Which levers actually move the numbers
Because every environmental quantity in this dataset is derived from a small set of coefficients, you can see exactly how sensitive the totals are to each one. Treat the outputs as parameter arithmetic, not prediction.
Adjust the parameters
Framed scenarios
Scenario water demand by state
m³ per year under the current parameter settings.
How each lever propagates
Scales energy, and therefore carbon, water and waste heat, one-for-one. The single most consequential coefficient in the model.
Scales water and scarcity-weighted water only. Does not touch energy or carbon.
Scales carbon only. A clean grid removes carbon without reducing energy, water, heat or noise.
Contractual, not physical: it changes market-based accounting, not the electrons drawn or the local footprint.
Scales everything simultaneously, including population exposure if new sites follow existing siting patterns.
Land use, setback distance and who lives within one kilometre are siting decisions, not efficiency parameters.
This is a coefficient sandbox, not a forecast. It holds siting, load factor, technology mix and facility count fixed except where you change them, applies changes uniformly to every facility in view, and captures no feedbacks, no grid constraints and no policy response. Median record confidence of the underlying records is 2.25 of 5.