← Current level & trend
The record — not a forecast

El Niño, La Niña & Lake Travis

Lake Travis fills in sudden flood pulses — usually spring or fall, not steady rain. La Niña years tend to drain it; the refills usually come in an El Niño or neutral year. In Central Texas, droughts end in floods.

Last updated

of La Niña years, the lake fell
71%of La Niña years, the lake fell
when floods actually come
May–Jun · Sep–Octwhen floods actually come
full pool — it sits below by design
681 ftfull pool — it sits below by design
record flood — 1952, no hurricane
441k cfsrecord flood — 1952, no hurricane

How Lake Travis actually fills

Locals hear “El Niño fills the lake.” The data says floods do. Lake Travis is a flood-control reservoir on the Colorado River, engineered to sit below its 681-foot full pool — it drains for months, then takes it all back in a few violent days, almost always in late spring or fall and almost never in winter. Two separate things decide a year. The storm type to deliver the rain — typically a traditional storm that stalls over the Hill Country, not a hurricane. The Pacific only tilts the odds: El Niño and neutral years deliver most of the big refills, while La Niña stacks the deck toward a draining lake. But the flood itself can't be forecast months out — the odds are all you can read.

Lake Travis Water History

Every flood, drought, and El Niño / La Niña year on one timeline — since 1940.

Read it top-down: the dots are floods, the Niños strip is the Pacific phase, the bars are each year's rain. Big refills cluster in El Niño & neutral years; La Niña tilts the basin dry. Floods fill Lake Travis — the Pacific tilts the odds.

The flood recipe

What it takes to fill Lake Travis — you need all five at once

  • Always hereBalcones EscarpmentThe hills shove wet air sharply upward, wringing the rain out of it.
  • Always hereThin, rocky soilWater can't soak in — it runs straight off the limestone into the rivers.
  • The variableDeep moisture plumeA soaking-wet air mass off the Gulf — or the leftovers of a tropical storm.
  • The variableA storm that stallsRain has to park over one spot for hours or days. This is the rare ingredient.
  • The variableA drought-primed lakeA low lake has the most room to jump — the biggest refills follow the worst droughts.

The Hill Country and the rocky soil never change — Lake Travis is built to flash-flood. The data shows these floods are nearly impossible to forecast months in advance. All you can read is the odds — and the clearest one is simple: La Niña years usually decrease lake levels. El Niño years have better odds at increasing lake levels, but the data shows they are no better than neutral years.

When floods come

Hill Country river flood peaks by month — all gauges, since 1916

Two seasons do almost all the work — late spring and fall. Winter almost never floods. So if Lake Travis is going to jump, the odds favor May–June, then again September–October.

El Niño vs. La Niña: which fills Lake Travis?

Average flood-strength index by Pacific phase — 0 = a normal year, 1950–2025

  • El Niño↑ +0.71
  • Neutral↑ +0.66
  • La Niña-0.24

These bars are a standardized index, not a percentage — 0 is an average year, positive means stronger floods. (Don't confuse El Niño's +0.71 here with the 71% below: one is a flood-strength score, the other is how often the lake fell.)

La Niña is the one pattern that reliably dries Central Texas out — in La Niña years the lake fell 71% of the time. When the floods come back, it's usually an El Niño or a neutral year doing the work.

But the lake doesn't read the rules. The two biggest floods on record (1952, 1935) weren't driven by a hurricane, and the catastrophic July 2025 flood hit during a La Niña. So La Niña stacks the odds toward a low, draining lake — it does not promise one, and a fill can still break through in any year.

Every major Lake Travis flood since 1935

The floods that refilled the lake — and what caused them

  • Sep 11, 1952Pedernales
    441,000cfs peak

    Largest flood on record. A stalled storm dumped ~29" near Fredericksburg, ending the 1950s drought of record. No hurricane — an ordinary storm that parked.

    Stalled storm — no hurricaneEl Niño
  • Jun 14, 1935Llano
    380,000cfs peak

    The Llano River’s all-time crest. Spring thunderstorm complex.

    Stalled storm — no hurricane
  • Oct 16, 2018Llano
    278,000cfs peak

    The Kingsland flood — moisture from Hurricane Sergio’s Pacific remnants. Took out the FM-2900 bridge.

    Tropical-fueledEl Niño
  • Jun 23, 1997Llano
    260,000cfs peak

    A stalled June storm on the Llano — among the biggest peaks in the record.

    Stalled storm — no hurricaneNeutral
  • Sep 8, 1980Llano
    210,000cfs peak

    Tropical Storm Danielle’s remnants parked over the Llano basin.

    Tropical-fueledEl Niño
  • Oct 5, 1969Llano
    154,000cfs peak

    A stalled fall storm on the Llano.

    Stalled storm — no hurricaneEl Niño
  • Oct 13, 1973Llano
    154,000cfs peak

    A stalled fall storm on the Llano — a major fill that broke the dry tilt.

    Stalled storm — no hurricaneLa Niña
  • Nov 4, 2000Llano
    151,000cfs peak

    A stalled November storm on the Llano.

    Stalled storm — no hurricaneLa Niña
  • Oct 4, 1959Pedernales
    142,000cfs peak

    A stalled fall storm on the Pedernales.

    Stalled storm — no hurricaneNeutral
  • Aug 3, 1978Llano/Pedernales
    139,000cfs peak

    Tropical Storm Amelia stalled over the Hill Country — ~48" at Medina, the wettest U.S. tropical cyclone on record at the time.

    Tropical-fueledEl Niño
  • Jul 4, 2025Llano
    132,000cfs peak

    The July 4 Hill Country flood — remnant tropical moisture (TS Barry) during a La Niña. Proof a fill can come in any phase.

    Tropical-fueledLa Niña
  • Jun 2007(lake +35 ft)
    Major refill

    Wet summer ended a multi-year drought; Lake Travis rose ~35 ft. Gulf moisture, no hurricane.

    Stalled storm — no hurricaneEl Niño
  • May 25, 2015Blanco/Pedernales
    Major refill

    Memorial Day floods broke the 2011–2015 drought; the lake jumped ~36 ft. A neutral-ENSO spring storm — not the El Niño everyone credits.

    Stalled storm — no hurricaneNeutral

What raises and lowers Lake Travis

The lake rises and falls mostly on rain, not water use. It fills when storms send runoff down the Hill Country watershed — and between those floods it slowly drops as water leaves three ways.

  • EvaporationThe sun is one of the biggest steady losses — the Highland Lakes give up an estimated 150,000–200,000 acre-feet a year, six to eight times the whole volume of Lake Austin.
  • Water supplyCities and industry draw firm supply year-round; the City of Austin is the largest single customer. Firm customers keep their water even in the worst drought.
  • Downstream releasesLCRA sends water downstream for river flows, freshwater into Matagorda Bay, and farms. The releases people argue about most are the interruptible ones for rice farming — LCRA cuts those off first in a drought to protect cities, which is why agricultural water was nearly zero in 2024–2025.

How Lake Buchanan fits

Lake Travis works as a pair with Lake Buchanan upstream. Buchanan is a water-supply reservoir; Travis is the only Highland Lake built to hold back floods. LCRA manages them together as “combined storage” — which is why it doesn't simply drain Buchanan to top off Travis: moving water between lakes adds no new water, Buchanan loses more to evaporation, and Travis is kept ready to catch the next storm.

Sources: LCRA (water supply, dams & lakes, evaporation) and KXAN (Buchanan vs. Travis).

Glossary

What these terms mean
ENSO (El Niño / La Niña)
A Pacific Ocean temperature cycle that tilts Central Texas winters wetter or drier. El Niño is the warm phase (leans wetter); La Niña is the cool phase (leans dry — the pattern that drains the lake).
RONI
Relative Oceanic Niño Index — NOAA's current measure of El Niño/La Niña strength. Past ±0.5 counts as an event; the bigger the number, the stronger.
PDO
Pacific Decadal Oscillation — a much slower Pacific cycle. On our lake it's only a weak nudge, not a reliable driver, which is why this panel doesn't lean on it.
Water year
October 1 through September 30 — the standard window for tallying a season's rain and runoff, since the wet season spans the calendar-year boundary.
cfs
Cubic feet per second — how a river's flood flow is measured. For scale, the record 1952 flood peaked at 441,000 cfs on the Pedernales.
Flood pulse
A sudden surge of river inflow from a big storm parked over the watershed — how Lake Travis actually refills, as opposed to gentle, steady rain.
Full pool
681 feet — the top of the lake's normal (conservation) range. Above that is flood storage; the reservoir spends most of its life below it by design.

El Niño, La Niña & Lake Travis FAQ

Does El Niño fill Lake Travis?

Not on its own — it only tilts the odds. El Niño and neutral years host most of the big refills, but the lake fills from a few floods that can land in any year. In the other direction the signal is stronger: in La Niña years Lake Travis has fallen about 71% of the time.

How does Lake Travis refill?

In sudden flood pulses, not steady rain. Lake Travis is a flood-control reservoir that drains for months and then regains water in a few violent days, when heavy storms park over the Hill Country watershed upstream. Those pulses come almost always in late spring or fall, and almost never in winter.

When does Lake Travis usually flood?

Late spring (May–June) and fall (September–October) do almost all the work; winter floods are rare. Across the gauge records since 1916, late spring is the single most common time for major flood peaks, with a second peak in the fall.

Is La Niña bad for Lake Travis?

It stacks the odds toward a draining lake: in La Niña years Lake Travis has fallen about 71% of the time, including the 1950s drought of record and the 2022–2024 drawdown. But that's odds, not a verdict — a fill can still break through, as the catastrophic July 2025 Hill Country flood showed during a La Niña.

Will Lake Travis ever be full again?

Almost certainly — eventually. Every Texas drought on record here has ended in a flood, and Lake Travis is built to refill to its 681-foot full pool. What can't be predicted is when: a refill depends on a single big storm parking over the watershed, which can't be forecast months ahead.

A note on what this is: the record, not a forecast. These patterns describe the odds over many years — they can tell you when the lake tends to fill, never when the next flood will hit. The flood itself can't be called months out. The views span different windows because each uses the full reach of its own dataset — the lake-level line since 1940, the El Niño / La Niña record since 1950, the by-month flood tally since 1916, and the biggest floods back to 1935. Sources: USGS streamflow (Llano, Pedernales, San Saba), NOAA ENSO & PDO, NOAA HURDAT2, LCRA.

ShareXRedditFacebook