Anthony's Science Blog

Geoengineering, ctd.

From an article in The Atlantic on using technological solutions to reduce hurricane intensity:

…researchers would dump their particles out of large cargo planes, some of which can carry 125 tons or more. In Hoffman and Alamaro’s scenario, the planes would disgorge the soot above the hurricane’s eye and the storm would disperse it outward. In the other group’s plan, planes would disperse the salt particles at the storm’s outer edges, to be hoovered up by the storm’s churn and delivered to its heart. In both cases, the immediate impact on the hurricane’s intensity would probably be negligible, but the effect would compound over time as the storm drifted west. All told, it might take a dozen or so flights a day to set in motion the degradation of a big storm.

More evidence that these types of schemes could be on their way sometime in our lifetimes.

Salix pulchra time lapse

Time lapse video by Ed Metzger showing leaf out in a common arctic tundra species, Salix pulchra.

Nice splom() wrapper for more informative scatterplot matrices

Thanks to this post, I just discovered a great wrapper for splom() called chart.Correlation (package: PerformanceAnalytics), which juices up scatterplot matrices with some nice loess curves, r² values, and density plots for each variable: pretty much the exact additions I want for every scatterplot matrix I make.

The only thing I don’t like are the significance stars, but it was easy to delete those lines of code and run without them. Here’s an example without the stars for some data I’ve been looking at. The numbers in the upper right boxes are the r² values and the text in the histogram/densityplot boxes are the variable names.

Fall in the arctic tundra, ctd.

Got a much better picture yesterday afternoon, this time a panorama of Toolik Lake. The picture is a little big for the blog, so follow the link.

Fall in the arctic tundra

We’re getting some beautiful fall color as the plants wind down for the season.

More cool Brooks Range photos

Fellow Toolik researcher Pat Tobin has a bunch of great Brooks Range hiking photos in his Picasaweb. This is me on a hike known as “not Dalton” that is near Dalton Peak, which rises above Atigun Pass.

Fun Scrabble study

As a fan of both scrabble and statistics, I enjoyed this fun article about a study of Scrabble scoring. Among the findings:

• The blank is worth about 30 points to a good player, mainly by making 50-point “bingo” plays possible.
• Each S is worth about 10 points to the player who draws it.
• The Q is a burden to whichever player receives it, effectively serving as a 5 point penalty for having to deal with it due to its effect in reducing bingo opportunities, needing either a U or a blank for a chance at a bingo and a 50-point bonus.
• The J is essentially neutral pointwise.
• The X and the Z are each worth about 3-5 extra points to the player who receives them. Their difficulty in playing in bingoes is mitigated by their usefulness in other short words.

Future directions in environmental science, ctd

Geoengineering is the emerging field that is evaluating possible global-scale technological solutions to climate change. There could be a lot of potential for research into the effects and side effects of geoengineering. From an NSF press release:

Bitz, working with University of Washington researchers Kelly McCusker and David Battisti, analyzed the impact of the leading geoengineering solution, the release of volcanic aerosols into the upper atmosphere.

“The equivalent of Mount Pinatubo going off every year,” Bitz said, referring to the eruption in the Philippines in 1991, the largest in recent memory.

Interpreting standard errors as confidence intervals

A lot of people don’t realize that standard errors can be interpreted as confidence intervals that vary in the confidence level depending on sample size. And what better way to get the word out than on my widely read blog? Knowing that these two concepts are closely related can help to interpret graphs on which people use standard errors. Here’s a handy table:

  n    CI
  1 50.00
  2 57.73
  3 60.89
  4 62.60
  5 63.67
  6 64.40
  7 64.93
  8 65.34
  9 65.65
 10 65.91
 15 66.68
 20 67.07
 50 67.78
100 68.02
Inf 68.26

So, if your sample size is 3, your standard error will represent a 61% confidence interval. If your sample size is 100, it will represent a 68% confidence interval. So, if sample size is not equal between groups on a graph, standard errors will represent confidence intervals of varying confidence. This is kind of a weird approach to graphing things when you think about it, but a few percent difference isn’t big enough to be important in most cases.

Plant-soil-microbial systems

Nice writeup by Richard Bardgett on plant-soil-microbial systems:

In fact, some work suggests that as plant growth increases because of elevated CO₂, more carbon not only flows into the plants themselves, but also exits their roots to impact the growth and activity of soil microbes. This causes a net increase in CO₂ and other greenhouse gases escaping from the soil and entering the atmosphere, thus adding to anthropogenic levels.

This scenario illustrates the complexity of controls on ecoystem carbon exchange and shows how something as simple as CO2 fertilization will not be able to explain the whole story. The rest of the article is great too and worth a read.

Future directions in environmental science

Interesting NYT magazine article about investor Jeremy Grantham. Climate change is a huge current research topic in the environmental sciences, but it won’t remain so forever. Articles like this suggest that human resource limitation could be an up and coming research area.

People are naturally much more responsive to finite resources than they are to climate change,” he said. “Global warming is bad news. Finite resources is investment advice.

. . .we’ll muddle through to a solution to Peak Oil and related challenges. Peak Everything Else will prove more intractable for humanity. Metals, for instance, “are entropy at work . . .Local water shortages will cause “persistent irritation” — wars, famines. Of the three essential macro nutrient fertilizers, nitrogen is relatively plentiful and recoverable, but we’re running out of potassium and phosphorus. . . And he rates soil erosion as the biggest threat of all.

Toolik blogging, ctd

Another Toolik blog, this time by field tech Verity Salmon. I’d say these are some of the best Toolik photos on the web!

Those are some mosses on the mountain known as Flat top. I did that hike a few weeks ago and saw the same mosses but my picture isn’t as nice as this one. Verity just left Toolik and will be starting a Ph.D. in the fall at University of Florida. She will be missed!

NEON to break ground

Looks like NEON is about to be up and running:

“If these guys are successful, they will have the goods on some really big ecological questions,” Wofsy says. “Scientists who are interested in addressing those problems will have to make the effort to learn how to use those data.”

Gates of Kiev

One of the cooler hikes near Toolik is Gates of Kiev, the tallest peak visible from camp (7,775 ft). A couple folks from our research crew and I hiked this yesterday and had a great time. Above is a picture of the glacier beneath the summit. The mountain has two sharp peaks within a few hundred feet of each other and is named after an architectural structure in Kiev that had a similar shape. I wanted to get a nice picture of the two peaks, but alas, the summit was foggy all day.

We hiked from Galbraith campground to a crag just below the taller of the two summits for a round trip of about 15 hours. As we knew before starting, the summit is inaccessible from our approach (north ridge) due to an almost vertical rockface (below). A good climber (i.e. not me) could probably do this without too much trouble, but perhaps not in the snowy conditions we had yesterday.

More pictures here.

More on climate change in the arctic tundra

In addition to the phenomenon of thermokarsts, the other really obvious and interesting thing going on up here is the increase in tundra fires. There was a massive tundra fire in 2007 along the Anaktuvuk River that burned 1000 km². Here’s a satellite image (MODIS) of it:

Some leading researchers here at Toolik recently published a Nature paper (and got some nice press coverage) on the amount of carbon lost from the fire (a lot). My fellow Toolik postdoc Adrian Rocha has also been doing some cool work on carbon balance post fire. I flew out to the burn to help him one time with his towers, but I didn’t see the actual burnedness since it was still snowy at the time.

Toolik blogging

Journalist Jennifer Smith just spent a couple weeks up here at Toolik and did some excellent reporting on the various projects that folks have been working on here at Toolik. This is a great snapshot of the variety of projects that people are working on up here this summer:

http://3290milesfromli.tumblr.com/

A cool picture from Jennifer’s blog of one of Toolik’s leading researchers, Gus Shaver, in his native tussock habitat.

Annual review review

I just went through the tables of contents for the last few years of Annual Review of Ecology and Systematics and poked around a few of the papers. Here are some cool things I read:

From a review by Yiqi Luo:

The coupled carbon-climate models reported in the literature all demonstrate a positive feedback between terrestrial carbon cycles and climate warming. A primary mechanism underlying the modeled positive feedback is the kinetic sensitivity of photosynthesis and respiration to temperature. Field experiments, however, suggest much richer mechanisms driving ecosystem responses to climate warming, including extended growing seasons, enhanced nutrient availability, shifted species composition, and altered ecosystem-water dynamics.

I thought that was a nice summary of the reasoning behind a lot of climate change-focused research in ecology. Can we hope to uncover all of these mechanisms in every ecosystem before they occur on their own? Probably not, but there are some cool discoveries yet to be made that will probably alter forecasts of atmospheric carbon.

From a paper called The social lives of microbes:

It used to be assumed that bacteria and other microorganisms lived relatively independent unicellular lives, without the cooperative behaviors that have provoked so much interest in mammals, birds, and insects. However, a rapidly expanding body of research has completely overturned this idea, showing that microbes indulge in a variety of social behaviors involving complex systems of cooperation, communication, and synchronization.

Not really my area of research, but sounds awesome.

Biogeochemical mind-bender from Hedin et al. about why there is high N availability in tropical forests:

The putative source for tropical N richness—symbiotic N fixation—should, in theory, be physiologically down-regulated as internal pools of bioavailable N build.

Why does this not occur?! Their answer was basically that heterogeneity, particularly vertical heterogeneity throughout the canopy, leaves some areas N-limited. N fixers grow in those N-limited spots, bringing a lot of N into the system. Then when they die, those N-rich tissues rain down to fertilize the rest of the ecosystem.

Landscape-level nitrogen import and export in an ecosystem with complex terrain, Colorado Front Range

Another paper from my dissertation just came out in Biogeochemistry. This was a fun collaborative project that I worked on with a couple members of my committee and a group in France that did some great isotopic work on our stream samples from Niwot Ridge.

I used some of the spatial data from my study site to estimate the relative contributions of two major inputs: winter atmospheric deposition of N on snow and N fixation. You can see that there is a somewhat complementary relationship in which there is a greater abundance of N-fixing plants in areas with less snow and substantial snow inputs in areas with low N fixer abundance.

In addition, the isotopic data showed that:

During the initial phases of snowmelt, mixing model end members for oxygen isotopes in nitrate (NO3-) indicated that a substantial quantity of NO3- is transported downhill into the forested subalpine without being assimilated by soil microbes.

July 31: snow day

The best side benefit of arctic field work is meeting fun people; next best is the hiking. We got snowed on yesterday on our hike near Atigun pass.

These Brooks Range hikes are really incredible due to the vast treeless terrain. There are no trails and it is rare to see another hiker. It’s typically slow and grueling to cross the landscape, but always worth it.

Online science

Really interesting article here about the future of scientific discourse in the digital age. Quote:

What we did not anticipate was another kind of resistance to the web, based not on an unfamiliarity with the digital realm or on Luddism but on the remarkable inertia of traditional academic methods and genres—the more subtle and widespread biases that hinder the academy’s adoption of new media.

As I’ve been publishing my dissertation, I’ve been wondering when (probably not a question of if anymore) our publishing will be done in an exclusively online format and how this will affect the peer review process. For now, I absolutely need traditional publications on my CV, but part of me feels like I should prepare sooner rather than later for the use of online media as the primary forum for scientific discourse.

Cool thermokarst video

Fellow Toolik researcher Sarah Godsey made this awesome time lapse video of a land formation created by melting permafrost, also known as a thermokarst. A lot of people have seen pictures of structures that have toppled due to melting permafrost, but less people have seen the effects on the arctic tundra. Thermokarsts are cropping up all over this biome due to climate change.

Nitrogen deposition decreases acid buffering capacity of alpine soils in the southern Rocky Mountains

New paper out in Geoderma this week based on some work I did with my advisor Bill and an undergrad in our lab, Anna Lieb. Anna put a ton of work into this project and did a great job, so I’m really happy to see this in print. Anna has gone on to a Ph.D. program in math at U.C. Berkeley.

We took soils from Bill’s long-term N addition plots and added acid to them in the lab. As you can see above, the plots that had been most heavily fertilized (lowest line) were also most sensitive to acid addition, with greater pH drops for the same amount of acid added, suggesting that the ongoing N deposition in this ecosystem makes these soils more susceptible to acidification.

Hot Spots of Inorganic Nitrogen Availability in an Alpine-Subalpine Ecosystem, Colorado Front Range

I was excited that one of the main papers from my dissertation came out recently. This paper went through a crazy number of iterations in which it was one giant paper in my dissertation (one part field data, one part mathematical definition of hot spots), then was split into two papers for submission to journals, then recombined into one paper at the request of the reviewers. I’m happy with how it came out though.

Eye of the science storm

Today is the rare day when I feel like I have a chance to catch my breath, the eye of the science storm. It’s been busy times finishing my dissertation, publishing its chapters, and working on my postdoc project over the last year and half. Now with a series of papers from my dissertation more or less done and a mid-summer break from field work in Alaska, it’s time to take stock of the journey thus far and boldly forge some epic paths to follow in the future. I kid of course…my journey is less Charles Darwin and more Stan’s dad from South Park.

I left Toolik Field Station two weeks ago after kicking off our second summer of field research during a six week stay. My first task was to implement our snowmelt acceleration treatment at our study site. I was excited that the treatment worked really well this year, accelerating snowmelt by two full weeks. I then worked on getting the rest of our measurements and experiments up and running, including putting out some ion exchange resin membranes to measure soil nutrients in our plots. I’m hoping these resins will provide a nice complement to our microlysimeter data.

After doing a lot of solo work during my dissertation (as is inevitable for such an undertaking), I have really enjoyed working on a team in my current postdoc project. Everyone kind of has their corner of the big project that they do a really good job on, and then hopefully it all comes together in the end. I am excited for that phase as I think everyone on our team will have something cool to bring to the table.

I’ve also enjoyed working at Toolik Field Station. It is a surprisingly engaging and fun environment in a “summer camp for scientists” sort of way. When journalists visit Toolik, some of them inevitably write the obvious story along the lines of “these people are having too much fun up here…food is delicious, there’s a sauna, your tax dollars at work blah blah.” Then they conveniently leave out the part about how everyone at Toolik (scientists and the awesome staff) busts their ass for minimum 60 hours a week (not an exaggeration) to get their projects done during their stay. Also, station residents more often than not forge meaningful–often lifelong–very productive scientific collaborations that have been a driving force in our understanding of arctic ecosystems for decades now. I’ve personally made some great friends. Because of the total scientific immersion at Toolik, I’d say it’s unquestionable that my skills as a scientist have been greatly sharpened in just the short time I’ve spent up there.

Next up I spend a month here in Toledo during which time I am hoping to whip out a little methods paper concerning the use of OPAME to measure amino acids using a microplate reader (a topic I wrote about on this blog last year; there are updates to that story). Then it’s back to Toolik to finish out the field season. When I return in September, more lab samples to process, data to analyze, papers and grants to write, and of course the foreboding prospect of applying for jobs! Jackie and I are both quite curious about where we will end up next.

Fun short video on the project I’m working on in the arctic

I really liked this video made by the crew at WGTE’s Plugged In series. Thanks to Amanda and Dave and whoever else helped for putting it together!

Arctic Seasonality Conference Day 2

During our second day at the conference, we broke out into groups and talked about linking projects together and defining next steps for research into changing arctic seasonality. One of the ideas that I was excited about was creating a formal research collaboration structure that allowed modelers and observational scientists to work together to create better models.

Apparently, this type of structure is already in place for many climate modeling teams. As I understand it, the climate modelers describe what empirical data they need and the observational scientists then attempt to collect that data or to convince the climate modelers to modify their models to include more realistically measurable data. This has apparently worked well for climate modelers and our working group agreed that this could be a successful approach for looking at arctic biogeochemical cycling as well.

A couple other scientific loose ends that I found interesting: (1) It is apparently unknown when the last time in geologic history that the Arctic Ocean was ice free. There are theories but it is up for debate. Would have thought we knew that, but I guess not. (2) Changes in snowfall and snowpack were a theme common to many projects. One of the breakout groups suggested that a comprehensive review of snow manipulation experiments would be useful if it does not already exist.

Finally, it was great to talk to a collaborator on our arctic plant-soil interactions project, Seeta Sistla, who is doing some cool modeling of soil decomposition processes that includes microbial enzyme activity. I’m looking forward to working on matching up that model with measurements from our experiment in Alaska.

Conference at UVM: Changing seasonality in the Arctic

Today I’ve been enjoying attending a small conference at the University of Vermont on the effects of changing seasonality in Arctic ecosystems. The basic issue is that the course of the seasons is changing in the arctic: in addition to warmer temperatures, the length of the warm snow-free summer season is being extended at the beginning and end of the season. In other words, spring is coming earlier and fall is coming later.

This change in seasonality will have far reaching effects on many aspects of the arctic system and we’ve seen presentations addressing effects on soils, streams, tundra, boreal forest, sea ice, and even day-to-day operations of the diamond mining industry. I’ve also learned a lot about the climatic drivers creating the changes.

Here are some of the findings I heard about today and found interesting. Climatological and remote sensing data show that with the retreat of sea ice, we are observing increasing snow and increasing heat anomalies in arctic weather. These are some factors that those of us doing terrestrial ecology may want to pay attention to. There appear to be good correlations between sea ice extent, air temperature, and NDVI (a proxy for plant production), suggesting a relatively deterministic system with mechanisms tractable for study.

At a smaller scale, I presented some information about local-scale plant soil interactions in the tundra. I was interested to see some of the same nutrient cycling issues discussed in talks about streams. One workshop participant pointed out that streams do not appear to show the same trends in NO₃^– that our research group has observed in soils. We tend to see more nutrients available early in the season whereas he said that NO₃^– concentrations tend to go up in streams throughout the summer. I think everyone agreed that it would be cool to try to link these terrestrial and aquatic nutrient dynamics throughout the season.

Microplate OPAME for total free amino acids

I have spent some time over the last few weeks measuring total free amino acids in soil solution collected with microlysimeters. I’m posting some tips here in case it helps anyone else who might be doing this. The abstract of the primary reference paper on this method explains why it is used:

The spectrofluorometric procedure relies on the reaction of free amino acids with o-phthaldialdehyde and -mercaptoethanol [OPAME]. The fluorometric method is much more sensitive (working range 0.1-50M) than typical spectrophotometric analysis procedures for free amino acids which employs the ninhydrin reagent (working range 10-500M). In addition, the method only requires small sample volumes (1-50l), is rapid (1min sample^-1), simple to perform, and linear over a concentration range of three orders of magnitude (0.1-100M).

We have been using this procedure on 96-well microplates and when it works well, it is fantastic because there is only one reagent to add to the samples (OPAME working reagent) and the plates can be read immediately. The reagent is also sensitive to NH₄⁺, but it is much more sensitive to amino acids (e.g., for our Leucine standard, it is ~40x more sensitive to Leucine on a molar basis) so it is easy to subtract out the small contribution of NH₄⁺ to fluorescence.

In the arctic tussock tundra soils I’m working with, we have to run an additional plate with no OPAME reagent (buffer and sample only) in order to subtract out the autofluorescence of the brownish soil water samples that we are collecting from the microlysimeters, but that is fairly easy to do.

The tricky part of this procedure is working with the OPAME reagent because it is a bit fussy. It has a high background fluorescence that changes over time once the mercaptoethanol is added to the OPA. To measure the low concentrations of amino acids in soil solutions, you want to hit the sweet spot where background fluorescence is minimal but reactivity is still high.

My best and admittedly pretty rough guess at this time is that 30-50 hours after adding mercaptoethanol to the working reagent is the best time to use it. This can require some advance planning and we have begun keeping two batches of the stuff so we can have some ready on any given day. (You can regenerate the working reagent by periodically adding more mercaptoethanol).

The other challenge in using OPAME with microplates is that the fluorescence of the samples will decay rapidly (on the scale of minutes) after you add the reagent to them. This means that you have to add the OPAME as simultaneously as possible to all wells of the plates that are being used. Slow pipetting will cause wells on the last-done part of your plate to have inflated values.

Also, it’s very important to draw all OPAME reagent from the same source over the course of pipetting, so don’t refill your pipetting reservoir mid-plate. When using a multi-pipette, it is good to pre-wet the tips by pipetting OPAME reagent in and out a few times before adding the reagent to the plates to make sure volumes are consistent among wells.

The background fluorescence can be minimized by using a low OPAME reagent:sample ratio. We have been using 50 ul of sample and 100 ul OPAME. Even less OPAME could probably be used as I don’t think it’s anywhere close to saturating at these levels. This dramatically improved our results over when we were originally using 20:200.

Hope this helps!

First dissertation chapter in print

Can be found here!

I won’t deny it: this paper is pretty abstruse and is only relevant to a handful of soil scientists who study this stuff. My father-in-law printed a copy to read in the bathroom and honestly I don’t recommend that. That said, it actually does have some new and interesting findings in it.

I examined levels of landscape heterogeneity (quantified using semivariograms) in soil organic matter carbon and nitrogen from two different soil density fractions using soils from Niwot Ridge. The density fractions contain soil organic matter with divergent residence times and thus represent soil organic matter of different ages. Though many studies have examined landscape patterns in soil organic matter, few have examined patterns in the different chemical constituents of soil organic matter. The results contribute to a growing body of evidence that suggests that microbes essentially homogenize the widely varying chemical inputs that plants provide. Pretty cool.

Plots melted out and instrumented

Our experiment is progressing well. Here’s a video of the removal of our fabric on our accelerated snowmelt plots.

About 4 days later, our control plots also melted out and they are now fully instrumented with passive warming chambers and our ‘mantis’ instrument arrays.