Ecosystem Services, Natural Capital, and Biodiversity–Ecosystem Functioning

Ecosystem services, natural capital, and biodiversity–ecosystem functioning (BEF) unite experimental community and ecosystem ecology with environmental economics to quantify how biological diversity regulates biogeochemical fluxes, biomass production, and ecological stability, and how these biophysical processes generate the material, regulatory, and cultural flows that sustain human civilization. Brought to global prominence in 1997 by Gretchen C. Daily's edited volume Nature's Services and Robert Costanza and colleagues' global monetary valuation in Nature, and institutionalized by the 2005 United Nations Millennium Ecosystem Assessment (MEA), ecosystem services are defined as the direct and indirect contributions of ecosystems to human well-being.

In ecological-economic accounting—synthesized in Sir Partha Dasgupta's 2021 Economics of Biodiversity: The Dasgupta Review—natural capital represents the planetary stock of living biota, soils, freshwater aquifers, mineral resources, and atmospheric reservoirs (analogous to physical produced capital and human capital in an economy's inclusive wealth), whereas ecosystem services represent the recurring interest or dividend flows generated by those intact natural capital stocks. Underpinning the reliable delivery of ecosystem services is the experimental science of Biodiversity–Ecosystem Functioning (BEF), which reverses the traditional paradigm of ecology: rather than asking how abiotic environments determine species diversity, BEF tests how species, functional, and phylogenetic diversity act as causal drivers of ecosystem productivity, nutrient retention, and temporal resilience.

Supporting, Provisioning, Regulating, and Cultural Services

The 2005 Millennium Ecosystem Assessment (MEA)—authored by over 1,360 scientists worldwide—classified ecosystem services into four functional categories, finding that ~60% (15 of 24) of assessed global services were being degraded or used unsustainably:

1. Supporting Services: The foundational ecological processes—including oxygenic photosynthesis and primary production, pedogenesis (soil formation), nutrient cycling (N, P, S, C), and the hydrological cycle—that are necessary for the production of all other ecosystem services, operating over long timescales.
2. Provisioning Services: The tangible material goods extracted directly from ecosystems, including agricultural crops, livestock, wild-capture marine and freshwater fisheries, timber and wood fiber, biomass fuels, clean freshwater, and bioprospecting genetic and pharmaceutical resources (such as paclitaxel from Pacific yew Taxus brevifolia, vincristine from Madagascar periwinkle Catharanthus roseus, artemisinin from Artemisia annua, and thermostable Taq DNA polymerase from Thermus aquaticus).
3. Regulating Services: The benefits obtained from the biophysical regulation of ecosystem processes, including climate regulation via terrestrial and marine carbon sinks (~55% of anthropogenic CO₂ emissions absorbed annually by forests, soils, and oceans), coastal and flood attenuation by mangrove forests (Rhizophora, Avicennia), salt marshes, coral reefs, and riparian wetlands (which dissipate storm-surge wave energy), water purification and denitrification by riparian buffer strips, biological pest control by arthropod predators and parasitoids, and animal-mediated crop pollination (by wild bees, syrphid flies, butterflies, bats, and birds, which enhance yield or fruit quality in ~75% of leading global food crops, valued at $235–577 billion yr⁻¹ by IPBES).
4. Cultural Services: The non-material benefits people obtain from ecosystems through cognitive development, scientific discovery, outdoor recreation and ecotourism, aesthetic inspiration, and spiritual or Indigenous cultural identity.

In 2018, the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) complemented the MEA framework with Nature's Contributions to People (NCP) (Díaz et al., 2018), grouping 18 categories into Material, Non-material, and Regulating contributions while explicitly incorporating relational values and Indigenous and Local Knowledge (ILK) alongside instrumental economic metrics.

Cedar Creek, Jena, and Loreau & Hector's Additive Partitioning

To test whether biodiversity loss impairs ecosystem functioning, ecologists in the 1990s initiated large-scale manipulative experiments in which plant species richness (S) and functional group richness (e.g., C₃ grasses, C₄ grasses, N₂-fixing legumes, and deep-rooted forbs) were randomly assembled across hundreds of replicated plots—most notably David Tilman's Cedar Creek Biodiversity Experiment in Minnesota (initiated 1994, 168 plots of 1 to 16 perennial savannah species), the pan-European BIODEPTH project (Hector et al., 1999), the Jena Experiment in Germany (2002), and global TreeDivNet forest experiments. Across grassland, forest, and aquatic syntheses (Cardinale et al., 2012), standing biomass and resource capture increase with species richness according to a saturating power-law or Michaelis–Menten curve (Y = a · S^b, with mean scaling exponent b ≈ 0.26 in short-term plots and rising to b ≈ 0.40–0.50 over multi-decade timescales as soil feedbacks develop; Reich et al., 2012), frequently achieving transgressive overyielding (where a diverse polyculture produces more biomass than the single highest-yielding monoculture).

In 2001, Michel Loreau and Andy Hector derived an exact mathematical identity—based on the Price equation—to partition the net biodiversity effect (Δ*Y*), defined as the difference between observed total polyculture yield (Y_O = ∑ Y_O,i) and expected yield based on component monocultures (Y_E = ∑ RY_E,i M_i), into two additive mechanisms:

Δ*Y* = Y_O − Y_E = N · Δ*RY* · M̄ + N · Cov(Δ*RY*, M)

where N is the number of species in the mixture, M_i is the yield of species i in monoculture, RY_O,i = Y_O,i / M_i is the observed relative yield of species i in mixture, RY_E,i is its seeded proportion (1/N), and Δ*RY*_i = RY_O,i − RY_E,i:

1. The Complementarity Effect (N · Δ*RY* · M̄): Positive when species in mixture on average perform better than expected from their monocultures (Δ*RY* > 0). Driven by niche partitioning in space and time (e.g., deep-rooted lupines and forbs extracting deep soil water and nitrate below shallow fibrous grass roots, or phenological separation between cool-season C₃ and warm-season C₄ grasses), inter-specific facilitation (legumes fixing atmospheric N₂ that enriches soil N for neighboring C₄ grasses), and dilution of host-specialist soilborne pathogens.
2. The Selection Effect (N · Cov(Δ*RY*, M)): Positive when species with high monoculture yields (M_i) competitively dominate the mixture (positive covariance between monoculture yield and relative yield gain Δ*RY*_i), analogous to a sampling probability effect. Long-term data from Cedar Creek and Jena show that while the selection effect is prominent in year 1–2, the complementarity effect grows progressively stronger over decades as niche differentiation and soil organic nitrogen/carbon accumulation develop.

The Insurance Hypothesis and the Portfolio Effect

Beyond raising mean biomass (μ), biodiversity stabilizes total community production over time against droughts, heatwaves, and pest outbreaks—quantified by temporal stability (S_temp = μ_tot / σ_tot), the inverse of the coefficient of variation (Tilman et al., 2006). Formalized by Shigeo Yachi and Michel Loreau (1999) as the Insurance Hypothesis and by Daniel Doak and David Tilman as the statistical portfolio effect, ecosystem stability increases with species richness because different species exhibit asynchronous responses (negative or zero covariance, Cov(Y_i, Y_j) ≤ 0) to environmental fluctuations (response diversity, Elmqvist et al., 2003). During Cedar Creek's severe 1988 drought, for example, drought-sensitive C₃ grasses declined sharply, but drought-tolerant deep-rooted C₄ grasses and forbs increased via compensatory dynamics, buffering total community biomass in species-rich plots while species-poor plots collapsed. Moreover, Hector and Bagchi (2007) demonstrated that maintaining ecosystem multifunctionality (simultaneously sustaining carbon storage, nitrogen retention, decomposition, pollination, and drought resistance across multiple years and locations) requires far higher species richness than maximizing any single function in a single year.

Inclusive Wealth, SEEA, and Real-World PES Programs

Because conventional Gross Domestic Product (GDP) measures market output flow while ignoring the depreciation of natural capital stocks (recording clear-cutting a primary rainforest purely as positive timber revenue without deducting the lost flood regulation, carbon storage, and soil fertility), the United Nations System of Environmental-Economic Accounting (SEEA), the Natural Capital Project (InVEST spatial modeling suite developed by Gretchen Daily, Stephen Polasky, and Kareiva), and Partha Dasgupta's 2021 Review advocate measuring national economic progress through Inclusive Wealth—the social shadow value of Produced Capital + Human Capital + Natural Capital.

To internalize positive ecological externalities into economic decision-making, governments and beneficiaries implement Payments for Ecosystem Services (PES)—voluntary, conditional transactions in which a service buyer pays a landholder or steward to maintain a defined land management practice that secures an ecosystem service (Wunder, 2005):

• New York City Catskills–Delaware Watershed Protection (1997 Memorandum of Agreement): Facing an EPA Safe Drinking Water Act mandate to build a mechanical filtration plant costing $6 billion to $8 billion in capital construction plus $300–500 million yr⁻¹ in operating costs to treat drinking water for 9 million residents, New York City instead invested ~$1.5 billion in natural capital restoration across the 4,100-km² Catskills and Delaware watersheds—purchasing conservation easements, upgrading dairy farm manure management and septic systems, and restoring riparian forest buffers. Intact watershed soils and microbes purified the water below regulatory thresholds at less than one-fourth the cost of engineered gray infrastructure (Chichilnisky & Heal, 1998).
• Costa Rica's National Pagos por Servicios Ambientales (PSA) Program (1996 Forestry Law 7575): Administered by FONAFIFO and funded by a dedicated 3.5% national fossil-fuel tax alongside hydroelectric utilities and international conservation agreements, Costa Rica's PSA program pays private landowners ~$60–$300 ha⁻¹ yr⁻¹ to conserve primary forest, allow natural regeneration, or establish agroforestry for four bundled services: carbon sequestration, hydrological regulation, biodiversity conservation, and scenic beauty—helping reverse national forest cover from a low of ~21–26% in 1987 to over 52–59% today.
• REDD+ and Blue Carbon Markets: Under the UN Framework Convention on Climate Change (UNFCCC), REDD+ (Reducing Emissions from Deforestation and Forest Degradation in developing countries) and coastal Blue Carbon initiatives (conserving mangroves, tidal salt marshes, and Posidonia/Zostera seagrass meadows, which bury organic carbon in anoxic sediments at rates up to 10-fold higher per hectare than terrestrial forests) channel climate finance toward ecosystem conservation.

What is the difference between the Complementarity Effect and the Selection Effect in BEF experiments?

Derived by Michel Loreau and Andy Hector (2001), the Selection Effect occurs when a diverse plant mixture achieves high productivity simply because it has a higher statistical probability of containing a single highly productive species (measured in monoculture) that competitively dominates the plot. The Complementarity Effect occurs when multiple species yield more in mixture than expected from their monocultures due to niche partitioning (e.g., differing rooting depths or phenologies), facilitation (such as legumes fixing nitrogen for grasses), and reduced host-specific pathogen buildup.

How does the Insurance Hypothesis explain why diverse ecosystems are more stable during droughts?

Formulated by Shigeo Yachi and Michel Loreau (1999), the Insurance Hypothesis states that communities containing multiple species with different physiological tolerances (response diversity) buffer total ecosystem functioning against environmental extremes. When a drought or pest outbreak suppresses sensitive species, drought- or pest-resistant species within the same functional guild expand via compensatory growth (and statistical portfolio averaging), stabilizing total community biomass over time (S_temp = μ/σ).

How did New York City use natural capital to avoid building an $8 billion water filtration plant?

In the 1990s, fertilizer and sewage runoff in the Catskills–Delaware watershed threatened the unfiltered drinking water supply of 9 million New York City residents. Rather than spending $6–8 billion to construct an industrial filtration plant (plus $300 million annually to operate it), NYC invested ~$1.5 billion to purchase watershed conservation easements, restore riparian forest buffers, and upgrade agricultural waste systems, restoring the natural water-purification service of the watershed ecosystem.

1. Costanza, R., d'Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., et al. (1997). The value of the world's ecosystem services and natural capital. Nature, 387(6630), 253–260. doi:10.1038/387253a0
2. Loreau, M., & Hector, A. (2001). Partitioning selection and complementarity in biodiversity experiments. Nature, 412(6842), 72–76. doi:10.1038/35083573
3. Tilman, D., Reich, P. B., & Knops, J. M. H. (2006). Biodiversity and ecosystem stability in a decade-long grassland experiment. Nature, 441(7093), 629–632. doi:10.1038/nature04742
4. Cardinale, B. J., Duffy, J. E., Gonzalez, A., Hooper, D. U., Perrings, C., Venail, P., et al. (2012). Biodiversity loss and its impact on humanity. Nature, 486(7401), 59–67. doi:10.1038/nature11148