“All the DNA across the world, plants and animals both, adds up to at least 53 nonillion (a number equal to 5.3 × 1031) megabase pairs. A megabase pair, equal to one million base pairs, is the unit of length for nucleic acids. That amount of DNA would weigh 50 billion tons, according to the estimates published in PLOS Biology.”
In terms of pure information, biological diversity matters. In terms of entropy, biodiversity matters even more, because each unique organism has a unique set of freedoms, so unique organisms can share less space with fewer confrontations than identical organisms. DNA can store about 700 terabytes per gram. Obviously, that’s optimised, but if you think about the amount of DNA in an old-growth forest, the quantum of data is staggering. Much of the information may be duplicate or junk, but considering that more than 99.9% of the digital information we store is duplicate and junk, the duplication of information within each organism can’t be counted against DNA.
The disappointing thing about information theory is the notion that there is more information now than ever before. In terms of raw information, almost every electron and proton now used in computers has been on earth since at least the late bombardment, and their information content hasn’t changed. The information is organized differently and some of it is really useful for predicting events and phenomena, but even in terms of organized information, it’s hard to say that the average data center contains more useful information than the topsoil it displaced. Moreover, nobody would say that the world is more predictable or stable now than it was 50 years ago, and most information rich nations are building walls to show their lack of faith in their own information to predict or control anything.
Deforestation and mono-cultivation in the remaining forests have almost certainly deleted more unique information in the DNA of eliminated organisms than will ever be created in digital technology. At some point, all of the data centres in the world might contain as many bits as were lost, but because those bits only relate to other bits of information, they could not possibly replicate the information that has been obliterated with the loss of wilderness since the advent of digital computing. It isn’t possible to compare the meaning of Spanish moss with, say, a weekend worth of duck face instagram photos, but the flood of trivial data within the digital record makes it very hard to find anything worth saving, let alone something good to eat.