Basin Futures · Explorable models

How Basin Futures models water storages

Every storage in a Basin Futures model is a daily water balance: inflow in, evaporation and releases out, spill once it is full and shortfall once it is empty. This page runs that calculation on Lake Eildon — 3,400 GL on the upper Goulburn, supplying one of Australia's largest irrigation districts. Set the demand, and fifteen years of recorded inflows (including the Millennium Drought) determine how the reservoir fares.

The daily water balance re-simulates in your browser as you drag. Grey dots are satellite estimates of the actual storage. Try to find the operating demand that reproduces them, then see what happens when you ask for more.

Ground · one day at the dam

Each day the reservoir gains inflow I from five gauged tributaries, loses evaporation E from its surface, and releases water through the outlet: irrigation demand plus a minimum environmental flow. When it's full, the excess goes over the spillway spill; when it's empty, deliveries fall short. Press ▶ or scrub the charts; the window chips zoom into the drought and the refill that ended it.

Live storage — Lake Eildon

Lake Eildon · Goulburn River, Victoria
3,400 GL capacity, 3,877 km² catchment in the Basin Futures Goulburn case-study model.
Inflows: 5 gauges (2,423 km², scaled ×1.6) · observed storage: GloLakes satellite (Landsat + ICESat-2) · evap: CRU PET − MSWEP rain over the lake surface · window 2000–2014

window

inflow (GL/day)

release + spill (GL/day)

your storage satellite observed inflow spill shortfall

Rung 1 & 2 · operate it, then see the family

Start gentle: at low demand the reservoir rides high and spills, while the grey dots show the real Eildon drawn far lower. Raise the demand until your trajectory tracks the satellite record: that number is the district's actual draw. Keep raising it and watch reliability break in the drought years. Sweep shows the whole family of futures one slider controls.

Operations — drag to operate

match r— reliability— spill—

Rung 3 · the yield–reliability landscape

Every pixel below is a full fifteen-year simulation: a demand asked of a dam of some capacity, coloured by how reliably it was delivered. The cliff edge is the storage yield: the most you can promise before drought years start breaking it. Notice how little extra capacity buys once demand passes the mean inflow (~1,460 GL/yr). Click anywhere to jump the whole page to that design.

Reliability over demand × capacity (env flow & season held)

demand = 200annual demand (GL/yr) →2,600

~2,700 full simulations.

About this page. A simplified daily storage balance (inflow − evaporation − release, with spill above capacity) built for teaching. Basin Futures' full storage engine adds operating rules, dead storage and supply levels, hydropower and per-use demands on top of the same water accounting. Inflows are the five gauged Eildon tributaries (Goulburn at Dohertys, Big, Howqua, Jamieson and Delatite rivers; 2,423 km² gauged, scaled to the 3,877 km² catchment). Observed storage is the GloLakes satellite series (Landsat + ICESat-2). Surface evaporation uses CRU PET minus MSWEP rain over a level-dependent lake area. Release rule: seasonal irrigation pattern + constant environmental flow, curtailed only when the lake runs dry.

Inspired by Bret Victor's Up and Down the Ladder of Abstraction.