Census and Detectability
You Can Count on It
When a lightning bolt strikes the ground, a hundred million volts and 30,000-degree temperatures vaporize minerals to create a fulgurite, a hollow tube with glassy walls of fused silica and minerals turned from red to black. I acquired my fulgurite in the Namib Desert of southwestern Africa. Because they are the spawn of lightning, the abundance of fulgurites should parallel the frequency of cloud-to-ground lightning and should be predictable from the lightning density (strokes per square kilometer), which varies enormously around the globe and with season, as NASA satellites have revealed. About 10 to 50% of flashes are cloud-to-ground, capable of making a fulgurite, and this frequency is higher in the tropics and at higher elevation. Florida has the highest lightning density in North America with about a hundred strikes per square kilometer per year, and indeed, some spectacular, meters-long fulgurites have been found in Florida’s sand.
But Florida’s lightning density is pitiful compared to the more than 250 bolts per day for 300 days per year, or 250 strikes per square kilometer per year around Lake Maracaibo in Venezuela. The eastern Congo of Africa is another lightning capital with 150 zaps per square kilometer per year, so you might expect that the Maracaibo and Congo areas are simply paved with fulgurites. Even if they weather rapidly (die), the fulgurite population density should be high, determined as it is by birth rate and longevity, much as is true for animal or plant populations.
By contrast, the Namib Desert, the home of my fulgurite, experiences less than one strike per km2 per year (see map below). Of course, in the Namib Desert, although lightning is very rare, there are no trees, so what is there to strike but the ground (score a point for the Namib)? Still, not exactly a fulgurite factory, and even if they weather very slowly, the fulgurite population is probably very low, and your chances of finding one in the red dunes of the Namib Desert would be pretty slim, even compared to Florida, let alone eastern Congo or Lake Maracaibo. However, I am pretty sure that mine was not The Lone Fulgurite of Namibia. Overall, the much higher rate of formation in the tropics should overwhelm the longer lifespan in the Namib, and we should expect a lot more fulgurites in the two tropical sites.
To test our theory, we would have to census fulgurite populations in both locations, but now reality clashes with theory because this census depends very much upon being able to find them, that is, on detectability, and this is vastly different for the Namib and the tropics. Imagine yourself under the dripping canopy of a Congolese or Venezuelan rain forest, soaking wet, monkeys screaming at you and throwing feces from the treetops while you search a litter-covered slope of slippery clay. The trees are dense, the light is dim, and rotting logs lie here and there, ready to trip you. Even if fulgurites are common, your chances of spotting one are mighty slim (score negative points for the tropics).
Now picture yourself strolling across the massive red dunes and white plains of the Namib Desert, the sun blazing down, a good westerly breeze stirring the dune crests and scouring the plains. Fulgurites, though probably very rare, would readily appear on the surface as the wind strips away the sand (positive points for the Namib), and their black color against the red dune sands would make spotting them easy (more points). Formation may be very rare in the Namib, but lifespan and detectability are very high. So perhaps it is not as surprising as it first seemed that my fulgurite’s home was in the lightning-deprived Namib Desert.
When we try to estimate abundance of any number of things, we often face the reality of detectability. For example, my efforts to estimate the population of colonies of the tiny ant Pheidole adrianoi has much in common with censusing fulgurite populations (yes, really!), because the census results in both are confounded with detectability. The worker ants of my chosen subject are so small they look like walking apostrophes (‘), and are difficult to detect even againt the the white sands of the forest where they live. Their nests too are devilishly difficult to spot, often marked by a mere smudge of yellow sand on white sand. When the soil is covered with a granular organic litter, the tiny black workers and their nests are close to undetectable. We are smack up against the problem of census and detectability, just as we were with fulgurites (though on a much smaller scale).
When I claimed that the fulgurite population around Lake Maracaibo ought to be dense on the ground, but rare as hens’ teeth in Namibia, proving my claim was hampered by the vastly different detectability in these regions. Similarly, if I claimed that my ant mostly preferred bare white soil and avoided black littered soil, this claim too would be hampered by the different detectability in the two types of surfaces. In both cases, my ability to know depends on my ability to detect.
Detectability, aside from adding uncertainty to my surveys, determines whether a bird eats an insect, a pig finds the truffle, a soldier navigates a mine field, a political survey reflects reality, a bat snags a moth, or a satellite finds the enemy tanks. Our physical world is painted with the admittedly limited palette of our senses, but detectability is not simply about our senses, it is about contingencies--- black ants on black backgrounds or white sand, black fulgurites on dry red sand or muddy, wet clay, a gaudy, red flower in a sea of green, a moth that resembles the dried leaves among which it rests. Detection is the first step in estimating abundance--- you can count on it!
Politics polls are notoriously inaccurate for some of same reasons you mention ie detection. As a former political reporter, I became aware of this a long time ago. Getting a true sample of voters who will go to the polls is difficult, and what is reported re polls often is inaccurate because people often don’t want to participate or are not reachable by cell phones, etc. Too much attention is paid to flawed polls vs issues pre-election. Detection of the popular vote ahead of an election is hard to accomplish.