Author Archive

Ode to six-legged wonder

September 8, 2017

If four legs are great, six legs are better. Right? For forty years now, the Portal Project has primarily focused on two-legged creatures trapping, studying, and sometimes cuddling small, furry four-legged creatures. But we haven’t ignored the six-legged inhabitants of our long-term research site, and I am going to tell you more about them now.


Which one of these is least like the other?

I am the Ernest lab entomologist, who doesn’t consider herself an entomologist. I am interested in biodiversity, community, and macroecological patterns like those studied by generations of Portal rodent researchers. I just happen to study them using bees. There are over twenty thousand species of bees in the world, and about four thousand in North America. My research so far has focused on the community ecology of native bees in a global hotspot of bee diversity in California. But as fate or luck would have it, another documented haven for native bees lies just down the road from our long-term rodent site in Portal. Between 2000 and 2007, bee researcher Robert Minckley documented 383 different species and 69 genera of bees from the San Bernardino Valley of Arizona and Mexico. The Smithsonian’s Southwestern Research Station, which is nestled up in the hills only a few miles from our rodent site, also celebrates the insect diversity of this area with field courses focused on bees or ants that attract dozens of eager entomologists from all around the world every summer.


Look familiar? This is the San Bernardino Valley on the border of Arizona and Mexico, just 50 miles south of our Portal research site. By BAlvarius – Own work, CC BY-SA 3.0,

A few months ago, I set out to see what kind of six-legged creatures we have scurrying around at Portal. My adventurous and ever-curious lab mates were willing to help, and (I think) even had a bit of fun learning to collect, pin, and curate bugs. During the March and May 2017 rodent censuses, we used a variety of passive (let the insects come to the trap) and active (go get ’em!) methods of insect collection to sample the local diversity on six legs:

  • Pan traps are brightly colored bowls full of soapy water that (passively) attract flying insects like bees that visit the bowls, fall into the water, and can be strained out and examined later.
  • Aerial nets (or “butterfly nets”) can be used to actively snag insects flying by or foraging on flowers, or can be used to more methodically sweep the ground to sample insects hiding in the grass or shrubs.
  • Blacklights lit underneath a white sheet at night passively bring in moths, roaches, and other nocturnal creatures that are drawn to ultraviolet wavelengths, perhaps because they confuse them with the moonlight they use to navigate, and can be scooped into collecting vials from their perches on the sheet.
  • Pitfall traps are plastic containers that are buried in the ground with their top openings flush with the ground level and covered with a coarse mesh with a hole in the center. Insects walking along the ground cross the mesh, fall into the hole, and are preserved in a small amount of ethanol at the bottom of the container. Do you think this is active or passive insect collecting? (Hint: we leave them out all day and only come back at dusk to check what is there.)
  • The crowd favorite was probably the “beat sheet,” which is simply a white sheet placed on the ground underneath an intriguing shrub, which is then vigorously beat with a piece of PVC pipe, a shovel, a hockey stick — really any bludgeoning tool will work. Insects that fall onto the sheet are then sucked up into an aspirator device — like a tiny, scientific, human-powered vacuum — and transferred into a collecting vial. This method of collecting is pretty active!

The spoils of our collecting efforts were creepy, crawly, and diverse.  The field crew had fun learning to pin them using the cooler and truck tailgate as our insect lab. Then I mailed them back to the lab and have been working on labeling, curating, and identifying our new Portal insect collection.


Then just last week I took them to campus for a photo glam session with the fancy microscope camera. Take a look:

This slideshow requires JavaScript.

I’m certainly not the only one who has been interested in the insect fauna in and around the rodent plots at Portal. Ants were actually a big part of the original research design. Between 1977 and 2009, ant colonies were censused once a year at every stake on all the plots, and between 1988 and 2009 ants were also baited (with Pecan Sandies!) and counted at 25 stakes in the thirteen unmanipulated plots. Research papers using this data, which is available to the public on the Weecology GitHub PortalData repository, have been authored by Tom Valone, Mike Kaspari, and more. Other brilliant ecologists like Deborah Gordon and Nate Sanders have also studied ant diversity, behavior, and community composition in the valley around our research site. I was actually following some of these researchers’ work long ago when I was first looking for graduate advisers and nerding out over my giant book about ants. And now here I am, working with the rodent branch of the legendary Portal Project while focusing on bees for my own work. It’s a small world after all, full of small six-legged wonder.


Learn more by checking out some of the published science on the ecological entomology around Portal:

Gordon, Deborah M. 1999. Ants at work: how an insect society is organized. Simon and Schuster.

Davidson, D.W., Inouye, R.S., Brown, J.H. 1984. Granivory in a Desert Ecosystem: Experimental Evidence for Indirect Facilitation of Ants by Rodents. Ecology, 65(6), 1780-1786.

Kaspari, M., & Valone, T. J. 2002. On ectotherm abundance in a seasonal environment—studies of a desert ant assemblage. Ecology, 83(11), 2991-2996.

Minckley, R. 2008. Faunal composition and species richness differences of bees (Hymenoptera: Apiformes) from two north American regions. Apidologie. 39: 176–188.

Sanders, Nathan J., and Deborah M. Gordon. 2003. Resource‐dependent interactions and the organization of desert ant communities.” Ecology 84.4: 1024-1031.
Sanders, N. J., & Gordon, D. M. 2000. The effects of interspecific interactions on resource use and behavior in a desert ant. Oecologia, 125(3), 436-443.
Valone, T. J., & Kaspari, M. 2005. Interactions between granivorous and omnivorous ants in a desert grassland: results from a long‐term experiment. Ecological Entomology, 30(1), 116-121.


2017-02-25 06.57.53

This four-legged creature thinks Portal is just the best place she’s ever been.



Walking the beat

October 12, 2016


Last month we highlighted the brains (and a little brawn) of The Portal Project, with a description of the new regime shift research and the requisite hardware-cloth-battle of 2015.  This month we bring out the big brawn guns (and some brains) to show you how the site keeps its youthful glow year after year in the unforgiving desert.  If we could bottle this Portal magic, it would be a best-seller for sure. Here’s our best attempt:


Do you like long walks in the desert?  Do you love cute, cuddly animals?  Do you like to take long walks in the desert with rattlesnakes, scorpions, and tarantulas when it’s 115 degrees out, and do you still love cute, cuddly animals when they bite you and poop on you?  If you answered yes to the last question, you might be Portal Protectors! material. (By the way, Morgan is recruiting another PhD student to work at Portal.  Please see her announcement about that here.) If you answered ‘yes’ to questions one and two but ‘no’ to question three, this blog post, accessed from the comfort of your armchair, is our gift to you.

To be a Portal Protector! is to be a biologist, a naturalist, an outdoor enthusiast/athlete, a team player, and a bit of a masochist. We get up early. We go to bed late. And we spend many of the intervening hours digging trenches, moving large rocks, pounding stakes into the ground, lugging heavy equipment to and fro, crawling around on our hands and knees amidst sharp objects, and catching small critters.  If any of these images below speak to you, then we’re on the same mental page:

Before you assume that these types of activities must only be for blockbusters, game shows, soldiers, or insane people, let us instead convince you how they are necessary and beloved steps in the quest towards scientific truth (I sense a blockbuster there…).  In fact, walking the beat of the Portal Project, including all of the aforementioned digging/crawling jobs, was the start to a successful career in science for a long and distinguished list of people.  Maybe you are one of them!  For that, we salute you.

Nearly every month, for approximately the last 453 months, one of the Portal Research Assistants has traveled to the site with a brave volunteer to census the rodent community.  Journeying from the original University of New Mexico headquarters, then from Utah State University, and most recently all the way from the University of Florida, they give up their weekends to keep the research going. (Keep the Lights Burning, Abbie…anyone?)  The current Portal RAs, Erica Christensen and Ellen Bledsoe, leave Florida on a Thursday to travel to Tuscon, Arizona, where they then pick up the truck, gather supplies and drive three hours to the site on Friday to set rodent traps on 12 of the 24 quarter-hectare plots before the sun goes down. Early the next morning they process the rodents caught in the traps, recording the species, sex, hind foot length, weight and pit tag number. They then bait and set traps on the other 12 plots to census the next nights’ activity, hauling wooden boxes of metal Sherman traps around a dizzying desert maze — like a Halloween corn maze, but made of spiny acacia bushes and located even more in the middle of nowhere.  After recording details about those rodents the next morning, they download data from the local weather station and make their way back to Florida by Monday night. It all looks something like this:

Many interesting scientific findings about how body size, metabolism, rain, species interactions, and temporal dynamics influence the structure of a rodent community, as well as a unique and impressive long-term community ecology dataset have come out of this workAnd since ecology is all about understanding the ways in which groups of living things interact with each other, additional results have come out of pairing the rodent census data with information collected twice a year about plants.

In both March and September of 2016, the current conglomeration of the Portal Protectors! traveled from Florida to Arizona to chase rodents, crawl around on our hands and knees counting plants, and perform our biannual trench-digging, rock-moving, stake-pounding, equipment-lugging rituals.  The Science Gods demand some weird sacrifices to ensure the everlasting bounty of Portal data.  And we must obey.  For the biannual plant census, we work together to record the sources of greenery around the site. We use meter quadrats (squares made out of PVC pipe, 0.25m on a side), placed at 16 permanent locations evenly spaced across each plot, to count and identify every tiny bud and blade of grass. We then string transect tape in a giant “X” across each plot and measure the width and height of every shrub it crosses. It’s spiny, tedious work. But the show must go on. In blistering heat and unrelenting sun, or through shivers and downpour, the plant census process takes about four long days each season, and looks something like this:

We often return from a plant census bruised, bloody, and burned, ready for a soft bed, a warm shower, a flushing toilet, and a clean shirt.  But we are only out there for a week.  Portal herself must endure the savage sun, devastating aridity, and seasonal deluges year-round.  Four-hundred and fifty-three rodent censuses and thirty-five plant censuses under the daily assault of the desert would take their toll on anyone.  So when the Portal Protectors! make our biannual pilgrimage from the lab to the site, we usually reserve some time, and some brawn, for a little Portal TLC.

Next time on the Portal blog, we’ll unveil both the long-overdue makeover of Plot 24, and our brand new shiny weather station!  This is where all the real trench-digging, rock-moving, stake-pounding, and equipment-lugging comes into play.  We might even include some original pictures of us doing these activities, instead of just George Clooney’s excellent portrayal of our work.  So stay tuned.  And stay safe.  It’s a crazy, dusty, desert world out there. But someone has to keep the lights burning.



August 3, 2016

We return this week from our special, breaking-news post about the recent reappearance of our one-hit-wonder, Twitter-sensation, spectabulous Banner-tailed Kangaroo Rat. This T-Rex of Portal may not be here to stay, but we’re sure excited she stopped by. What is here to stay is that pesky plot switch we mentioned last month. We’re going to continue our series of Portal science updates and tell you all about that now:


The last time we checked in at this blog prior to the 2015 plot switch, Erica was battling monsoon season to record desert rodent dynamics on the twenty-four long-term experimental plots that have been censused almost monthly since the site was established in 1977 by James Brown, James Reichman, and Diane Davidson. That’s thirty-nine years of tracking the occurrence of various species of small mammals. That’s over four hundred visits to Portal, AZ to trap, measure, weigh, and tag rats. And it all started before scientists had thought very much about why fluctuating species abundances in a community might be interesting.

This is cool because now we have decades of data (most of it publicly available) on how these scurrying, hopping, burrowing creatures have been interacting at Portal, just in time to see a surge of researcher interest in community ecology and species dynamics. For example, we were able to document an abundance of the small, pink flower Erodium cicutarium in plots where the competitive seed-foraging Kangaroo rats were excluded (Allington et al, 2013). And, after many years of pondering its absence, we recorded a resurgence of the Northern pygmy mouse, Baiomys taylori, a peculiar trend that would have gone undetected without regular sampling efforts. We’ve seen shrubs increase in both size and abundance, changing the entire look of those plots, not to mention the nature of the local foraging and shade resources. Ecology, in all its complexities, happens over long time scales. To understand it, we must record it over short ones. Few studies exist that have managed to do both, and we are excited about the scientific opportunities the history at Portal affords us and others who use our data.

Erodium cicutarium (left) has been increasing in abundance in plots where Kangaroo rats are excluded.  Baiomys taylori (right), the Northern pygmy mouse, has resurged from apparent rarity.

So now that we have amassed this monstrous dataset and finally understand more about how these rodents have been faring, cohabiting, and influencing the plants of these twenty-four plots in long-term treatment groups over nearly forty years, naturally we decided to turn it all on its head. Yes, after over four hundred samples of the twenty-four plots in their original treatment states, we got a grant from the National Science Foundation to switch them all around. Why? Because we’re scientists, and we like to poke systems to see what happens.

Because we scientists often want to be useful in addition to curious, we also like to simulate real expected ecological change so that we can predict likely outcomes and plan for them. Our world is rapidly changing. We need to understand how ecological communities will respond. When we’re not watching and recording, or sometimes even when we are, seemingly small changes can add up to big shifts. A breeze, a little water vapor, a small temperature change can suddenly turn into a monstrous hurricane, for example, that introduces a whole new set of rules and challenges to human existence.  Similarly, ecological systems can undergo extreme, abrupt changes in state that are very hard to understand and manage unless they have been tracked before, during, and after that transformation.

Questions surrounding the idea of regime shifts, described as dramatic changes in populations, communities, and ecosystems over short periods of time (Hare & Mantua, 2000), represent relatively new challenges in the field of ecology, ones that the Portal project may be uniquely situated to address.  Understanding regime shifts, or how communities of rodents may suddenly shift in their relative species abundances or resource usage, at Portal may help us understand what to expect from other ecosystems undergoing unprecedented levels of environmental change.  We’re not promising X-(woman)-like vanquishing of Hurricanes. We study rats, not wolverines after all. But this is important stuff.


While hurricanes are exceedingly rare in the Arizona desert, there are subtler forces at work which may cause shifts in our Portal rodent community. Knowledge of these forces may help us, and other scientists, understand similar sudden disruptions in unmonitored groups and ecosystems. It’s the scientists’ mutant superpower — studying one thing ‘over here’ can help us predict and manage another thing ‘over there’ which we may have actually never seen…except in our mind’s eye.

The figure below, from Dr. Ernest’s 2014 grant proposal, shows three scenarios in which ecosystems and drivers (e.g. climate, nutrient input, biotic interactions) can be related in ways that might represent system-disrupting regime shifts. Sudden shifts in drivers (e.g. hurricanes) could cause a corresponding shift in ecosystems (e.g. massive urban destruction), which would constitute a regime shift. The same ecosystem pattern, however, could be triggered by simply crossing a threshold along some seemingly innocuous linear increase of a driver (like a slowly rising sea level that causes the sudden collapse of a city when it finally submerges the business district, or the state of water as it’s gradually heated past its boiling point). Regime shifts, like the melting of sea ice caps, can be hard to undo. Sometimes an ecosystem can be brought back down from a boil by turning down the driver dial. But sometimes an ecosystem will fail to revert back to its original state after a driver has increased and then decreased again, taking the ecosystem along a new trajectory. At Portal, where neither massive urban destruction nor hurricanes are a major concern, this might come in the form a plant that fails to return to a location after a particularly dry summer even after rains have resumed, or a Banner-tail Kangaroo rat who once reigned mighty and may never be seen again.


To poke the system and test this at Portal, in March 2015, we reversed some of the long-standing plot treatments. We also, of course, maintained some of the plots in their original treatments to serve as reference plots, against which we can test the existence and magnitude of potential regime shifts caused by introducing a large granivore as a driver. This has rarely been experimentally tested because few systems have the long-term data, ability to simultaneously cause a jump in an ecosystem driver in both directions, and enough plots to maintain replicates with reversed and reference plots.


The Portal project’s twenty-four plots and well-monitored, gated rodent communities is an ideal system to study regime shifts because it does not have these common experimental design limitations. Because we manipulate rodent access to plots via gates of different size, it is also shockingly easy to change the rodent community on a plot – we can make new gates by clipping holes in the fencing, remove gates by applying patches over them, or change the size of gates (and the species that can enter the plot) by either patching or clipping to create new hole sizes. So last spring, Dr. Morgan Ernest led the hardware-cloth-stripping-team in reassigning plots to their new experimental treatment, ending an era, but never the science:

Erica securing the new hardware cloth pieces to close the old gates (left).  This little guy (right) wishes we could strip the hardware cloth off him.  But we are done with that task.

In a gesture of inclusiveness and diversity characteristic of the Weecology lab group ideology, we broke down barriers between different rodent communities by enlarging or creating gates in some of the plots that had previously excluded the granivorous, large-skulled (and adorable) Kangaroo rats:


A Kangaroo rat, Dipodomys merriami.  What plot wouldn’t want these cuties?

But because this is science and not social revolution after all, we are also testing the influence of these large-skulled favorites as ecosystem drivers by excluding Kangaroo rats from some of their previous home plots by patching up or narrowing gates in plot fences that had previously allowed them to pass into these areas.

Example of a gate in a Control plot, open (left) and closed (right), which provides free, open access for all knowledge-(or seed)-seeking rodents!
Example of a gate in a Kangaroo rat exclosure plot, open (left) and closed (right), which is very oppressive of Krats, but great for controlling their influence in plots as a driver.


In case you have ever doubted the dedication of field biologists, thinking perhaps that we enjoy a life of sipping Coronas in a scenic field station after a day of wandering through the hills sniffing at plants, we would like to explain to you the unnatural nature of hardware cloth. Hardware cloth is not cloth. It is, however, hard. It is also pointy and sharp and vexingly narrow. To enact the plot treatments switch described above, the valiant hardware-cloth-team-of-2015 used pliers to strip countless thin layers of wire off of small metal squares of fencing, creating exposed, pointy wire ends along the hardware cloth, which must then be woven through the tiny squares in the existing plot fence to close off or narrow old treatment gates. They then cut or adjusted new treatment gates in the existing fence hardware cloth to create new access for the Kangaroo rat ‘drivers.’ Science is not all high-tech gadgets and sophisticated computer algorithms. Sometimes real science is stabbing yourself with hardware cloth one thousand times and squatting on the ground in the blazing hot desert to sew patches over gates in metal fences because you want to see what the rats will do. We happen to actually love this type of self-brutality in the name of science, and how it makes our evening Coronas, enjoyed together around the campfire at our beautiful field station, taste that much better.


The field crew taking a break at the Ramada, our two-sided field station.

And that is the story of how the plot switch, and the finger-massacre, of 2015 was recorded in Portal history. We won’t show you the blood, but we will show you the final plot map result, and a much cleaner schematic of plot treatment shifts:




Next time on the Portal blog we will fast-forward to a year after this historic plot switch when, wiser, hardier, and with more Band-Aids and guacamole in tow (though there is never, ever enough guacamole), the Weecology field crew hit the road again, in our new University of Florida vehicle, to undertake the 447th sampling of desert rodents, and the twelfth under a possible new regime.

PORTAL: The times, they are a’changin

July 14, 2016

Updates on temporal community dynamics, and a whole new project scheme.


Things have been quiet on the portal blog lately.

But in the lab and the field, it has been anything but.


Over the past year there have been big changes afoot for the Portal project.  In the summer of 2015, Weecology lab headquarters relocated from Utah State University in small, mountainous Logan to the massive University of Florida campus in subtropical Gainesville. So now we study Arizona’s desert rodents from the mossy groves of the southeast rather than the alpine forests of the Rockies, like true cosmopolitan, ever-curious ecologists.


The Portal project headquarters relocated from Utah to Florida in summer 2015. Leaping Krat photo illustration by Molly Zisk, taken from

If you’re going to box up your life, you might as well reorganize it too. In the midst of planning her transcontinental move, Dr. Morgan Ernest, supported in her decision by funders at the National Science Foundation, decided it was time for a shuffle of Portal’s long-term experimental rodent community plots as well, a reorganization of the plot gates that allow some rodent species in, and keep others out. Yuri Kochiyama once said that movement is contagious. said that science has finally proven that devoted dog owners often resemble their dogs. Maybe these things have nothing to do with the rigorous study of kangaroo rats in the Arizona desert. Maybe Yuri was talking about politics, not ecology and nobody actually thinks Peyton Manning looks like his dog. But maybe, when you study the ebb and flow of desert rodent populations for long enough, they have to move when you move. That’s not science. But stay tuned and we’ll tell you all about the real science of the Portal project plot move, and the people doing the moving. In the full words of Yuri, “the movement is contagious, and the people in it are the ones who pass on this spirit.”

YuriDogsActivist Yuri Kochiyama ( and people who look like their dogs (
This is silliness.  Stay tuned for science.

It takes a community of people to shuffle a community of rodents, and to pass on the spirit of the Portal project.  To facilitate all this contagious movement, the Portal project lab group, like all strong biological populations, has grown in numbers over the past year. One might even call Dr. Ernest the Pied Piper of Ecology, enlisting a group of young followers in her slow migration from the mountains towards the sea. In 2015, she recruited another smammal (small mammal)-loving Portal RA, Ellen Bledsoe, to join Erica Christensen in conducting the monthly plot rodent census, officially appointed our own Glenda Yenni as Portal project manager, and brought in another graduate student, Joan Meiners, from Utah to help communicate the many exciting new projects radiating out of the Portal, Arizona site. And even though Morgan does know quite a lot about rats, we’re all pretty confident she’s not trying to drown us.

So now that we’ve gotten the debate about doppelgänger dogs, social activism, and old nursery rhymes out of the way, we can dive right in to the nitty gritty science of the new Portal project movement – what changes we made on site, and why.

Next time on this blog series, we’ll (re)begin our Portal project story in March of 2016, one year after the great plot shuffle of 2015, when this whole crew, plus two willing workers from Dr. Ethan White‘s side of Weecology, Shawn Taylor and Kristina Riemer, headed to the Arizona desert to check in on the plot switch, conduct the Portal project’s 447th rodent census and the 34th plant census (with a new high tech twist!), and dig a giant trench in a desert near-tornado.

PiedPiperThe Pied Piper, keeper of the rodent gates.