Genome banking is a civilizational insurance policy, not a substitute for habitat protection. The minimum strategic goal is to preserve DNA, viable cells, gametes, embryos, seeds, and associated ecological metadata for the largest feasible fraction of Earth’s biota before extinction cascades erase the raw material of future restoration.
Executive assessment
- The latest IUCN Red List update reports 175,909 assessed species, of which 49,505 are threatened with extinction[1].
- That means roughly 28% of assessed species are currently at risk, and the true number is higher because many species remain unassessed[1][2].
- A species archive must therefore be built for scale, not rarity: the survival target is not a few flagship mammals, but tens of thousands of threatened lineages across plants, fungi, invertebrates, freshwater species, corals, and vertebrates[1][2].
How many species are at risk
- Current IUCN figures put 49,505 species in the threatened categories: Vulnerable, Endangered, or Critically Endangered[1][2].
- The IUCN also states that 44% of reef-building corals, 41% of amphibians, 38% of trees, 37% of sharks and rays, 34% of conifers, 26% of mammals, 26% of freshwater fishes, and 12% of birds are threatened[2].
- A 2026 IUCN-linked update also highlighted severe pressure on specific groups: over half of molluscs reliant on hydrothermal vents are threatened by deep-sea mining[1].
Storage media for genome banking: durability and limits
### Silicon chips
- Silicon-based digital storage is best for compressed genomic data, annotation, and redundancy copies, not for raw biological preservation.
- Practical lifetime depends on the full storage stack, but silicon media can be re-copied and migrated indefinitely if powered archives are maintained.
- Civilizationally, silicon is useful for information persistence, not for restoring living organisms by itself.
### Synthetic DNA storage
- Synthetic DNA is attractive because it stores enormous information density and can be read with sequencing.
- It is not yet the primary medium for live biodiversity banking because synthesis, write costs, and error correction still limit routine deployment at planetary scale.
- Its strategic value is highest for encoding reference genomes, ecological metadata, and redundancy backups.
### Crystal storage
- Crystal-based storage is a long-horizon concept for extremely dense, potentially durable archival records.
- It is still experimental relative to conservation operations and should be treated as a future archival layer, not a present biodiversity solution.
- For continuity planning, crystal media matter as a hypothetical millennial cold archive, but not as the main near-term preservation tool.
Frozen Ark Project
- The Frozen Ark Project is a global effort to preserve the DNA, tissue, and viable cells of endangered animals.
- Its purpose is to create a genetic backstop against extinction, preserving biological material for future research, conservation, and possible restoration.
- This is the correct model for a genome bank: sample now, annotate fully, store in multiple formats, distribute geographically, and refresh metadata continuously.
Seed bank vs genome bank
- A seed bank stores plant seeds, usually dried and frozen, to preserve the ability to grow whole plants later.
- A genome bank stores genetic material from any organism: DNA, tissues, cells, sperm, eggs, embryos, spores, and somatic samples.
- Seed banks preserve reproductive potential of many plants directly; genome banks preserve genetic information and, where possible, living cells for animals, fungi, and plants.
- Seed banks are already operational at scale; genome banks are broader, harder, and more technologically dependent because many species cannot be preserved as seeds or simple dormant propagules.
Minimum viable genetic diversity requirements
- A restoration archive must preserve multiple unrelated individuals per species, not a single exemplar.
- For immediate breeding rescue, a minimum starting target is typically 50–100 individuals per species when available, with broader sampling preferred to retain rare alleles and avoid inbreeding bottlenecks.
- For long-term evolutionary resilience, the target should be hundreds to thousands of genomes per species in heavily threatened taxa, especially where populations are fragmented.
- The operational rule is simple: one genome is a reference; many genomes are a population.
- For species with strong geographic structure, samples must cover distinct subpopulations, not just one park or captive line.
De-extinction feasibility
- De-extinction is technically plausible only for a narrow subset of recently extinct or near-extinct species with close living relatives, usable DNA, and a compatible surrogate reproductive system.
- It is most feasible for species lost recently enough to recover high-quality genomes and where embryos can be gestated or propagated using a related extant species.
- It is not a general solution to biodiversity loss: ecological niches disappear, behavior is not fully encoded in DNA, and restored organisms often lack the original habitat and coevolved partners.
- De-extinction should be treated as a last-resort recovery tool, not a substitute for preventing extinctions.
- For civilization continuity, the priority is genome preservation before extinction, because once a species is gone, the restoration problem becomes dramatically harder and usually incomplete.
What matters most for long-term survival
- Preserve tissue + DNA + viable cells whenever possible, not just sequence files.
- Sample many individuals, many populations, many taxa.
- Store in multiple media and multiple locations.
- Link every sample to location, date, sex, age, health, microbiome, habitat, and pedigree data.
- Treat genome banking as part of a larger survival stack: habitat protection, seed banking, cryopreservation, captive assurance populations, and ecological monitoring.
If needed, this can be converted into a Moon-ark acquisition standard with explicit sample quotas by taxon, storage tier, and prioritization rules.