21.2.16

Drones in Dunes: A Weekend Mapping !Nara



Gobabeb has begun a few newer research projects, instigated by the arrival of some new research equipment graciously donated by Dartmouth College after their annual visit to Gobabeb in November 2015. Among these research tools is the much-hyped drone, which had been used by the Dartmouths this year to begin mapping hummocks of !nara around Gobabeb and the Kuiseb Delta (read their report HERE). Our ongoing research on the !nara is benefiting from the arrival of our new ‘eyes in the sky’, as we endeavor to continue this !nara mapping around Namib-Naukluft Park.

Which is what brought us out to Far East, the weekend of 18th December this past year: mapping the !nara. We drove past Saagberg Mountain and the vast valley of stubby grasses and big sky, bumping along the euphemistically-labeled ‘track’ towards a secret series of !nara-filled hummocks. Eliza Hoffman, our Dartmouth intern, had previously arranged several flights over these hummocks, which Gillian and her !nara team have already done research (ongoing research) on the phylogeny and condition over time on these dense populations of !nara.

Setting up the "batcorder" over beautiful sunset. Saagberg
("Saw tooth") mountain in the background
After we had overcome the tremendously bumpy track, with the help of fearless off-road driver Jess Sack, we arrived in between our targeted hummocks in time for sunset. While we set up camp and enjoyed the view, we began setting up a “batcorder” (or bat echolocation recorder) to record in the area while we were there. This is part of the ongoing bat surveys that one of our newest arrivals, Angela Curtis, will be working on for her Master’s degree. Since we’ve little idea of where and what kind of bats frequent different corners of the Namib, she’s looking for evidence of bats in all kinds of places - !nara hummocks included! (You can read more about her project HERE). The drone flying would happen in the morning, but we would have to find the approximate “take-off” points that Eliza had programmed into the flight paths for the two hummocks. The take-off scouting of course gave us the chance to explore the area – and the stunning scenery that is the Far East. While I was scouting around the hummocks, I got to hunting after the subject of my would-be pet-project: the Namib golden mole. The trails were fantastic (and fresh!) around this hummock, and you could clearly see the diversity of spoor left by this one little critter. Check out some of these meandering trails and surprising spoor...
 


Popcorn and wine for dinner, set to the background chorus of barking geckos. We all embraced the productive silence around us – just knowing that there was so much life around these dunes, that it existed inside this bubble of Namib-Naukluft Park, is somehow very comforting. And there really is nothing like star-gazing in the desert!
Panorama of the dunes surveyed; courtesy Jessica Sack

The next morning we set up for the drone launch. Here’s the general gist of the drone-flying set-up: there are four modes for the drone, STD (Standard) = hovering, LTR (Loiter) = hold position, AUTO (Command to take the planned flightpath; essentially ‘autopilot’ for drones), and RTL (Return to Launch). For our drone at least, it’s not recommended that you free-fly it with the joysticks – remote-controlled car-style. This uses up a lot of battery, and since our drone doesn’t have a lot of juice to each of its lithium-ion batteries, it’s more productive to use the pre-planned flightpath feature. Once we’ve set up the ‘launch pad’ for the drone – consisting of a tarp laid on the ground for the drone, and the computer perched and ready with the flightpath and control panel up – you can start setting up our drone for flight.

HOW TO FLY (THE GOBABEB) DRONE (ish)
Step One: Plug in the battery at the base. Usually, we let it sit for a half-minute or so once it’s lit up and started searching for its satellites. The controller and computer panel will tell you when it picks up satellites. Once it’s R2D2-beeped at you relaying this message, you’re good to continue.

Step Two: Attach the camera to the base of the drone. We’ve hijacked a shock-proof Canon point-and-shoot to take continuous photos at regular intervals, which we just Velcro-strap onto the belly of the drone and turn on before lift-off.

Step Three: Connect the drone to the computer. This is an important step, to ensure that the line of communication between what the drone is doing and the computer is registering is clear. The drone’s flight path will be on the control panel, and they’ll be ‘talking’ during the whole thing – relaying everything from its current height, speed, perceived wind direction, etc so that you can read it all from the computer and foresee any possible issues with the drone.
Pilot Eliza and Meg watching drone lift-off at Dune 1

Step Four: “Arm” the drone. This basically tells it to get ready for flight, and it starts up its (surprisingly noisy) engines.

Step Five: Lift off! For our quadcopter, you just flick the left toggle down towards the right, and slowly, carefully throttle up with the joysticks to the needed height before you select the AUTO mode and it launches into its flight plan. For the drone to keep connection with the computer, the controller needs to retain at least 40% connection; the controller will conveniently beep for you if it senses it’s getting too far away from the drone.

Once the drone is up, up and away, you’re mostly just watching it – for us, we had at least one person watching the computer control panel and relaying information as necessary to Eliza, who was piloting with the controller. Although it does mostly ‘fly itself’, it’s important to watch the drone anyway – sometimes flukey things will go wrong, and it’s better to be watching in case you have to make an emergency landing, or if it strays from its path have it RTL. Our first flight, for example, went pretty swimmingly – until the end, when it was returning to launch, hesitated about 200m away from the launch pad, and promptly began plummeting towards the ground. Since we saw the battery was declining too rapidly, we were able to anticipate this, and Eliza caught the drone before it could get itself tangled up in a spiny thorny !nara mess. And we also knew why this happened – we saw on take-off, once satellites had registered the global position of the drone, that our chosen start-point was a little bit further than the programmed one. So the drone ended up traveling further from the launch pad to the start of the planned route, and was thus a tad short on juice on its return flight.

Our second flight went swimmingly too, as we all watched the drone zooming in line-transects above the swaths of !nara. Following the flights, we did some sample surveying of the !nara being photographed, noting the sexes of the plants (one of the things we’re interested in looking for patterns about with widespread !nara mapping; since !nara plants are dioeceous [plants are either male or female], there might be interesting insights regarding where male versus female plants predominate) and noting general condition. Worryingly, many of the plants we observed were getting absolutely hammered by the wildlife in the area – the female plants always get a little beaten up in the frenzy over the fruits, but many of these plants’ conditions was cause for concern according to our resident !nara expert, Gillian.
Eliza, Meg and Gillian survey !nara hummocks

And the future of these plants is a big reason for the ongoing work that Gobabeb is doing into the !nara plant, given the relatively recent developments in !nara tourism products, and the resulting conflicts facing the Topnaar community. Read more about !nara and the work that Gobabeb’s team is doing on our website.







x

Ten Days in RITS’ shoes. Or, The Week Meg Got Promoted Thrice [Sept 2015]



As much of a meeting place as Gobabeb often is, of people and creatures alike, there are times when our staff gets a bit pared down. It is during times like these, and as in this episode relating to a particular week in September of this past year, that we like to emphasize the point in all Gobabeb job descriptions (and emphatically noted in the Grinnells’ job postings) of being flexible. Gobabeb is a small family, with plenty of responsibilities to take care of, and there are times when it really becomes ‘all hands at hand’ around here. To illustrate, let’s hop aboard a time machine back to mid-September of this year…

It is around September 17th, 2015. I am just returning with Gillian from my short excursion to Windhoek, where I successfully sought an intact (but, as it would turn out, structurally deficient) boot for my newly Fickle Foot. I arrive back to much fanfare – less for my arrival and more in celebration of our lovely IT guru (and featured previously in a particularly memorable encounter with a horned adder) Doris’s birthday – and cheesecake. (Glittery cheesecake, no less! A marvelous treat of our generous chef Hendrik’s creation) No sooner have I arrived back at the station then everyone else left – well, not everyone, but the bulk of the research section was leaving for about a week for a series of conferences in Windhoek.
The upshot? Doris, Gillian, Johanna (our chipper receptionist and accommodation assistant), and myself were on our own for the week. Which left us with all of the research tasks to take care of, for about 10 days. Here’s how we made it work out:
Doris
Meg
Pitfall Traps (M-W-F)
Weather (Thrice daily)
BSRN (Daily at 8h)
NOAA Sampling (Thurs or Fri)

FogNet Monitoring

MPI Sampling (Thurs)






Gillian and Johanna were around for support, while Doris and I basically tag-teamed it as needed. Fortunately, there were no needy school groups or such things around in Training’s hair for the week. A little background on each of these tasks, before I go into storytelling of my week as RITS:

Pitfall Traps: How intriguing-sounding, hey? An easily-deducible answer – it’s literally spelled out in the name. A pitfall is nothing more than a bucket buried so that its lip is flush with the surrounding soil surface, where it waits for anything crawling around to drop in for a visit. They’re an easy, cheap method used to survey insect and crawling-critter populations anywhere, and a particularly useful one around Gobabeb. GBB has been using pitfall traps to sample the area’s famed variety of beetle species for many decades. There are two key, long-running projects that have been making good use of the hundreds of buckets (and bucket-like objects) around Gobabeb: regular beetle sampling around the three ecosystems, and the East-West dune sampling.

I have previously referenced the East-West Dunes excursions, which happen semi-regularly every three-ish months, and take you out into the less-frequented dunes of the Namib Sand Sea, east of Gobabeb. I went on one-such outing in August, when we were testing the methods of the anticipated “Watering the Desert” experiment of this year’s SDP 19 programme in December, in addition to the standard tasks of typical E-W outings. Here’s how it usually works:

[Recipe for East-West Excursion]
2 Research Technicians (more hands are more helpful!)
40 buckets
2 spoons
Weather sampling kit
Food for 3-5 days
Water for 3-5 days
Beetle ID book
Sampling vials
2x 25 m measuring tapes
1 square-meter quadrat
Plant ID book
Spade
Cook stove, table, chairs, eating utensils
Toilet paper
Sleeping bags/personals

The responsible researchers head out early on a Monday morning to trek through Namib-Naukluft Park in what amounts to a massive loop back to the dunes that lie east of Gobabeb – if we could go directly from Gobabeb as the crow flies, it would be a much more direct route. Unfortunately, our vehicles don’t yet fly over the dunes, unless you’re driving them in a vehicle-lifespan-shortening fashion. It is reported in GBB gossip that this direct-if-slightly-treacherous route has been previously attempted, but the dunes were deemed too sharp-angled for our vehicles to survive the route intact. In any case, the route to the start of the E-W transects is a beautiful drive, especially when begun in the wee hours of the morning: you get a great view as both the sun and temperature slowly rise over the horizon of the gravel plains. A couple of hours into the drive, you’ll reach the gate – we actually re-enter Namib-Naukluft from a sort of back entrance, with permission of MET of course, and begin the trek along the far-less-traveled path. The tire-treads in this part of the park are best described loosely as ‘paths’ because they are often little more than a faint shadow on the surface of the wash-ridden, boulder-strewn plains that stretch before you. If you’re able to catch a glance of the landscapes wrapping around the route, you’ll be amazed by the hills and valleys, the storied and still mysterious so-called ‘fairy circles’, the silent herds of zebra and groups of oryx that pop up suddenly. (Had you not just left a major motorway bisecting this portion of Namibia and frequented by many the white bakkie racing through these hills, the transition to these landscapes might not be so dramatic) In any case, you soldier on through the bumps and jolts of the pathways, until you find three different sites parked along the dunes – the first, Mniszechisvlei, occurs on the far side of what looks like a more modest ‘beginner’ dune. Continuing on up through a dried creekbed and past several challenging running-start-worthy dune fields, there’s the second site tucked away past a spring frequented by surprising numbers of wildlife. This one is Noctivaga. To find the third one, you’ll venture out along the rim of the X Canyon and its sweeping views of dark stone crevasses towards the furthest station, aptly named Far East.

Once you’ve arrived at your station, the set-up is as follows: usually one technician will start by setting up camp, including a shade if accompanied by other vehicles, tent, cooking station. The other technician will busy themselves with setting in the pitfall traps – loading oneself with about forty buckets and digging them into the slipface, the dune slope, and the interdune in sets of ten. These traps are what will be monitored by the researchers at different intervals – all of the traps must be surveyed at least thrice a day, once first thing in the morning, once around lunchtime, and then again in the evening. The slipface traps undergo special monitoring, and must be checked hourly for a full day. This labour of love is for the beetles. The 240-some species of beetle that are found in the Namib Desert are key indicator species, and their monitoring makes up an essential foundation of a lot of the environmental monitoring that Gobabeb has been doing for the longest. For example, one of the superstars of Namib ecology is Onymacris unuicularis, popularly known as the Fog-Basking Beetle, and is suspected to be changing its range and concentrating more on coastal dunes with more regular fogs rolling through. With the long-term monitoring that Gobabeb does on these further-inland dunes, we can prove or challenge these suppositions – all those hours of pitfall checks, thrice-daily weather measurements, dozens of data logbooks are the evidence on the ground that tells the story.

The more regularly-performed pitfall trap monitoring involves a series of pitfall traps in the three ecosystems around Gobabeb - the gravel plains, the riverbed, and the dunes - which are checked thrice weekly. It's the same concept as the Far East sampling, except these pitfall traps are permanently installed, so there's none on the slipface of the dunes (they'd be buried within hours, with all of the quickly-shifting sand at the peaks of the dunes!). The responsible researcher heads out to check these pitfalls early in the morning, before things heat up, armed with a spoon, the Beetle ID book, and the record book. Each bucket is thoroughly scooped to evict any of the temporary residents. Each of these is then identified, before being registered in the record book, and released away from the pitfalls (to avoid another visit too soon). 
Pitfalls Map around Gobabeb

The pitfall traps are just one among the wide range of projects that Gobabeb has been doing and hosting for the long run. These Long-Term Ecological Research projects (LTER) are what Gobabeb excels in. Speaking of long-term, one of the other pretty-fundamental-to-GBB research tasks I was tasked with was Weather.

Gobabeb has been taking weather data for its entire nearly-54 years of existence. For the most part, this has been in the form of our First Order Weather Station, where we continue to take weather measurements thrice daily at various times according to season. Within the weather station, you’ll find your basic weather gear – wind vane, thermometers galore, rain gauges – in addition to some less-easily-identified instruments. My favorite is the mystical-looking Sun Recorder: Gobabeb’s divining orb, this glass sphere concentrates the day’s sun rays onto a strip of paper, which is extrapolated into hours of sunlight received per day. The patterns in the line give us the daily report of sky conditions; for example, a slow start to the sun-burned line likely indicates the presence of fog that day. These strips are then photographed or scanned and sent in to the Namibian Meteorological Service, where they inform the long-term patterns of cloud cover, surface radiation and such.

Also at the weather station is our array of soil thermometers, a set of six reaching depths ranging from 5cm to 1.2 meters below ground; these tell us more about the soil profile over the course of the day (which has important ramifications for the wildlife around here!). There’s also an evaporation pan, an anemometer, and several fog screens, in addition to the Stephenson screen, which houses our full array of thermometers and our hydro-thermograph.

Gobabeb’s long-running and extensive weather record is one of its most valuable data sets. Particularly in an extreme if fragile environment, observing changes in weather phenomena and temperature is extremely valuable – because it is an environment of extremes, any changes manifest themselves much more obviously over the relative short-term. And, in today’s flurry of research on climate change, long-term weather records are really valuable to informing current/future/ongoing research, by providing a baseline on which other studies can be based and change can be observed and tested for.

We now have an automated weather station on site, as part of the FogNet tower at Gobabeb. However, we continue to take manual measurements at the FOWS because we want to ensure the relative continuity of our data in the long-term; if we switch to relying on the automated weather station at FogNet, then we need to see if there are any major differences with how they both measure maximum temperature, for example. If there are, our new data will have to make clear that the older data is not on the same relative scale/sensitivity.

What the NOAA box looks like
NOAA, the research task named after the American National Occeanic and Atmospheric Administration, is probably one of my favorite research tasks – it’s simple and easily performed, and it takes you out to the dunes! What’s involved: you take the box pictured below, the one which I am so enthusiastically toting, onto the quad bike and drive out to the designated spot. Haul box to northern edge of dune, and open ‘er up. Slide the extend-o-pole tube sampler to its highest reaches. Flip a few switches, and allow the gas flasks to flush for a few minutes while you take the wind reading (standing Statue of Liberty style towards the wind). Flip a few different switches and record a few readings in the box (pressure of pump, etc), before walking away again to let the flasks to fill with fresh Air of the Namib. Meanwhile, you get the opportunity to take in the view, marvel at the perfectly glazed quartz stones along the slopes, or even hunt for some Golden Mole trails…

After the sampling is done, you strap the box back onto the back of the quad bike in as secure and convoluted a manner as possible, and jet back to the back lab, where the flasks are placed with the filled-out info sheet into their appropriate box, bound for somewhere in the US of A in ~6 months or so (with a legion of its fellow flasks). The gas samples are part of a global data set of air samples that NOAA documents and tracks over time for a wide variety of trace gases and the like, and Gobabeb has been working with NOAA to provide air samples since 1997.

As long as we’re onto the alphabet-soup portion of the research projects, BSRN is one of our atmospheric-sciences projects that requires daily monitoring. The purpose of this robot-looking machine-upon-a-rock is to look at long-term trends and patterns in solar irradiation, both atmospheric and reflective off of the ground. It takes continuous measurements over the course of the day, and the Micky-Mouse-heads-on-a-stick portion of the machine tracks the sunlight directly. Our job is to ensure that the sensors are cleaned daily, and to record what type of cleaning we did – to correspond with local weather events and accompany the solar irradiation readings the machine is taking,
BSRN array at Gobabeb; picture courtesy Reyk Borner 2015

Most people don’t relate directly to the need for data on solar irradiation, unsurprisingly. So why do we painstakingly record whether sensors are wet when wiped (code 1) and whether the balance is accurate on each one (0 or 1) +-365 days/year? Solar irradiation is an elaborate term for how much energy (via light) we get from the sun. Understanding how much of that energy comes directly from the sun to the surface of the Earth, and how much is reflected back off of the surface into the atmosphere, is pretty important for a number of reasons: (1) weather! Energy in the system is largely what drives weather patterns, on regional and global scales. Getting a grip on how much energy is coming directly from the sun, and how much that varies daily, seasonally, geographically can improve our understanding of weather and climate. (2) sun power. Direct data regarding how much energy we’re receiving from the sun can support future research into and justification for different forms of renewable energy. In this part of the world, and as it happens at our very own station, we have a solar array that generates most of the power we use. This is an extremely fitting example of what is termed an “appropriate technology” – one which not only enhances the quality of life for a group of people, but does so in a relatively low-cost, efficient, and environmentally-conscious manner. The data provided by BSRN, or Baseline Surface Radiation Network, details where and when the greatest amounts of solar energy are received, and where they’re going (whether that is back into the atmosphere, or into the ground).

MPI is another of our more complex regular research tasks. It is another atmospheric science monitoring project, from the Max Planck Institute in Jena, Germany, which goes a little bit further than the NOAA weekly samples do.  MPI installed a shipping container filled with beeping, buzzing, clicking machinery and manned by four different computers connected by miles of thin metal wires and neatly-taped-together bundles, a suite of massive gas cylinders, a cryotrap, and a briefcase-sized gas sampler in 2012. This container with its ~12m tall tower are continually measuring trace gases in the local atmosphere. Like many of our atmospheric projects, this is one of a network of stations around the world, which all report back to the mothership in Germany, creating a more complete idea of global atmospheric gas concentrations. With all of that intimidating machinery and glam tech inside the temp-controlled container, you may wonder why we humans must enter this elaborate techy trap at all. Our weekly work with MPI requires us to manually capture several (four, to be precise) air samples to send back to MPI for further analysis. We also perform checks on the rest of the system, and answer any queries the MPI staff back in Jena may have for us – usually they can see everything that is (and isn’t) happening with the system through the manifold computer presence of the MPI box itself, but sometimes the computers need a little human help after spats with dust storms and such.
Meg carries the NOAA sampling box effortlessly


Usually MPI goes swimmingly. You ride out the 2km east of Gobabeb to the tower via quad bike or bakkie, open the gate (to keep out those wayward goats with an interest in atmospheric sciences), wave hello to the camera outside the container itself, unlock the container, and pop in. Once you’re in, you tell the computer who you are and what you’ve just done/are coming to do: for example, “Meg and Chris, entered container”. Then you start going through the regular procedure – which involves a lot of checking on things, like reading off the gas pressures for all of those tanks I mentioned, gingerly replacing the cryotrap, entering data into the spreadsheet, and of course performing the gas sampling. However – when MPI is not working, it usually takes a whole fleet of Gobabebeans and lots of correspondence to figure things out. Mostly, you just pray that the science gods are on your side when you enter that nicely-cooled container. It’s easy to get intimidated by the container, being a living caricature of science – but once you get used to the sampling process, and as long as you have the trusty guide developed by RITS alum Taylor Chicoine, you’re in good hands!

Finally, FogNet Monitoring. FogNet is a German-sponsored initiative to get a better idea of fog frequency and distribution along a 120 km gradient stretching from the coast (Walvis Bay) inland towards Gobabeb. The project consists of a network (thus Net) of nine towers stationed across the gravel plains, each equipped with a local weather station array, a fog harvester, and several pieces of fog-specific weather monitoring equipment. These stations collect real-time weather data and ‘collect’ the fog that passes, delivering that data directly to a website (accessible here). Our resident FogNet technician, Ruusa Gottlieb, oversees the ongoing operation of these towers and collects the data for later analysis. The researchers behind this project are looking for patterns in fog production and frequency over time across this gradient – specifically with change in mind. Climate change was a key driver behind the instigation of this study, because of observations by
FogNet Station at Gobabeb: two fog harvesters pictured
Photo courtesy Reyk Borner 2015
long-time Gobabebeans that fog production and frequency at the station have been declining over time. There are concerns that climate change is the culprit behind this too, owing to changes in ocean circulation and currents. The Benguela Current, which flanks the Namibian coast with upwellings of cold and nutrient-rich water from the Antarctic south of us, may be warming slightly. And it is this current that provides the cool air that then collides with the hot, dry air coming off of the Namib to reach the critical point of condensation, which is then blown inland by the coastal winds. If fog production is declining over time, it will have pretty serious consequences for the wildlife of the coastal Namib: both furry/scaly and photosynthesizers around Gobabeb obtain most of their moisture from fog, which is an obvious choice when it’s about five times more available as a source of moisture than fog is. Much of the unique biodiversity that calls this desert home is dependent on those fog banks that roll through, and if this region were to receive less fog (whether that’s in fewer events over the course of the year, less of the right kind of (usable and accessible) fog), it would spell trouble for the long-term sustainability of these unique and bizarre animals and plants.
Novald shows Meg a lizard during FogNet Monitoring, November

FogNet monitoring, while at the Station, involves ensuring that all of the towers are reporting their data to our servers here, and checking out the fog harvesters after fog events. When a fog comes through, we have to take a sample of the fog that is collected. This is then labeled and stored in our refrigerators until Ruusa takes a whole load of these samples down to UCT, where she is able to use their laboratory equipment to run analyses on what that fog might contain. Her Master’s degree research is a study of a known fog-collecting plant, Trianthema hereroensis, and whether it (and other plants/organisms) might be getting more than just moisture in the fog. Regular FogNet maintenance just has us measure the amount collected by the fog harvesters, and to ensure that all of the equipment is operating. Once a month, several technicians go to the eight towers across the gravel plains, where they clean all the equipment, collect the fog harvested for the month, and do general maintenance as needed at the stations.

Meg marvels at aforementioned lizard

These are, of course, hardly all of the research tasks that Gobabeb technicians are tasked with, in addition to their own research. Besides this is the Oxford Dust Models, several different towers operated by KIT, the Salt Spring CameraTrap, and Kahani Beetle sampling – just to name a few of the ones I’m familiar with. I don’t always get to do research work, but I enjoy learning everything from the reason we have four thermometers to read for ambient temperature, to getting to know some of the 80-plus species of beetles during pitfall monitoring. Some of my latest research exploits were had on a weekend out for Kahani Beetle Sampling – stay tuned for that story!