Skip to content

Reservoir-Flux Network

Version
v1 · 2026-08-24 · History
Prime #
1140
Origin domain
Systems Thinking & Cybernetics
Subdomain
stock and flow modeling → Systems Thinking & Cybernetics

Core Idea

A reservoir-flux network decomposes a system into named reservoirs (stocks, compartments, pools) linked by fluxes (flows, transfer rates) under a conservation closure that holds the total content invariant except across a declared boundary. The closure is the discriminating feature — it is what separates the pattern from a bare graph and from a single isolated flow, and what makes mass-balance bookkeeping calculable.

How would you explain it like I'm…

Buckets and Pipes

Picture a row of buckets with little pipes between them, and water moving from bucket to bucket. None of the water disappears — it only moves from one bucket to another, or out a marked drain. If you know how much is in each bucket and how fast the pipes flow, you can figure out what happens next. The trick is that water is never lost, just relocated.

Tanks That Never Leak

A reservoir-flux network is a set of named tanks connected by flows, where the total amount of stuff is kept fixed unless it crosses a marked boundary. Each tank is a stock you keep track of; each flow is a pipe carrying stuff from one tank to another in a set direction. The big rule is conservation: whatever leaves one tank must show up in another tank or cross a labeled boundary to the outside — nothing just vanishes. Because of that rule, you can do real bookkeeping: inflow equals outflow plus whatever piles up. Deciding how to draw the tanks — one big tank or several small ones — is the main choice you make, and it decides where you can later step in to fix things.

Stocks, Flows, and Conservation

A reservoir-flux network decomposes a system into a finite set of named reservoirs (stocks, pools, accounts) connected by fluxes (directed flows) under a conservation closure that holds total content fixed except across an explicitly named boundary. Three commitments are jointly load-bearing: reservoirs individuate what accumulates, fluxes specify the directed pathways content moves along, and conservation guarantees that whatever leaves one reservoir arrives elsewhere inside the system or crosses a named boundary. Conservation is the discriminating feature — it separates this from a bare graph of nodes (no conservation) and from a single isolated flow (no graph, no stocks). That closure unlocks the reasoning toolkit: mass-balance accounting, steady-state analysis, residence times, time constants, and shock propagation. Naming the reservoirs — deciding where to draw their boundaries — is the major modeling choice, since it fixes where every later intervention can be located.

 

A reservoir-flux network is the structural arrangement in which a system is decomposed into a finite set of named reservoirs — stocks, compartments, pools, accounts, levels — connected by fluxes — flows, transfer rates, directed channels — under a conservation closure that holds the total content invariant except across an explicitly declared boundary. Three commitments are jointly load-bearing and dropping any one collapses the family of reasoning the pattern licenses: reservoirs individuate the things whose accumulated levels matter; fluxes specify the directed pathways along which content moves; conservation asserts that whatever leaves one reservoir arrives somewhere else inside the bounded system, or crosses a named boundary, so the summed contents (corrected for boundary exchange) do not change under internal dynamics. The conservation closure is the discriminating feature — it separates the pattern from a bare graph of connected nodes (no conservation) and from a single isolated flow (no graph structure, no stocks). The closure also makes the reasoning catalogue available: mass-balance accounting (inflow = outflow + accumulation), steady-state analysis (levels at which all fluxes balance), residence-time calculation (reservoir size over through-flux), time-constant estimation, and shock propagation. A fourth feature is naming: reservoirs are labelled rather than anonymous, and where to draw their boundaries — lumping versus splitting — is the major modelling choice that determines where every later intervention can be located. Once reservoirs are named and fluxes drawn, 'where should we act?' becomes a question about a specific edge or node, and 'is our account complete?' becomes an auditable arithmetic identity.

Broad Use

  • Biogeochemistry: the carbon cycle as atmosphere/ocean/biosphere/soil reservoirs linked by photosynthesis and respiration, total carbon conserved.
  • Pharmacokinetics: compartmental PK/PD models with first-order rate constants and elimination, drug mass conserved minus clearance.
  • Epidemiology: SIR/SEIR partition a population into Susceptible/Infected/Recovered reservoirs with conserved total.
  • Macroeconomics: flow-of-funds and sectoral balances conserved by double-entry bookkeeping.
  • System dynamics: the entire Forrester stock-and-flow discipline of stocks, rates, and equations.
  • Ecology and hydrology: trophic energy budgets and watershed water balances under thermodynamic or hydrological closure.

Clarity

Separates a flow (one directed channel), a network (graph topology, silent on conservation), and a reservoir-flux network (graph plus stocks plus closure), and makes the closure boundary an explicit modelling choice whose misplacement a failed conservation check flags.

Manages Complexity

Compresses an arbitrarily complicated multi-substance system into a short specification — how many reservoirs, what fluxes, what rate laws, what total — from which steady states, residence times, and shock propagation follow, with conservation as an auditable identity.

Abstract Reasoning

The dynamics are coupled rate equations (each reservoir's change equals inflows minus outflows); steady states solve a linear system, perturbation response follows the Jacobian's eigenstructure, and looped topologies can oscillate where feed-forward ones relax monotonically.

Knowledge Transfer

  • Pharmacokinetics → epidemiology: one- and two-compartment dosing solutions carry over to SIR-family models, both linear reservoir-flux networks.
  • Macroeconomics → ecology: the sectoral-balance identity ("one sector's surplus is another's deficit") is the same closure that yields trophic energy budgets.
  • System dynamics → climate/inventory: the "bathtub" result (a stock rises even as inflow levels off) transfers to atmospheric carbon and to backorders.

Example

The two-compartment PK model — central and peripheral compartments with rate constants k12, k21, and elimination k10 — makes the mass-balance identity numerically explicit: total drug equals both compartments plus what was eliminated, equal to the dose, and a failed balance points straight at a forgotten elimination flux.

Relationships to Other Abstractions

Local relationship map for Reservoir-Flux NetworkParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Reservoir-FluxNetworkPRIMEPrime abstraction: Conservation Laws — presupposesConservationLawsPRIMEPrime abstraction: Source-Sink Role — presupposesSource-Sink RolePRIMEPrime abstraction: Network — decomposeNetworkPRIMEDomain-specific abstraction: Biogeochemical Cycle — is a kind ofBiogeochemicalCycleDOMAINDomain-specific abstraction: Lake ecosystem — is a kind ofLake ecosystemDOMAINDomain-specific abstraction: Ventricle (heart) — is a kind ofVentricle(heart)DOMAINDomain-specific abstraction: WAFLEX — is a kind ofWAFLEXDOMAIN

Current abstraction Reservoir-Flux Network Prime

Parents (1) — more general patterns this builds on

  • Reservoir-Flux Network presupposes Conservation Laws Prime

    A Reservoir-Flux Network presupposes a conservation law that closes internal transfers over its declared boundary.

Children (6) — more specific cases that build on this

  • Biogeochemical Cycle Domain-specific is a kind of Reservoir-Flux Network

    A biogeochemical cycle is the Earth-system specialization of a conserved reservoir-flux network, fixing the reservoirs and transformations to planetary bio-geo-chemical ones.

  • Lake ecosystem Domain-specific is a kind of Reservoir-Flux Network

    The proposed strict upward parent is prime:reservoir_flux_network.

  • Ventricle (heart) Domain-specific is a kind of Reservoir-Flux Network

    The proposed strict upward parent is prime:reservoir_flux_network.

  • WAFLEX Domain-specific is a kind of Reservoir-Flux Network

    WAFLEX strictly specializes Reservoir-Flux Network.

  • Source-Sink Role Prime presupposes Reservoir-Flux Network

    Assigning a source or sink role requires a bounded stock-flow account linking the component to a reference reservoir.

Hierarchy path (1) — routes to 1 parentless root

Not to Be Confused With

  • Reservoir-Flux Network is not a bare Network because it adds stocks and a conservation closure that license mass-balance reasoning, whereas a bare network carries only connectivity.
  • Reservoir-Flux Network is not Turnover because turnover (and residence time) is one derived rate read off the structure, whereas the prime is the whole conserved system.
  • Reservoir-Flux Network is not Equilibrium because the prime is the structure whose dynamics may or may not reach a resting state, whereas equilibrium is that state itself.