Concept: ecological analogue of WEM and HEM · cumulative stress beyond project-level YVA
WEM (Winter Endurance Monitor) tracks the electricity system's multi-day endurance capacity. Its core insight is that instantaneous power adequacy does not imply sustained endurance — a system can be balanced in a given hour while being structurally fragile over a 168-hour stress window. HEM (Hydrological Endurance Monitor) applies the same logic to water systems: a lake's current level does not reveal whether it is on a trajectory toward minimum records.
Biodiversity exhibits the same structural property. An ecosystem can appear intact in any given project-level assessment while being on a trajectory of cumulative fragmentation that will produce irreversible state change before any conventional threshold triggers a response. SYKE's May 2026 report on wind and solar power ecological sustainability identifies this gap explicitly: Finnish YVA (environmental impact assessment) processes evaluate individual projects adequately but have no mechanism for cumulative impact across the full pipeline (~30 GW simultaneously in assessment in 2026).
BEM is the instrument that would close this gap — not by replacing project-level assessment but by providing the system-level view that project-level assessment cannot generate.
WEM asks: does the electricity system have enough firm capacity to survive a 168-hour cold windless period?
HEM asks: is the lake on a trajectory toward minimum historical water levels?
BEM asks: is the regional ecosystem on a trajectory toward fragmentation thresholds beyond which ecological continuity cannot be restored?
All three instruments share the same epistemological structure: they measure accumulated stress over time, not instantaneous state. All three are diagnostically useless if applied only at the moment of crisis.
BEPP v1 (additive index) is a valid starting point but has three structural weaknesses: additivity assumes component independence; linear weighting misses ecological threshold behaviour; and the formulation is a snapshot, not a trajectory detector. BEPP v2 addresses all three.
Each raw input is transformed through a sigmoid function σ(x) = 1/(1+e⁻ˣ) before entering the stress computation. This enforces saturation behaviour — consistent with observed ecological dynamics where stress accumulates slowly until a threshold, then accelerates.
| Component | Symbol | Raw input | Data source (candidate) |
|---|---|---|---|
| Fragmentation | D_f(t) = σ(k_f · F(t)) | Loss of contiguous habitat corridors vs reference period | SYKE ILMAVERSO · Metsäkeskus · CORINE land cover |
| Species stress | D_s(t) = σ(k_s · S_bio(t)) | Indicator species population trend. Two groups: (1) Forest: liito-orava, palokärki, metso, kalasääski. (2) Waterfowl: tukkasotka, telkkä, tavi — sensitive to eutrophication and riparian habitat loss. NDVI canopy continuity proxy. | FinBIF warehouse · SYKE biodiversity monitoring · iNaturalist Finland |
| Cumulative disturbance | D_c(t) = σ(k_c · C(t)) | Effective footprint weighted by irreversibility (not raw area) | YVA database · Fingrid grid permits · Metsähallitus wind permits |
| Recovery capacity | R(t) = σ(k_r · R_cap(t)) | Protected and conservation area fraction — modulates threshold dynamics | SYKE protected area registry · METSO agreements |
The multiplicative structure means that high fragmentation combined with high species stress produces a non-linear amplification rather than a simple sum. Exponents α, β, γ, δ are calibration parameters — regional pilot data determines their values. Recovery capacity R(t) acts as a fragility modifier: low recovery capacity accelerates the approach to tipping thresholds rather than simply subtracting from stress.
Where P(t) is the rate of new ecological pressure (incoming YVA permits, land-use change); λ·R(t)·S(t) is recovery damping (resilience slows stress accumulation); and μ·S(t)² is self-reinforcing degradation — fragmentation accelerates further fragmentation. The μS² term is the mechanism that produces tipping point behaviour.
M(t) = dS/dt is the momentum signal — the BEM analogue of WEM's stress persistence (SP) component. A system with moderate S but positive M is on a deterioration trajectory; a system with high S but negative M is recovering. The distinction is operationally critical and invisible to snapshot indices.
Tipping risk — a single administrative summary if needed:
T(t) = S(t) · max(0, dS/dt) · (1 − R(t))T(t) rises rapidly as level, trend, and vulnerability converge — the characteristic signature of a system approaching nonlinear state change. This is the institutional "alarm" trigger: T(t) exceeds threshold → regional YVA review mandated.
BEPP v1 answers: how much stress is present?
BEPP v2 answers: which direction is the ecosystem moving, how fast, and how close to a state transition it cannot reverse?
BEM does not invent from nothing. Three separate research and policy traditions address related problems; BEM is the first instrument to combine them into a single repeatable, institutionally neutral operational framework.
| Tradition | Examples | Limitation BEM addresses |
|---|---|---|
| Static biodiversity monitoring | Biodiversity.fi (110+ indicators, Finland); VMI-moni (LUKE forest structure); SYKE species monitoring | Designed for long-term state description, not real-time trajectory detection or tipping point identification. Cannot respond to rapid YVA pipeline dynamics. |
| Cumulative Impact Assessment (CIA) | SCAIRM (Spatial Cumulative Impact Risk Management); CEMPRA (Cumulative Effects Model for Prioritizing Recovery Actions); Finnish applications in reindeer herding / forestry coordination | Technically sophisticated but project-specific and retrospective. Not designed as a continuously updated regional operational instrument. Requires expert input per application. |
| Nature risk financial pricing | TNFD (Taskforce on Nature-related Financial Disclosures); EU Nature Credits tiekartta July 2025; voluntary nature credit markets | Addresses the institutional incentive problem (who pays for nature) but requires a measurement layer before credits can be priced. TNFD provides the reporting framework; BEM would provide the underlying signal. |
The EU Nature Credits roadmap (July 2025) is particularly significant for BEM's institutional positioning. Nature credits are a financial instrument designed to incentivise biodiversity conservation and restoration through market mechanisms — directly analogous to carbon credits in the ETS system. For nature credits to function, the ecological state they represent must be measurable, verifiable, and dynamically tracked. BEPP v2's output vector (S, dS/dt, M, R) provides exactly the measurement substrate that a nature credit verification system requires. BEM is therefore not only a diagnostic instrument but a potential measurement layer for an emerging EU regulatory market.
Project-level YVA is designed to assess whether a specific project is permissible given its direct impacts. It is not designed to assess whether the cumulative portfolio of projects in a region is pushing the regional ecosystem toward a state change. The distinction is structural, not a failure of implementation.
Consider the wind power pipeline: 401 projects totalling 56.3 GW are in various YVA stages in Finland (January 2026). Each project's YVA can conclude that its local impact is acceptable. None of them can conclude that the aggregate impact is acceptable — because no instrument currently measures the aggregate. SYKE's report notes this gap and recommends stronger cumulative assessment. BEM would be the instrument that makes such assessment possible in operational time rather than as a post-hoc academic study.
The relationship between project-level YVA and BEM is analogous to the relationship between individual bank risk assessment and systemic financial risk monitoring. Individual banks can each be solvent while the system accumulates fragility. The 2008 crisis demonstrated that project-level adequacy does not imply system-level stability. Macroprudential regulation (stress testing, systemic risk buffers) was the institutional response.
BEM is the ecological analogue of macroprudential monitoring: not replacing project assessment but adding the system-level view that project assessment structurally cannot provide.
BEM extends the ACI endurance monitoring family. WEM, HEM, and BEM form a three-instrument suite covering energy, hydrology, and ecology respectively. All three share the same diagnostic logic: measure accumulated stress over time; identify trajectories before thresholds; make slow-moving systemic change visible before institutional response latency has already closed the correction window.
CN-013 (Spatial Value Capture) identifies ecological continuity as a topological value without institutional owner: the value exists in the physical layer (intact ecosystems, migration corridors, carbon sinks) but no market mechanism prices it and no actor is assigned responsibility for its trajectory. BEPP would constitute the signal layer — the precondition for any future institutional mechanism. You cannot allocate responsibility for a variable you do not measure.
Peruskivet (fiction, 2026) captures the pre-instrument condition in Luku 14C: "Tarvittaisiin HEM luonnolle. Joku joka tallentaa sen ennen kuin se on mennyt." BEM is the formal articulation of that requirement.
BEM at concept stage faces three prerequisites before pilot development:
The Rautalammin reitti watershed is the natural BEM pilot region: HEM is already operational there, CN-012 documents the aquatic biomass loop, SYKE ILMAVERSO III covers the lake network, and wind power YVA applications affect the surrounding forest landscape. A BEM pilot here would close the monitoring triangle: energy (WEM) · hydrology (HEM) · ecology (BEM).
The Suomen Uusiutuvat wind power project map confirms an active YVA-stage project within the Rautalammin reitti pilot area. Tervalamminvuori (Solarwind Finland Oy) is located in Rautalampi municipality, Pohjois-Savo — directly within the BEM pilot bbox. Status: permitting/studies ongoing, land use planning started.
| Field | Value |
|---|---|
| Project | Tervalamminvuori |
| Municipality | Rautalampi |
| Region | Pohjois-Savo |
| Type | Onshore wind power |
| Capacity | 30–34 MW (5 turbines) |
| Status | Permitting / studies ongoing |
| Developer | Solarwind Finland Oy |
| BEM relevance | D_c input — confirmed YVA pipeline pressure within pilot bbox |
This confirms that D_c = 0.61 (estimated from national 56.3 GW pipeline) has a concrete local driver. TURRI methodology (2027) will provide the empirical link between this permitting pressure and actual watershed ecological impact. When TURRI data becomes available, Tervalamminvuori serves as a calibration case for the D_c component in the Rautalammin reitti pilot.
A second, substantially larger concrete D_c driver has been identified within the Rautalammin reitti pilot area: Fingrid's planned Harjulinja (Murtoperä–Kauppila), a 2×400+110 kV transmission line crossing four regions. Unlike Tervalamminvuori, this is a national transmission-grid project — its entire rationale (data-centre and industrial electricity demand growth in southern Finland) is fully external to the Rautalammin reitti watershed's own ecosystem, which is exactly the structural blind spot BEM is designed to make visible: project-level EIA assesses Harjulinja's own impact, but not the fact that the same watershed simultaneously carries both the wind-power load (Tervalamminvuori) and the transmission-grid expansion load.
Methodological caveat (added 2026-07, applies to §06c and §06d alike): all findings below are read from the EIA assessment programme's map appendix (Liite 1, dated 9/2025) — this stage shows multiple parallel route alternatives (east/west options on several segments) under consideration, not the final selected corridor. The actual EIA impact statement (YVA-selostus), expected late 2026, will narrow this to a single, considerably narrower alignment within the currently-mapped wide review corridors, confirm or discard specific mitigation choices (e.g. the Mehtiönkoski avoidance routing discussed in §06c below), and may remove the route entirely from some currently-flagged sensitive points (such as the flying-squirrel habitat in §06c) if the alternative on that side is not selected. Everything below should be read as describing the current possibility space, not a confirmed outcome — this section needs revisiting once the impact statement is published.
| Field | Value |
|---|---|
| Project | Harjulinja (Murtoperä–Kauppila) |
| Developer | Fingrid Oyj |
| Type | 2×400+110 kV transmission line |
| Length (routes under review) | ~594 km total; final build ~174–255 km |
| Corridor width | ~96 m (new corridor) / ~74 m (alongside existing line) |
| EIA phase | Assessment programme reviewed here (9/2025, multiple route alternatives); impact statement with final route expected late 2026 |
| Construction | Estimated 2030–2032 |
| BEM relevance | D_c input plus direct D_s overlap (see below) |
1. Viitasaari–Vesanto (map sheets 10–11 of the EIA programme's map appendix, 1:30,000, FCG/Fingrid 9/2025). Both sheets carry the label "Rautalammin reitti, Kuhankosken yläpuolinen vesistö, MUU090023" directly on the route corridor — confirming this segment crosses the Rautalammin reitti watershed itself, even though the map sheets administratively cover Viitasaari and Vesanto municipalities (not Rautalampi). Watershed boundaries do not follow municipal boundaries, so this is a third, more northerly touchpoint with the BEM pilot area, in addition to the two segments below.
2. Tyyrinvirta–Vehkamurto (Rautalampi–Suonenjoki): runs alongside the existing 110 kV line, corridor widening ~74 m. East and west options under review; the eastern option runs ~5.4 km alongside the existing line (Lonkari–Vehkamurto), the western entirely through a new corridor.
3. Ahveninen–Mehtiö (Suonenjoki–Rautalampi–Pieksämäki): new corridor, ~96 m width. East and west options; the eastern option is ~500 m longer than the western, but the developer itself notes this as a deliberate mitigation — the western option would cross Mehtiönkoski, a wilderness fishing rapid already documented by the local Myhinpää village association.
Map sheet 10 marks a site directly adjacent to the planned route: "88. Ahvenisojan liito-oravan elinympäristö" (Siberian flying squirrel habitat), alongside sites 86–87 ("Ahvenisoja 1/2"). The Siberian flying squirrel (Pteromys volans) is one of BEM's own D_s indicator species named in TN-015 (forest group: flying squirrel, black woodpecker, capercaillie, osprey) — this is not merely a D_c input, it directly touches the D_s component at the same location, at the same time.
This matters for BEPP v2's own multiplicative structure (D_f^α · D_s^β · D_c^γ): the model is specifically designed to capture non-linear amplification when multiple stressors converge on the same location at the same time, rather than simply summing linearly. This is the first concrete, located example of such an overlap in BEM's own pilot data — no longer only a theoretical modelling assumption.
The flying squirrel is also an Annex IV(a) species under the EU Habitats Directive (92/43/EEC); degrading its breeding or resting sites is prohibited in Finland without a derogation permit — which makes segment 2 (the Tyyrinvirta–Vehkamurto vicinity) the legally most significant of the touchpoints identified so far.
Map sheets 10–14 have now been reviewed (Viitasaari/Vesanto through Konnevesi/Hankasalmi) — the remaining map sheets, including the ones covering the Tyyrinvirta–Vehkamurto and Ahveninen–Mehtiö segments further south, have not yet been reviewed. Precise, machine-readable route geometry (e.g. GeoJSON/Shapefile) has not yet been obtained — the information above was read from map images, not digitised. Updating D_c = 0.61 based on this new information therefore remains open follow-up work. As noted above, this entire section describes the assessment-programme possibility space (multiple route alternatives) — revisit once the impact statement (expected late 2026) confirms the final route.
Reference: Fingrid Oyj — Harjulinja (Murtoperä-Kauppila) project page, updated 2026 · EIA assessment programme November 2025, Appendix 1 Map Sheets (sheets 10–14, FCG Rakennettu Ympäristö Oy, 1:30,000) · Myhinpää village association, Harjulinja statement, December 2025.
Map sheets 12, 13, and 14 of the EIA programme's map appendix cover the route through Konnevesi and into Hankasalmi (Central Finland region) — a segment substantially denser in protected-area designations than any other Harjulinja segment reviewed to date. All three sheets carry the same repeated label directly on the route corridor: "Rautalammin reitti, Kuhankosken yläpuolinen vesistö, MUU090023" — this is no longer an inference from administrative boundaries (as with the Viitasaari–Vesanto segment in §06c) but a directly, repeatedly labelled confirmation that this stretch of route runs through the Rautalammin reitti watershed itself.
Map sheet 14's title block reads "Konnevesi, Hankasalmi" — meaning the route crosses into Hankasalmi, a Central Finland municipality not previously identified among the nine core watershed municipalities (Pielavesi, Keitele, Tervo, Vesanto, Kuopio, Suonenjoki, Pieksämäki, Rautalampi, Konnevesi). This extends the set of municipalities with a direct Harjulinja touchpoint within or adjacent to the pilot watershed.
Across the three sheets, the legend itself expands to include categories not present on map sheets 10–11: Natura 2000-alue, Valtion luonnonsuojelualue (state nature reserve), Valtion suojelutarkoituksiin varattu alue, Vanhojen metsien suojeluohjelma (old-growth forest protection programme), Soidensuojeluohjelma (peatland protection programme), Koskiensuojelualue (rapids protection area), Rantojensuojeluohjelma (shore protection programme), and Valtakunnallisesti arvokas kallioalue (nationally valuable rock/cliff area).
Named protected sites visible on these three sheets include: Hajalan luonnonsuojelualue (YSA200111), Isosuo (MLO350104), Myllyniskan metsäkortekorpi, Kataja-ahon lehto, Kytömurronsuo (MUU090478), Kytömurronkankaan luonnonsuojelualue, "Sydänmaa – Suomi 100" (YSA239859, a reserve apparently named for Finland's 2017 centennial), Honkaneva peatland reserve, Kalajärvi-Kytäneva (F10900059), Linnunlaulu (YSA245386) and Linnunlaulun luonnonsuojelualue (YSA230723), Leppälehdon luonnonsuojelualue (YSA206616), Aholan ranta (YSA097389), and Veräjänranta (YSA097424) — considerably more named sites per map sheet than the Viitasaari–Vesanto segment in §06c.
Taken together with §06c's flying-squirrel finding, this confirms the Kuhankoski–Konnevesi stretch — immediately upstream of Virmasvesi/Iisvesi in the Rautalammin reitti flow chain (Pielavesi → Nilakka → Äyskoski → Rasvanki–Virmasvesi–Iisvesi–Niinivesi → Koskelo/Hankavesi → Etelä-Konnevesi → Siikakoski) — as the most protected-area-dense segment of the Harjulinja route identified to date. This is precisely the stretch flagged as of particular BEM interest: watershed network, forest corridors, valuable habitats, recreational use, and landscape value converge here, immediately upstream of HEM's own Virmasvesi/Iisvesi operating area (TN-014).
Two further map sheets showing Virmasvesi and Iisvesi directly have been referenced in discussion but are not yet available in this repository — once obtained, they should extend this section with the segment most directly relevant to TN-014. As with §06c, no machine-readable route geometry has been obtained; this entry is based on OCR-assisted reading of map screenshots, not digitised GIS data. This section, like §06c, describes the assessment-programme possibility space, not the final route — the Kuhankoski–Konnevesi corridor and the specific sensitive sites named above may narrow, shift, or partly disappear once the impact statement (expected late 2026) selects between the mapped alternatives.
Reference: Fingrid Oyj — Harjulinja (Murtoperä-Kauppila), EIA assessment programme November 2025, Appendix 1 Map Sheets (sheets 12–14, FCG Rakennettu Ympäristö Oy, 1:30,000).
BEM is not a research report. It is a continuously updating observation system — closer in architecture to a meteorological pipeline than to an analysis notebook. This distinction determines every subsequent technical decision.
The single most important decision is to not hardcode BEPP into a fixed formula. Instead, BEM maintains a model registry alongside a stable feature layer:
Feature layer (stable): D_f, D_s, D_c, R pipelinesThe feature layer processes raw data into ecological indicators. The model layer computes BEPP from those indicators. When the model is updated — better mathematics, better calibration, better understanding — the feature layer does not change. This is the distinction between research code and an instrument.
| Layer | Function | Technology |
|---|---|---|
| Data ingestion | Sentinel-2 (NDVI, land cover), FinBIF (species), YVA permits, Metsähallitus protection areas | SentinelHub STAC API · PostGIS · GeoJSON |
| Spatial-temporal storage | Indexed raster + vector data with time dimension | PostgreSQL + PostGIS · Cloud Optimized GeoTIFF |
| Feature engine | D_f: forest cover, patch size, edge density, corridor loss D_s: species presence, indicator trends D_c: disturbance rate, effective footprint R: recovery capacity, buffer ratio | GeoPandas · Rasterio · Xarray |
| BEPP core | S(t) computation via model registry | Python · versioned model config |
| Dynamic engine | dS/dt, d²S/dt², T(t), tipping probability | SciPy ODE solver |
| Visualisation | Regional map, BEPP heatmap, trajectory arrows, tipping zones, replay 2015→2026 | Kepler.gl · MapLibre |
The long-term architectural goal is convergence: WEM (energy), HEM (hydrology), and BEM (ecology) sharing the same ingestion layer, trajectory engine, and tipping framework — with domain-specific data sources and models on top. This transforms ACI from a document collection into a monitoring architecture. The 30-day pilot sprint (PostGIS setup → Sentinel NDVI pipeline → FinBIF integration → first BEPP heatmap) is the first concrete step toward that convergence.