The oxygen entering your lungs was made by life. Marine photosynthesizers account for roughly half of Earth's gross oxygen production each year, and the cyanobacterium Prochlorococcus is commonly estimated at up to 20% of that flow. Those are production shares, not the standing atmosphere. Most of that oxygen is respired again; the air you breathe is a geologic stock, maintained over immense time by organic-carbon burial. The point is simpler: the production system is still biological.
Land plants run the same fundamental process. Photosynthesis makes organic matter from sunlight, carbon dioxide and water. After plants spend some of that energy on respiration, what remains is net primary production, or NPP: the new plant matter available to build roots, wood, leaves, food webs and future growth.
In 2023, Richardson and colleagues reported that six of nine planetary boundaries had been transgressed. For the functional integrity of the biosphere, they proposed human appropriation of net primary production, or HANPP, as the control variable.
what the 30% measures
HANPP is not the percentage of nature humans eat. It is the sum of biomass harvest and land-use-induced change in NPP relative to potential vegetation: HANPP = harvested NPP + land-use-induced NPP change. In ecological terms, it measures how human land use changes the annual biomass energy left in terrestrial ecosystems.
Harvest includes removed and harvest-related biomass. Land-use-induced change captures the difference between potential and actual plant production caused by conversion and management. That term can even be negative where irrigation or fertilization raises production above the modeled potential baseline. The global number therefore describes an alteration of biological energy flow, not a single pile of material consumed by people.
Richardson et al. provisionally placed the boundary at HANPP equal to 10% of preindustrial Holocene mean terrestrial NPP, with the high-risk zone beginning around 20%. Against that Holocene baseline, their estimated 2020 value was approximately 30% (16.8 Gt of carbon per year). Against current potential vegetation the same appropriation is about 23.5%. They use the Holocene baseline because the extra production from carbon fertilization is a climate-resilience response, not a surplus available to harvest.
Richardson dates that Holocene-baseline transgression to the late 19th century. A risk boundary marks rising probability of destabilization. It is not an event date, and it is not a parcel underwriting threshold. Global HANPP tells an allocator the direction and scale of exposure. Place-level condition tells them what is still standing and what protection can do.
settled: life makes the production base
Photosynthetic organisms make the organic matter that supplies energy and material to most of the biosphere. Economies participate in that flow through food, fiber, timber and land use.
supported: HANPP measures the scale of that take
HANPP is a well-developed indicator of the combined effects of harvest and land-use-driven productivity change. It makes the physical scale of human demand visible against the annual production base of terrestrial life.
contested: the exact guardrail
The exact global guardrail and its translation across ecosystems remain provisional and debated. The 10% and 20% lines are proposed risk boundaries within a global framework.
Six of nine, in Richardson 2023, is context. It is not a countdown clock. The 2025 Planetary Health Check assessed seven of nine, with ocean acidification the newest transgression. The count moves. The HANPP argument does not depend on the headline tally.
what biodiversity loss actually means
Biodiversity loss is the erosion of the living processes that manufacture the conditions an economy already uses: biomass energy, soil function, water regulation, immune exposure, genetic options. It is not only a species-count headline. Richardson tracks that as two variables in one boundary. Genetic integrity (extinctions per million species-years) is estimated at more than 100 against a proposed boundary under 10. Functional integrity is HANPP. One measures the makers. The other measures how much of what they make is already spoken for.
from your body to a place
At body scale, every breath spends oxygen and every meal carries carbon first fixed by a photosynthesizer. Follow that dependence outward and it reaches places: a field whose soil community cycles nutrients, a forest whose canopy and roots regulate water, a grassland whose plants keep carbon and moisture moving through living tissue.
The maker is not an abstract category called nature. It is the particular assembly of plants, microbes, fungi, animals, water and soil making conditions in a particular place.
The title says eating because we spend the production. HANPP is the accounting of that spend — harvest plus productivity change — not a claim that people ate 30% of the plants.
An economy drawing down its biological production base is reducing the capacity on which future output depends. That transmission is supported, not guaranteed in every place: the loss may first appear as unstable yields, greater supply volatility, more treatment cost, or more erosion. Those are downstream entries from an upstream change in condition.
Appropriation is also uneven. Embodied HANPP in trade shows a spatial disconnect: high-income countries are typically net importers of the appropriation, while much of the land pressure sits elsewhere (Erb et al. 2009; Chen et al. 2023). A global 10% line would bind subsistence and agrarian economies hardest unless the reduction is assigned to high consumers.
makers can stop
Biological production is renewable only while the system retains the condition required to renew it. Remove a mature canopy, simplify a food web, compact living soil or sever water flows, and the maker may continue at lower capacity—or reorganize into a different system.
Protection and restoration are not interchangeable. Protection keeps an existing network, its accumulated structure and its current productive capacity intact. Restoration attempts to rebuild what has been impaired. Its clock is set by growth, reproduction, soil formation and ecological succession, not by a capital budget or reporting year. Soil structure and organic matter often take decades; some biocrusts take a century or more. Money can arrive quickly; biological organization cannot always be repurchased on demand.
This is why condition matters before loss, as described in the investment value of ecosystem condition. It is also why capital must be able to name the place and maker it intends to protect.
money points the other way
A procurement contract pays for tonnes delivered. A farm loan commonly underwrites cash yield and collateral. A timber sale recognizes removed volume. None automatically pays to preserve the biological production left in the place after extraction.
The result is specific misdirection: financial accounts recognize the harvest while much of the maker's continuing condition remains outside the transaction. FAO, UNDP, and UNEP, averaging 2013–2018, put global support to agricultural producers at about $540 billion a year, and found 87% of it price-distorting or harmful to nature and health. That money still pays, at scale, for volume and inputs more readily than for the condition that keeps production possible. Capital doesn't invest for nature examines that allocation problem; living money asks what changes when circulation is designed around continued life rather than one-way removal.
Who would fund this maker, and why? A food company would fund soil and plant condition in a named sourcing area to reduce input volatility; a municipality would fund an upstream forest to avoid rising treatment and flood costs; a long-duration owner would fund both to preserve productive capacity. If you hold land-dependent revenue, the practical ask is condition and appropriation intensity on the sourcing landscape — not yield alone.
Some dimensions of nature's value are infrastructural; the category stays selective; it does not claim the whole. The relevant mechanism is continued biological production in a defined place.
a selective capital filter
Criticality is not the same as worth. A living system does not need to pass an investment screen to deserve existence. The screen only identifies where protective capital has an unusually strong risk case. The test is older than our phrasing: critical natural capital (Ekins et al. 2003), the question of whether natural capital is really substitutable (Cohen, Hepburn & Teytelboym 2019), and cascading regime shifts (Rocha et al. 2018). Failure that is non-substitutable, irreversible, and cascading is the load-bearing case.
A living system does not become critical because an outside analyst says so. That claim has to be co-constructed with the science and with the Indigenous nations, local communities, and rights-holders of the place.
This filter prevents every green surface from being relabeled as essential infrastructure. The natural infrastructure consensus trap explains why selectivity makes the case stronger.
condition before instruments
Ensurance applies this logic at the scale where protection can be specified: a place. Establish the maker and mechanism, assess condition, identify the parties exposed to its decline, then structure capital for protection and—where necessary—restoration.
RealValue is the natural capital accounting engine used to make current ecological condition and flows legible to capital. It is a bridge between evidence and allocation, not a statement of the living system's total worth. Finance can be based on measured condition and present flows without pretending that price contains every ecological, cultural or intrinsic value.
Once that relationship is legible, a specific ensurance certificate — a claim that funds a named natural asset — can direct capital under defined terms. The instrument follows the place; it does not make the place real.
This is a place-level financing approach, not a planetary-boundary solution by itself. Global boundaries indicate systemic risk. Place-level evidence determines what is present, what makes it, what threatens it and what protection can realistically do.
read next
Continue with dirt is a health system.
Or explore how ecological stocks, flows and condition connect in natural capital.