water supplies.  This fall the researchers report techniques that can detect lead in water down to the EPA threshold of 15 ppb.  This cheap and accurate test is within the reach of ordinary people, and could be a great help for detecting lead contamination and assuring safe drinking water.

This is a great prospect, and it will be important to put together the whole package of instructions, access to materials, and software.

But first, this needs to be more broadly tested, to make sure the strengths and weaknesses are well understood, and to work out problems that may occur in filed use by untrained people.

So there is work to do, but it looks like it’s very doable.

This and similar techniques can be used for a variety of surveillance, including assessing microbes and other potential contaminants.  But each use needs to be carefully developed and publicly validated. (Start ups tend to keep their methods and performance secret–which is not good enough for public health.)

It would be really great to develop libraries of open source technology for as many environmental assessments as possible, especially for water and food.  (This would be something for the twenty first century public library to provide, no?)

I have expressed concerns about misguided DIY sensing. I think this project is an example of the right way to do it: carefully validated, published, and specifically targeted  to an application with a real need.  And, of course, open sourcing it, rather than rushing to commercialization.

Nice work, and I hope this becomes a real open source product.

  1. Jeannie Kever, Researchers Create Smartphone System to Test for Lead in Water, in University of Houston — News. 2018.
  2. Hoang Nguyen, Yulung Sung, Kelly O’Shaughnessy, Xiaonan Shan, and Wei-Chuan Shih, Smartphone Nanocolorimetry for On-Demand Lead Detection and Quantitation in Drinking Water. Analytical Chemistry, 90 (19):11517-11522, 2018/10/02 2018.
  3. Yulung Sung, Fernando Campa, and Wei-Chuan Shih, Open-source do-it-yourself multi-color fluorescence smartphone microscopy. Biomedical Optics Express, 8 (11):5075-5086, 2017/11/01 2017.


Arctic Ice Minimum For 2018

The ice is melting everywhere.

This fall the Arctic icecap reached its summer minimum, which appears to be the sixth lowest on record.

There isn’t much more to say, except maybe “glub!”.  (“I’ll keep it to myself. Until the water reaches my lower lip, and then I’m gonna mention it to SOMEBODY!”)

  1. Maria–José Viñas and Mike Carlowicz, Arctic Sea Ice Reaches 2018 Minimum, in NASA Earth Observatory. 2018.


New LIDAR Survey of Lowland Maya Ruins

One of the most romantic of all sciences is the archaeology of “lost” civilizations.  And none are more romantic than the recognition and documentation of the Mayan empire, which flourished more than 1,200 years ago.

The Mayan empire is known from relatively few sites with large, well preserved settlements.  Much of the civilization lived in areas that are thickly forested and difficult to access even today.  With so much of the remains hidden and sparsely populated today, the population, wealth, and technology of the Maya has been consistently underestimated.

In the twenty first century, remote sensing is bringing a flood of new data about ancient settlements and landscapes, including the regions of the Mayan empire. This fall a research team of the Foundation Patrimonio Cultural y Natural Maya report a major new study of the Maya low lands of Guatemala, combining LIDAR and field work [1].

Earlier studies have shown that airborne LIDAR is a very effective tool for detecting settlements and structures even under heavy forest cover. The new study extends earlier work to survey a larger sample areas of this region (more than 2,000 square KM).  The research group includes a multidisciplinary team which is making a variety of studies from this data.

The LIDAR was measured from an aircraft flying at 750m above ground level, with resolution approximately 1-3 metesr. The resulting terrain models were visually inspected and cross referenced with previous ground surveys.  The data can also be analyzed by algorithms.  This analysis was used to guide a round of excavations to evaluate the putative “interesting” features.  This led to a second iteration of visual analysis, to identify more such “interesting” features.

(One important advantage of remote sensing on Earth versus other planets is that it is actually possible to visit the site and obtain ground truth to verify and better interpret the remote sensed data.)

The results have identified thousands of structures, including canals and causeways.  The canals and causeways represent intensive infrastructure for transportation and agricultural production.

The researchers interpret the findings to indicate evidence of extensive agriculture, and varying levels of urbanization.

Based on the identified structures, heuristics suggest a population density of 80-120 people per square km.  If the sampled areas can be extrapolated to the whole lowland region, this suggests a total population of 7-11 million.  This is at the upper end of earlier estimates, but arguably may underestimate the population.

(This could have been a higher population density than, say England, of the same time period.)

This large of a population with large urban centers would require intensive agriculture to sustain it.  The study identified many canals of various sizes, which are interpreted as either irrigation or drainage channels for low land agriculture. In uplands, the survey also identified linear stone features, which are interpreted as fences and terraces.

“The PLI survey revealed a landscape heavily modified for intensive agriculture” ([1], p. 5)

Nothing is known of the horticultural practices of the times, but plausible estimates based on contemporary practices indicate that the putative food producing land could have supported the estimated population.  This opens the possibility that there was surplus food and/or production of non-food crops.

The samples indicate some areas with high density, urban habitation.  Interestingly, these locations were often near highly intensive agricultural developments.  Perhaps the development of denser populations pushed the intensification of food production, compared to other areas which could remain extensive.  Even so the urban areas would still not be self-sufficient without food from the countryside.

“The co-occurrence of dense settlement with agricultural improvements suggests that investments to maximize agricultural production were directed to areas where population sprawl limited the possibility for extensive farming.”  ([1], p. 7)

The study also identified extensive infrastructure for water supply, which would have required significant labor investment and political coordination (e.g., digging reservoirs). The estimated water supplies are consistent with an adequate or abundant supply for the estimated the estimated population.

“such features are monumental in scale and imply some form of centralized involvement in planning and execution. “ ([1], p.12)

The LIDAR also indicates raised linear stone “causeways”, which are interpreted as roads and ceremonial entryways. There are also a variety of works interpreted as defensive fortifications.  These are prominent in some areas, generally integrated with natural defenses.  The researchers suggest that these works indicate that some areas had  a greater perceived need for military defense, which is consistent with what is known of the military conflicts of the times.

This is an amazing study, with both astounding breadth and detail.  It is interesting to see how the different features can be interpreted to give multiple, mutually consistent, estimates of the population and settlement patterns.

“In the central Maya Lowlands, lidar survey will become in- dispensable for settlement research because of (i) the speed of the data-gathering process, (ii) the degree of detail attainable over large areas, and (iii) the ability to discern large and small features routinely undetected by traditional methods. “ ([1], p. 14)

The researchers point out that the LIDAR data may have other uses, including monitoring looting and estimating biodiversity.  The data will be published next year after the researchers have a chance to publish.

Unravelling how the Maya flourished in this area–and possibly failed to flourish–may yield insights that help contemporary residents achieve sustainable development.

  1. Marcello A. Canuto, Francisco Estrada-Belli, Thomas G. Garrison, Stephen D. Houston, Mary Jane Acuña, Milan Kováč, Damien Marken, Philippe Nondédéo, Luke Auld-Thomas, Cyril Castanet, David Chatelain, Carlos R. Chiriboga, Tomáš Drápela, Tibor Lieskovský, Alexandre Tokovinine, Antolín Velasquez, Juan C. Fernández-Díaz, and Ramesh Shrestha, Ancient lowland Maya complexity as revealed by airborne laser scanning of northern Guatemala. Science, 361 (6409) 2018.
  2. Anabel Ford and Sherman Horn, Above and below the Maya forest. Science, 361 (6409):1313, 2018.


Space Saturday

New Hummingbird In Ecuador

Let’s stop and stare for a minute at a newly recognized species of hummingbird.

Hummers are some of the coolest birds ever, and certainly some of the prettiest.  So there can be no such thing as too many species of hummingbird!

This fall ornithologists report a new species, tagged Oreotrochilus cyanolaemus, or blue-throated hillstar, living in the mountains in Western Ecuador [2].

(BBC) The male of the species has a dark blue neck and a white breast with a black stripe
(BBC) The female of the species lacks the blue throat colouring

This bird lives is a fairly small geographic area, and appears to be highly endangered by mining operations today and possible climate change in the future.

At least we got to see it before it is gone.

  1. BBC News, New hummingbird species spotted in Ecuador, in BBC News – Latin America. 2018.
  2. Francisco Sornoza-Molina, Juan F. Freile, Jonas Nilsson, Niels Krabbe, and Elisa Bonaccorso, A striking, critically endangered, new species of hillstar (Trochilidae: Oreotrochilus) from the southwestern Andes of Ecuador. The Auk:1146-1171, 2018/10/01 2018.

PS:   The Blue-throated Hillstars would be a great name for a band.