Duck Curve Explained

The duck curve creates two operational problems simultaneously: an oversupply problem at midday and a ramping problem at sunset. The solutions for each are different, and deploying one does not necessarily fix the other.

The oversupply problem occurs when midday solar generation exceeds what the grid can absorb, forcing curtailment or negative prices. The ramping problem occurs when solar fades at sunset and 10-15 GW of dispatchable generation must come online within 3 hours to meet evening peak demand. Storage addresses the oversupply (absorb midday excess) and partially addresses the ramp (discharge during evening). But current 4-hour battery deployments can only shave the peak of the ramp, not sustain output through the full evening.

CAISO's operational math. On a deep spring day in 2024, CAISO's net load dropped to 3 GW at 1 PM (oversupply); by 7 PM it had risen to 28 GW. The 25 GW swing in 6 hours is the operational challenge.

Can batteries deployed for oversupply also solve the ramp?

Partially. If 10 GW of batteries charge during the midday belly (absorbing oversupply), they can discharge 10 GW during the evening ramp. But the ramp requires 25 GW for 3-4 hours. Batteries alone cannot cover the full swing at current deployment levels. Gas peakers fill the remainder, which is why gas capacity has not declined in California despite solar additions. Solving the duck curve requires storage at a scale roughly 2-3x what is currently deployed, plus demand-shifting programs that move evening load to midday hours.

---