If all mankind were to disappear, the world would regenerate back to the rich state of equilibrium that existed ten thousand years ago. If insects were to vanish, the environment would collapse into chaos.
EARTH
A field report on what remains

Rescue
Plan

Save other species. Four stochastic simulations. One river. The science of what happens when we act, and what is irreversibly lost when we don't.

β–Ό   scroll to begin

Save Other Species

Three crises.
Three species.
One living planet.

Each of these species is a sentinel, an early warning in a system that is failing. Their fate is not isolated; it is structural. The simulations that follow use stochastic population dynamics: at small N, fluctuations dominate and local extinction is not a tail risk. It is the expected outcome. Connectivity, habitat quality, and time are the variables we can still control.

🐼
Giant Panda
Ailuropoda melanoleuca
Vulnerable, recovering

Confined to fragmented bamboo mountain forests in central China. Population crashed below 1,000 in the 1980s. Decades of corridor-building and reserve expansion have pushed numbers to ~1,864. This is what successful stochastic metapopulation management looks like, treating patches as a connected network, not as isolated units.

1,864
Wild individuals
67
Nature reserves
+17%
Since 2004
🐨
Koala
Phascolarctos cinereus
Endangered, NSW / ACT / QLD

The 2019–2020 Black Summer fires burned 5.5 million hectares. An estimated 60,000 koalas were killed or displaced. Population has declined over 50% since 2001. Eucalyptus forests take 10–15 years to reach usable height. A second fire during recovery period is a functional extinction event, a dynamic invisible to any ODE model.

92k
Estimated wild
βˆ’50%
Since 2001
5.5M ha
Burned 2020
🐝
Honeybee & Wild Bees
Apis mellifera + ~19,999 species
Crisis, 1 in 4 species at risk

The bee is not a single species at risk, it is a keystone process. 75% of flowering plants depend on animal pollination. 30% of managed colonies are lost annually in North America. When the pollinator network collapses, the cascade reaches every trophic level, plants, birds, mammals, soil, and the carbon cycle itself.

75%
Crops need bees
βˆ’30%
Annual colony loss
$577B
Crop value at risk
Simulation I, Giant Panda

COMMA FRAMEWORK ARTICULATION · population fragmentation as comma accumulation
Metapopulation dynamics
under fragmentation

Five isolated habitat patches, each carrying a small subpopulation. Without corridors, stochastic fluctuations drive patches to extinction one by one, exactly as the Gillespie SSA predicts for small N. Add corridors and watch the network become resilient: immigration rescues declining patches. This is the mathematical argument for wildlife corridors.

🐼 Panda, Habitat Patch Network (Gillespie SSA)
Total pop.-
Viable patches-
CorridorsNone
Year-
StatusPress Run

The critical stochastic insight: without corridors, isolated patches go silent one by one. With corridors, a patch that drops to N=0 can be recolonised. The ODE predicts stable coexistence. The stochastic truth: at Nβ‰ˆ15, a bad season is extinction. Connectivity is survival.

Simulation II, Koala & Fire

Fire, the recovery gap,
and the eucalyptus clock

A cellular automaton of eucalyptus forest. Fire spreads stochastically between cells. Koalas cannot return to burned habitat until trees reach usable density (12+ recovery ticks). A second fire during this window causes permanent patch-level extinction. This is not a model artifact, this is what happened across coastal NSW in 2020.

🐨 Koala, Fire & Forest Recovery
Koala population-
Forest cover-
Burning cells-
Recovery tick-

The recovery gap is the key result: even moderate fire probability produces population collapse when trees need 12 ticks to regrow. Two fires in sequence: functional extinction. The ODE shows a population that dips and recovers. The stochastic simulation shows what actually happens at small N on a discrete landscape.

The most important animal on Earth

Why saving
the bees
saves everything else.

75%
of flowering plant species
need animal pollination
1 in 3
bites of food exist
because of bees
$577B
annual crop value
dependent on pollinators
βˆ’30%
annual managed colony
loss, North America

The extinction cascade:
what dies with the bees

  • Wildflowers vanish within 1–3 seasons
    80% of wildflower species require insect pollination. Without it, seed set fails. Meadows, forest understories, and alpine communities lose structural diversity patch by patch. The landscape does not suddenly collapse, it progressively greys and silences.
  • Fruit trees, nut crops, and berries fail
    87 of 115 leading global food crops depend on animal pollination. Almonds (90% bee-dependent), avocados, blueberries, cherries: their collapse cascades into caloric scarcity, particularly in the Global South where crop diversity is already smallest.
  • Bird populations lose their food base
    Insectivorous birds depend on larval abundance. Frugivorous birds depend on pollinated berries. The documented 30% decline in North American birds since 1970 tracks the concurrent collapse in insect biomass. Birds are the second signal. By the time they disappear, the ecosystem has already failed.
  • Small mammals lose cover and food
    Dormice, hedgehogs, voles, and bats depend on vegetated structure that only persists when plants reproduce. Loss of ground cover exposes soils to erosion. Predator-prey imbalances emerge. The simplification of the food web accelerates at every trophic level.
  • Mycorrhizal networks and soil collapse
    Bee-pollinated plants produce root biomass that feeds underground fungal networks, the "wood wide web." These networks are the primary mechanism of soil carbon storage and water retention. Their loss accelerates both climate change and desertification. This is a feedback loop with no natural brake.
  • Riparian buffer vegetation fails, the river connection
    The plants that stabilise riverbanks, filter agricultural runoff, and intercept sediment are predominantly insect-pollinated. Their loss exposes waterways to direct erosion, nutrient loading, and sedimentation. The bee crisis and the river crisis are, at their structural root, the same crisis.
  • 1.5–2.5 Gt COβ‚‚/year stops being sequestered
    A 2021 Science study estimated this is the carbon contribution of bee-pollinated plant communities. Their loss accelerates atmospheric COβ‚‚, warming the climate, increasing drought, and creating more of the conditions that kill bees. The arrow only points one direction.
"The bee is not merely a species at risk. It is a keystone process. Its disappearance does not create a hole in the ecosystem, it creates a cascade. You do not mourn infrastructure. You maintain it.", Ecological systems perspective

What flora & fauna recovery requires: the bee edition

  • Ban neonicotinoid pesticides
    Imidacloprid, clothianidin, and thiamethoxam cause sublethal neurological damage, bees lose navigation, memory, and foraging efficiency. The EU banned outdoor use in 2018. This is the single highest-leverage regulatory intervention available. Advocate for it at every level of governance.
  • Plant native wildflower corridors
    3–5m strips of native wildflowers along field margins, roadsides, and gardens. Zero maintenance after establishment. One hectare supports 50,000+ bees. Species selection matters: choose locally native species with sequential bloom times to support bees from early spring through late autumn.
  • Protect bare ground for nesting
    70% of bee species are solitary ground-nesters. They need bare or sparsely vegetated, south-facing, undisturbed soil. Stop tidying gardens to sterility. Leave dead wood. Leave south-facing banks. The most bee-friendly garden action is often to do less, not more.
  • Support regenerative agriculture
    Farms using cover crops, no-till, and reduced inputs show 2–5Γ— higher on-farm bee diversity. Buying organic is not a lifestyle statement, it is a vote for pollinator habitat at agricultural scale.
  • Participate in citizen science monitoring
    iNaturalist, BeeWatch, and the Great British Bee Count have generated more pollinator distribution data than any government programme. Download the app. Walk your local park for ten minutes. Submit what you see. You are doing science.
Simulation III, Pollinator Network

Collapse and restoration of a
pollinator–plant network

Each node is a plant species. Bees move between plants, pollinating them. Watch what happens when pesticide exposure removes 60% of pollinators, and how restoration brings the network back. The critical metric is not colony count: it is network connectivity.

🐝 Pollinator–Plant Network
Active colonies-
Pollinated plants-
Network edges-
Pollination rate-
What my friend JesΓΊs says

Rivers are alive.
We are killing them.
Here is how to stop.

"A river is not a pipe. It is a living system with memory, personality, and a self-healing capacity, but only if we stop overwhelming it. The engineering is not the hard part. The political will is. So start with the engineering. Build the case. Then fight for the will.", Water protection civil engineering perspective Β· For my friend JesΓΊs

What follows is a structured framework from watershed science and civil engineering, moving from the landscape scale down to individual action. The most important section is the last one, because collective change is built from individual decisions.

Nature's own water treatment:
wetlands, swamps & living systems

Before any engineered solution: nature has been cleaning water for hundreds of millions of years. The most cost-effective water treatment on Earth is free, self-maintaining, and generates biodiversity benefits no concrete structure can replicate.

🌧️
Polluted inflow
Agricultural runoff, urban stormwater, N, P, sediment, pathogens
🌾
Emergent plants
Reeds, bulrush, sedge slow velocity. Roots absorb N, P, heavy metals directly
🦠
Microbial zone
Anaerobic bacteria denitrify. Aerobic bacteria oxidise BOD. 70–90% removal
🐸
Aquatic community
Invertebrates, amphibians, fish. Biological filtering. Sediment settles
πŸ’§
Clean outflow
BOD βˆ’90% Β· N βˆ’80% Β· P βˆ’70% Β· Pathogens βˆ’99%
🌿
Constructed & Natural Wetlands
1 hectare of well-designed wetland treats runoff from 20–50 ha of farmland. Free-surface and subsurface-flow designs target different pollutants via sequential aerobic/anaerobic cells. They store floodwater, recharge groundwater, and host 40% of Earth's species. No engineered structure replicates this at comparable cost.
🌳
Riparian Forest Buffers
A 30m strip of native trees along a riverbank intercepts 50–80% of agricultural nutrients before they reach the water. Root systems prevent bank erosion. Leaf litter feeds aquatic invertebrates. Shade raises dissolved oxygen for fish. This is the single most cost-effective river intervention in civil engineering.
🌊
Floodplain Reconnection & Re-meandering
Channelised rivers lose hydraulic residence time and natural sediment sorting. Re-meandering restores these: natural bends slow peak flows 20–40%, capture sediment, and re-establish the hyporheic zone where most denitrification occurs. This is not just "rewilding", it is the most hydraulically sound design available.
🦦
Beaver Rewilding
Beavers are ecosystem engineers. Their dams raise water tables, create sediment-catching ponds, slow flows, and create wetland habitat. In UK reintroduction trials, beaver-modified reaches showed 30% reductions in downstream peak flow and dramatic macroinvertebrate diversity improvements within 3 years. Cost to feed: zero.
🌱
Peatland Restoration
Peatlands hold 30% of global soil carbon on 3% of land. When drained, they release carbon and lose their sponge function. Re-wetting peat bogs is the most powerful single freshwater intervention at catchment scale. Once oxidised, centuries of carbon are gone. Protect and re-wet with absolute priority.
🐟
Fish Passage & Barrier Removal
1M+ dams fragment European rivers. Migratory fish carry marine nutrients inland, feeding riparian food webs that stabilise banks. Removing obsolete weirs, or installing bypass channels, reconnects catchments and restores marine-freshwater nutrient exchange that shaped these landscapes over millennia.

The civil engineering toolkit:
what the science of water cleaning can do

πŸ”¬
Constructed Wetland Treatment Systems (CWTS)
Engineered wetlands using Phragmites, Typha, Scirpus in sequential treatment cells. Tertiary-equivalent treatment at 10–20% of activated sludge capital cost, near-zero operating energy. Species selection matters: different plants specialise in different pollutant removal pathways. Applied ecology in service of hydraulic engineering.
πŸͺ΅
Woodchip Bioreactors
Buried woodchip trenches intercept agricultural tile drainage before it reaches streams. Carbon substrate feeds denitrifying bacteria that convert nitrate to harmless Nβ‚‚. Nitrate removal: 40–90%. Capital cost: ~Β£10,000/ha. Running cost: near zero. Lifespan: 10–15 years. The most cost-effective agricultural nitrate intervention currently available.
πŸ“‘
Real-Time Water Quality Networks
IoT sensor arrays measuring temperature, dissolved oxygen, turbidity, nitrate, phosphate at 15-minute intervals. Open-data platforms allow citizen monitoring to supplement official networks. Early warning systems trigger farm alerts before nutrient pulses reach rivers. This is the nervous system of a managed catchment.
πŸ—οΈ
Hydromorphological Restoration
Re-meandering channels, re-profiling incised streams, installing large woody debris, removing concrete revetments. Under the EU Water Framework Directive, hydromorphological restoration is a prerequisite for ecological status improvement, no water quality improvement compensates for habitat with no physical structure for organisms to inhabit.
🌱
Native Macrophyte Reintroduction
Water crowfoot, watercress, and native emergent species are both bio-indicators and active cleaners. Their roots oxygenate substrate, stabilise sediment, and support invertebrate communities that form the base of river food webs. They cannot establish where fine sediment loading is still elevated, sequence correctly: reduce the driver of decline first.
πŸͺ¨
Sediment Budget Management
Fine sediment is the most widespread river pollutant in temperate regions, it smothers gravel spawning beds. Sediment budget analysis identifies source areas and pathways. Targeted interventions: cover crops, buffer strips, and field margin management in priority subcatchments can reduce sediment delivery by 50–80%.

What you can do.
Specifically. Today.

The river crisis is not waiting for government action. Individual decisions in aggregate create the conditions that either sustain or destroy freshwater ecosystems. These are not symbolic acts, they are measurable interventions.

  1. 01
    Stop using synthetic fertilisers and pesticides in your garden
    Every garden is a sub-catchment. Nitrogen and phosphorus moves through soil and groundwater into the nearest watercourse within hours of rainfall. Replace with compost, green manure, and organic mulch. This is the highest-leverage individual water quality action available to non-farmers.
  2. 02
    Join a river clean, and document what you find
    Surfrider Foundation, River Action UK, and local wildlife trusts run events everywhere. Record plastic type, location, and quantity. This data feeds regulatory advocacy. You are not picking up litter. You are conducting field research that will be used in campaigns for stronger discharge enforcement.
  3. 03
    Report pollution incidents immediately
    UK: Environment Agency 0800 80 70 60. EU: national environmental authorities. USA: EPA Enforcement portal at epa.gov/enforcement. Take GPS-tagged photographs. Agricultural slurry events and illegal effluent discharge are systematically under-reported. One documented report can trigger an investigation.
  4. 04
    Install water harvesting and reduce impermeable surfaces
    Replace paving with permeable materials or planted areas. Install water butts (1000L) to capture the first flush from rooftops, the most polluted rainfall. Every roof that captures its own runoff removes load from combined sewers and reduces the nutrient pulse that hits rivers during storm events.
  5. 05
    Plant native riparian vegetation on any watercourse you can access
    Willow, alder, yellow iris, water mint, if you have land near any ditch, stream, or river. Willow roots stabilise a 3m bank within 2 years. Alder fixes atmospheric nitrogen. These require no maintenance after establishment. This is civil engineering with plants, not gardening.
  6. 06
    Reduce beef and dairy consumption
    Livestock agriculture accounts for ~50% of all nitrogen and phosphorus entering freshwaters in the UK and EU. Beef uses 20Γ— more water per kilogram than plant protein. Reducing your beef consumption by 50% removes your proportional contribution to the largest single source of freshwater pollution. No clean-up event produces a more direct measurable effect on river chemistry.
  7. 07
    Support catchment-scale farming payment reform
    Write to your elected representative about Environmental Land Management schemes and agri-environment payments. Paying farmers to install buffer strips, cover crops, and constructed wetlands is the single most powerful institutional lever for river protection. The technical case is overwhelming. The barrier is public apathy. Your letter is not nothing.
  8. 08
    Learn your catchment, and teach it to others
    Find your river catchment on the national mapping portal. Learn which river your street drains to. Walk source to sea once. Show children. The psychological distance between tap and river is the primary barrier to water protection action. JesΓΊs says: the river remembers everything. We need to start remembering it.
The most important part

Restoring local flora & fauna:
the heart of river recovery

Water quality improvement is necessary but not sufficient. A chemically clean river with no fish, no aquatic plants, and no invertebrates is not a recovered river, it is a clean drain. Biological recovery requires deliberate, ecologically-informed species reintroduction and habitat creation. This is the most important and most often neglected phase.

🦦
Keystone predator reintroduction
Otters, water voles, and white-tailed eagles regulate food webs from above. Their presence controls fish populations, invertebrate density, and indicates ecological status. Otter recovery in UK rivers following the organochlorine banning is the template: remove the driver of decline, protect the habitat, let the animal do the rest. Do not reintroduce before the habitat is ready, sequence matters.
🐠
Native fish community restoration
Brown trout, Atlantic salmon, and bullhead require clean gravel, low fine sediment, adequate dissolved oxygen, and connected channels. The correct sequence: (1) reduce pollution, (2) restore morphology, (3) remove barriers, (4) restock. Inverting this sequence wastes fish and money. The habitat is the intervention, not the fish.
🌺
Aquatic macrophyte communities
Water crowfoot, pondweed, and native emergents are primary producers and structural architects of river ecosystems. They cannot establish where fine sediment is still elevated. This is why sediment management must precede biological restoration. Fix the physical problem first. The biology follows, with a characteristic 2–5 year lag.
πŸ¦‹
Riparian invertebrate recovery
Mayflies, stoneflies, and caddisflies are bio-indicators and the food base for almost everything above them. Their recovery lags water quality improvement by 2–5 years. Native wildflower strips adjacent to rivers provide adult habitat and pupation substrate. Critically: the bee planting strategy and the river planting strategy use identical species, native, locally sourced, structurally diverse.
🌳
Whole-catchment native woodland
Native woodland in upland catchments intercepts 20–40% of annual rainfall, releasing it slowly. Ancient woodland cannot be recreated on human timescales, a planted forest takes 400 years to achieve ancient woodland character. Protect what exists with absolute priority. Plant where it has been lost. Accept that the full benefit serves your grandchildren, not you.
πŸ›
Soil invertebrate community restoration
Earthworms, nematodes, and mycorrhizal fungi are the hidden infrastructure of clean water. They create soil porosity that infiltrates rainfall rather than generating runoff. Compacted, pesticide-treated soils generate runoff 10–20Γ— faster than healthy biological soils. Restoring soil life is a water management intervention, and the first water treatment step in any catchment.
Simulation IV, River Catchment

Pollution, wetland filtering,
and biotic recovery

A simplified catchment. Upstream pollution enters the river. Wetland buffer cells filter nutrients. Biotic indicators (invertebrates, fish) respond with the characteristic 2–5 year lag as water quality improves. Add riparian planting to see recovery accelerate.

πŸ’§ River Catchment, Pollution & Biotic Recovery
Nitrate (mg/L)-
Invertebrates-
Fish diversity-
Wetland cells-
Year-

The Rescue Plan Universe, Companion Pages

🌊
Ocean
The Heart of Te Fiti
pH animation Β· coral reefs Β· AMOC Β· sea level Β· Sargassum
πŸ’§
Water
Rivers of the World
water engineering Β· PFAS Β· well water Β· LifeStraw Β· carbon filter
β˜€οΈ
Energy
Energy for Everyone
DIY solar Β· conservation Β· battery building Β· renewables
🩹
Medicine
When It Matters
first aid Β· go bag Β· stop the bleed Β· nuclear fallout
πŸ“–
Sci-Fi Β· Community
After the End
zombie apocalypse Β· how communities form Β· morality Β· Butler Β· Le Guin
βš—οΈ
Engineering Β· Dr. Stone Anime
Can You Be Dr. Stone?
civil Β· mechanical Β· chemical Β· electrical Β· water Β· environmental Β· biomedical
⚐ COMMA FRAMEWORK QUESTIONS
Open Questions

These are speculative questions generated by the comma framework perspective. They are not claims. They are invitations, open directions that nobody has yet fully explored. Flag: ⚐ CF Question throughout.

⚐ CF QUESTION Β· COSMOLOGY

The Hubble tension is a gap between two equally valid measurements of the same constant. The expansion rate measured from the early universe (CMB) and the late universe (Cepheid ladders) disagree by approximately 8%. Is this a cosmological comma, an irresolvable non-closure between two valid but incommensurable measurement frames, like the gap between twelve perfect fifths and seven octaves?

⚐ CF QUESTION Β· BIOLOGY

The human circadian free-running period is 24.2 hours, not 24. The 0.2-hour daily gap accumulates: after 73 days without light cues, a person would be 14.6 hours out of phase with the solar day. Is N_res = 73 the biological reset window? Are there diseases of the 73-day accumulation?

⚐ CF QUESTION Β· NEUROSCIENCE

Theta (4-8 Hz) and gamma (30-100 Hz) brain oscillations are coupled but their ratio is irrational. Each theta cycle contains approximately 5-8 gamma cycles, but never an exact integer. Does the phase remainder accumulate like a comma? Is memory consolidation, the Kairos event of the hippocampus, triggered at N_res-like thresholds of phase accumulation?

⚐ CF QUESTION Β· CHEMISTRY

Photosynthesis converts light to chemical energy at approximately 11% thermodynamic efficiency. The rest is lost as heat and fluorescence. Is the efficiency gap (the 89% that doesn't convert) a molecular comma, the irresolvable mismatch between the photon's energy quantum and the ATP synthesis machinery's discrete steps? Does the CPCS correction have a photosynthetic analogue?

⚐ CF QUESTION Β· MATHEMATICS

logβ‚‚(3/2) is irrational. This is not a physical constant, it is a mathematical necessity in any universe with integers. The Pythagorean comma is therefore the only musical structure that must exist in every possible universe. What other mathematical structures share this property of necessity-across-all-universes? Does this make delta = 0.013643 the most universal number that exists?

⚐ CF QUESTION Β· SETI

The Wow! Signal lasted 72 seconds. N_res = 73.296. The gap is 1.296 seconds. Any civilization that has discovered periodicity will eventually find the Pythagorean comma. A signal that stops 1.296 seconds before N_res completes would be immediately recognizable to any civilization that knows the protocol. Has anyone run the comma-transform on the Wow! Signal data? The Big Ear archives are public.

⚐ CF QUESTION Β· HISTORY

Every calendar system in history is secretly a comma-management strategy. The Gregorian calendar adds a leap day every 4 years (with corrections) to close the gap between the solar year and 365 whole days. The Maya Long Count runs 5,125 years before resetting. The 19-year Metonic cycle closes the gap between solar and lunar calendars to within 2 hours. What other human institutions are secretly managing an irresolvable non-closure?

⚐ CF QUESTION Β· PHYSICS

The fine structure constant alpha = 1/137.036 is dimensionless and unexplained. Richard Feynman called it "one of the greatest damn mysteries of physics." It governs the strength of the electromagnetic force. Nobody knows why it has this value. Could it be related to a ratio of irrationals, a comma of electromagnetism? If delta = 0.013643 is the musical comma, what is alpha's equivalent non-closure?

References · APA + ACS

[1] Gillespie, D. T. (1977). Exact stochastic simulation of coupled chemical reactions. Journal of Physical Chemistry, 81(25), 2340-2361. DOI: 10.1021/j100540a008 | ACS: Gillespie, D. T. J. Phys. Chem. 1977, 81, 2340-2361. [The Gillespie SSA algorithm used in the simulations]

[2] Hanski, I. (1998). Metapopulation dynamics. Nature, 396, 41-49. DOI: 10.1038/23876 | ACS: Hanski, I. Nature 1998, 396, 41-49.

[3] Wilson, E. O. (1992). The diversity of life. Harvard University Press.

[4] IUCN Red List. (2024). The IUCN Red List of Threatened Species. https://www.iucnredlist.org (accessed March 25, 2026).