The removal rate puzzle: why are in-district almond orchards being removed faster?

Part 9 of 10 in a series on the almond and pistachio sectors in California. Download a complete report here and explore the underlying data in a standalone application here.

One of the assumptions that runs through most discussions of water risk in Californian permanent crop plantings is that orchards inside irrigation districts - with access to surface water deliveries, in addition to groundwater resources - are better protected than those in white lands, where growers depend entirely on groundwater. The planting data broadly supports this: as we showed earlier in this series, pistachio growers in particular have actively redirected new planting toward in-district land since SGMA implementation began.

The removals data tells a somewhat more complicated story.

The finding

When we compute annual removal rates against standing almond acreage inside and outside irrigation districts, the result is consistent across every year in the dataset - almond orchards inside irrigation districts are being removed at a higher rate than those outside.

The gap varies year to year, but the direction is uniform. In 2022-2024, in-district almond orchards were removed at roughly 6% per year versus roughly 4% for those outside districts - a ratio of approximately 1.6x.

California almonds removed inside and outside irrigation districts 2016-2024. Source: Demeter

It isn't just a Westlands effect

We might initially assume that this pattern is driven by a handful of districts with exceptionally high removal rates. Westlands, Belridge, San Luis and Wheeler Ridge all recorded removal rates well above the in-district average, with Westlands seeing removal rates of 10-16% in 2022-2023 while Belridge exceeded 30%. These districts do indeed pull the aggregate up. But even excluding these high-removal districts entirely, the inside-district rate remains higher than the outside rate in most years. The 2022-2024 average excluding them is 4.8% inside versus 3.9% outside.

The within-GSA comparison

For a cleaner test, we can compare removal rates for the inside-district and outside-district portions of the same GSA, which allows us to control for geography, aquifer and general growing conditions. The picture is mixed at the individual GSA level: some GSAs show much higher inside rates, others show the reverse. But the weighted average across all GSAs with both in-district and outside-district land shows inside rates approximately 1.15x higher than outside.

The age profile doesn't explain it

If inside-district acreage were simply weighted older - with a higher proportion of orchards in the peak removal age brackets - that alone could account for a higher removal rate. But the standing inventory data does not support this. Approximately 20% of inside-district standing acreage is in the 16-25 year age bracket (the ages that account for the majority of removals), compared to 19% outside. The broader 16+ bracket is 25% inside versus 24% outside. The age composition of the two populations is not materially different, so the gap in removal rates cannot be attributed to a difference in the age structure of the standing base.

The pattern is different in pistachios

In pistachios, the direction is reversed: outside-district removal rates are higher than inside, at roughly 2.9% versus 1.1%. This is the opposite of the almond pattern. However, as we discussed in our companion memo on pistachio removals, the majority of pistachio removals are pre-bearing trees. The dynamics driving pistachio removals appear to be fundamentally different from those in almonds, where most removals are of mature, productive orchards. The two crops may not be directly comparable on this measure.

Potential forces at work

The data shows that inside-district almond orchards are being removed at a higher rate than outside-district orchards. It does not tell us why. Several mechanisms could contribute, including but not limited to the following:

Opportunity cost

In-district land often may carry higher underlying value due to its additional water access. A grower deciding whether to keep a marginally profitable almond orchard may face a more attractive alternative use, such as replanting to a higher-value crop or simply selling or leasing it with its attendant water rights - than an outside-district grower whose land is less valuable without district membership. Our final memo explores what happens to this ground after an orchard is removed.

Operating costs

The combined cost of surface and groundwater inside certain irrigation districts may exceed that in some areas outside districts, where lower pumping costs or an absence of groundwater levies may result in a lower aggregate cost of water and correspondingly more profitable growing operations.

SGMA baseline dynamics

Some GSA plans set pumping allowances based on recent historical use. This may have created incentives to maintain orchards - or even plant new ones - during assessment periods to establish a higher baseline allocation, with removal decisions deferred until after baselines were locked in.

Survivorship bias

The outside-district orchards that remain in 2024 may represent a selected population - the ones with the best well yields, the deepest aquifers, the most favourable hydrogeology. The most vulnerable white land orchards may never have been planted in the first place, or may have been removed early enough that the current standing base is biased toward viability.

Water rights grandfathering

Some water allocations granted during the SGMA transition period may be temporary or conditional. If certain rights are not renewed, a future wave of removals could be concentrated in specific locations in ways that are not yet visible in the data.

These are not competing hypotheses. Several may be operating simultaneously, and their relative importance likely varies by district, basin and individual operation.

What the data tells us

The relationship between surface water access and orchard survival is more complex than the standard narrative suggests. District membership is clearly valued by growers making new planting decisions. The planting data shows a strong and growing preference for in-district land. But for existing orchards, being inside a district does not appear to reduce the probability of removal. If anything, for almonds, the opposite is true.

Next in this series: After almonds: what happens to former almond ground in California

You can download a complete report here.