I'm writing this in the late afternoon of Day 3 of our mission - until now, we've been in a fair whirl of activity, with little time to sit, contemplate, and think about writing.

As I write, there's a small school of French grunts peering in at me inside the Aquarius, peeking in right over the display of this computer. I'm the one inside; they're outside, for a change. Beyond that, the scene fades to blue, the sandy bottom rolling away to the coral spurs. Not a shabby office view.

Our first day was more logistical than data-productive. Unpacking and stowing gear, trying out electronics, getting computers rigged and working. By the evening, we'd got ourselves pretty well fitted out for the next day.

Morning of Day 2 began to crank up the anxiety a bit. The computers (both of them) suddenly refused to talk with our data loggers (both of them). OK, OK. Think, think. We're experienced at this, we've debugged these systems before, in the lab, in the field. Test meters start flying, probes probing, adrenaline flowing. It's a little more than disconcerting to have the main data collection devices of your mission become null on Day 2, right after seeing them work on Day 1.

But, back to the plan, we do have the "no tech," "ultimate backstop" mode of data collection: the dissolving plaster models will still work. So, out they go. Ken and Matt headed out and placed three sites of three replicates of each of our three shapes. We now have something happening while we keep trying to salvage the technical side of things. But we've got no way to measure flow speeds around the plaster... Yes we do. Once again, Ken and Matt head out with a syringe of dye and a meter-long piece of steel marked as a ruler. With a divewatch as a stopwatch, they did excellent dye-movement timings over each plaster site. So we even have flow speeds, but still no tech.

Inside the habitat, we're narrowing the possibilities. The loggers themselves are working: we can interact with them using their keypads. The computers are working. The serial cables are wired correctly and check out. But as we probe communications with an RS-232 breakout box, we start to realize that there's a problem with the serial datalink between the computers and the loggers.

The most likely candidate for failure is our so-called "magic cable" that connects the Campbell CR10 data loggers to a standard serial line, doing the CMOS-to-RS232 translation with a couple of chips built into the cable connector. We try three sets of cables: all fail. But why did they work last night? Talking with ourselves... talking with shore folks. The only thing we can figure is that there's a slow pressure-dependent effect in the chips on those cables. Weird.

We send a cable to a boat at the surface, where it's zipped back to the shore base. Up there, it works fine. Hmmm... So what do we do about it? We concoct a scheme to embed a cable in epoxy as one possible way of pressure-protecting it. We also think about sending the data-logger case to the surface with a non-habitat diver, having it opened there (at surface pressure) and brought back down to us. Down here, we'd use it without ever opening the case, thus preserving the surface pressure. On the surface, they get ready to bring us new computers and try the logger-pressure approach. Meanwhile, we keep probing down here. It becomes apparent that at least one computer's serial port just isn't acting properly at all: it flashes its control lines when accessed, but doesn't seem to transmit or receive data. The other computer's serial port talks to itself just fine (with a loopback), but we still get no communications with the data loggers.

Then a suggestion comes from the surface: drop back out of Windows 98 and try running out of straight DOS. Oh my. It works. It works. It works! Thank you very much Windows for semi-hijacking the serial port and never giving us any indication. An excellent use of the better part of a day. But we're back on line...

Our heart rates return to somewhat normal, and we begin swiftly assembling cases and gear: we're now (at about 3:00 PM) where we should have been by 8:00 AM. Get out in the water, get some data. Amazingly, after only about one false start (uhmmm, I forgot to plug in the "magic" serial cable on the data logger), the sensor packages work and we get a flowspeed sensor signal.

The rest of the day is smooth - small problems encountered and solved. Video system runs and we get good pilot tape of plaster models oozing fluorescein under blue LED lights at night. By the end of the day... time for bed.

Lying down, of course, we're right next to the steel hull of the Aquarius habitat. It's night, and so the snapping shrimp are out and about. I've certainly heard the continuous crackle of snapping shrimp in the water at night, but never before from above my head.

Day 3

Day 3 begins for us at 7:00 AM, with intentions of getting one team in the water by 9:00. Amazingly, it happens. We're beginning to get into a data collection routine here. Today's activities (it's now night at the end of Day 3 as I write this part) went surprisingly smoothly. We retrieved yesterday's plaster dissolution models first thing, did flow measurements at their sites, got flow profiles done during the day, all as planned.

We did good runs with chilled brass models (using the rate of heating as a measure of the rate of "exchange" with the water at the models' surfaces). Basically, those runs boil down to swimming out with a Thermos bottle full of icewater (and a brass model), placing the container upside down on the seafloor (the model is bolted to the inside of the Thermos lid), and lifting off the Thermos container. That immediately exposes the model to ambient water temperature (and flow conditions). Using thermocouples in the models, we have instruments logging the temperature change as the models exchange heat with the water. It takes about 30 seconds for most of the change to happen; we give each model 3 minutes.

Though it's easy to say we're happy these observations are working, there's more than simple satisfaction. We've never attempted these observations in anything like these conditions. Most of the equipment, while tested in isolation, has never been put to field use before. It's really quite amazing (to me, anyway) that things work as well as they do. Barring the Day 2 equipment flail...

But that is, quite literally, why we're here: to do things that we have never before done (and nor has anyone else). Aquarius gives us access to the reef environment in a way not achievable by boat. Our instruments are plugged right through the habitat wall into our computers. We swim right in and out of the habitat entryway to our site, about 100 feet from the door (nice commute, eh?). When there's a problem, we don't scrap a dive, rethink our profile, and try to salvage something from the day. Instead, we just rendezvous back at the habitat, chat it over, and head right back on out. Amazing. Just as we'd hoped, the immediate access to the site, combined with extended bottom time, is letting us take the time with each experimental setup to get it right. We can get it to the point that data are flowing well, rather than thrashing out our 72.5 minute dive to get something resembling a recorded data point.

By now, the late evening of Day 3, we're beginning to figure out what it's like to live and work down here. It's rewarding, and intense. We are in close contact, both between ourselves (the four-person science crew) and the two-person NURC staffers. Delightfully, we're getting along well. I shudder to imagine what it could be like with interpersonal weirdness down here. But, to be fair, when Chris and I were looking for two additional divers for our mission, we very deliberately looked for people who could not only do the job, but with whom we could survive for 10 days in a can. The choices are paying off well. We're all working hard, while taking the occasional few seconds to simply look around and be amazed at being here.

Food is... fine. Not excellent, but fine. The "larger" meals are freeze-dried camping food. Quite edible, generally, if a touch strange. These are supplemented with a cartload of random foodstuffs that we picked up shortly before coming down here. Pop tarts, chocolate, granola bars, tortillas and cheese, and coffee. Coffeeee! Being a bit, well, funny about coffee, I managed to bring down three pounds of "SLO Morning Fog Lifter" coffee (my home favorite). It's a wonderful thing to come in from a dive and have an excellent cup of coffee. We've got an instant-hot potable water tap that makes drip coffee in a matter of seconds. No cappuccino, but... Ahhhh...

Speaking of seconds of time... one thing that does amaze me is how much time we spend on the most mundane of dive activities: getting de-geared at the end of a dive. Just to orient things: the habitat has essentially four rooms: a 6-bunk (3 x 3) sleeping room, the main eating, sitting, and habitat operations room, the entry lock (a couple of sinks for science work), and the wet porch. Starting in the wet porch, you come up in a pool of water, then up a metal platform to the habitat floor. This room is damp and warm, with a grating floor that drips through to the water's surface below. A pair of steel doors separates it from the entrance lock. These doors are hydraulic, and one of them is kept nearly closed except to pass through. That is to (mostly) separate the warm, humid air of the wet porch from the dryer, cooler air of the interior. The entrance lock is mostly for science use. We've got some electronics there (flow probe calibration setup), as well as a sink for making fluorescein tracer dye solutions. Another double steel door separates this from the main chamber (but these doors remain open for now). The main chamber is separated from the sleeping chamber by only a curtain.

Now, back to my discursion on getting off dive gear. Salt-watery people are allowed only on the wet porch - no salt water may pass into the habitat (we're living there, after all). So coming up after a dive entails, first, getting out of the twin steel 112-cubic-foot tanks we wear. (If you dive, you can guess what those weigh; if you don't, I'll just say that the rather strapping NURC staff on shore generally use a hand-truck or two people to move them about). They get refilled from a high pressure hose right at hand - that's about 10 minutes. Then one climbs up and strips off a full wetsuit. Then a quick freshwater shower, and towel off. Then dress for the dry zone. Walk on in. This is 20 minutes or so (at least) from "surfacing" in the wet porch, to getting into the main lock. It's got to be done - it would be awful to be living in a damp, salty can. But is sure sucks up a chunk of your day.

So, where do we stand, at the end of Day 3? It's now 22 minutes into Day 4, actually. I'm tired, but very, very satisfied with the progress we're making. We're here, in this amazing place, getting data that we couldn't get otherwise. Not because there are exotic instruments provided here, but because we can use simple instruments where we normally couldn't work.

We arise and peer out of windows to see fish peering in at us. Our bedroom window has a few fantastic bryozoans growing on it (transparent, from our point of view). Three of us excitedly followed a school of four 2-foot long permit jacks from a front porthole, around the back past our bedroom porthole, and around to our dining room window. We're getting used to the single huge tarpon that regularly patrols in a loose circle around the habitat.

It's funny how the four of us see different things, out on the reef. I mentioned that I hadn't noticed any moray eels, to which Ken replied that he's seen five or seven in our immediate work zone. I'm so focussed on our observations that I haven't spent enough consciousness to just see what there is to see. A touch of performance anxiety.

We'd all heard consistent stories of teams coming down, getting narced, and spending uncounted time giggling, or staring off into space. People assured us that we wouldn't manage to get any data properly collected without detailed, foolproof datasheets with prominent blanks for each number. This doesn't seem to be our experience. Are we just such dull people that we don't even have the breadth to get silly? Or are we just pretty focussed on what we'd like to get done? Besides a slowly accumulating fatigue (that happens when you work hard and sleep less than normal), I don't feel physically or mentally deficient. No more so than usual, I mean. Things are busy, we address problems, strategize, and work on. In that sense, It's like any other field project. But in a setting like no other field project.

We'll see how things progress in the next few days. We've only just started, and we've also been assured that it's the first few days that are productive, and then things start to go downhill...