Did you scream out “Southern Humboldt County”?

Give yourself a pat on the back. You are correct.

Over the last few years, one of the major political issues surrounding cannabis cultivation have to do with water — unsurprisingly, given California’s drought. How much do marijuana farmers drain from their watersheds? What are the impacts to those watersheds’ salmon fisheries? Can we build appropriate storage systems, ones that gather enough winter rainfall to slake thirsty plants through the summer? Where are those storage systems most urgently required?


These questions have been the subject of national news coverage, and decriminalization only moves them closer to the fore. Locally, environmental groups such as Humboldt Baykeeper and Friends of the Eel have stationed themselves on the front lines, demanding that upcoming county regulations address water usage. The State Water Resources Control Board has spent the last year or so crafting its own system of permits, enforcement and outreach to growers.

But which local watersheds are most impacted? Where are the grows? The Outpost’s “Weed or No Weed” experiment can provide some answers.

Let’s start by defining our terms. The attached map shows Humboldt County divided broadly by watershed. Specifically, we’re looking at the U.S. Geological Survey’s definition of the nation’s watershed “sub-basins” (HUC-8), as applied to Humboldt County particularly.

We can divide our estimates of Humboldt County marijuana cultivation by property by these watersheds, giving us a separate figure for each of the HUC sub-basins that extend into the county borders: the Lower Klamath, the Trinity, the South Fork of the Trinity, Mad River/Redwood Creek, the Lower Eel (including the Van Duzen), the South Fork of the Eel, and the Mattole.

So we did that.

Roughly 72 percent of the county’s marijuana-growing properties are in the Eel or Mattole watersheds — which together make up only about 46 percent of the county’s total land. 

The remainder of the county — that belonging to the Klamath, the Trinity and the Mad River/Redwood Creek — grows far, far less.

In sum, the numbers look something like this.

Watershed Total parcels
(>1 acre)
Est. # of parcels w/grows %
Lower Klamath 1,963 130-248 6.6% - 12.6%
Trinity 1,980 97-165 4.9% - 8.4%
South Fork Trinity 317 13-45 4.1% - 14.3%
Mad-Redwood 10,628 479-709 4.5% - 6.7%
Mattole 3,291 568-765 17.3% - 23.2%
Lower Eel 7,478 788-1,059 10.5% - 14.2%
South Fork Eel 2,888 685-903 23.7% - 31.3%

Full calculations here. Refer to yesterday’s post for instruction on how to read those tables. 

One thing to note: Breaking the data down this way increases the confidence interval in the statistically sampled portion of our study. As the set of parcels under study decreases, or as the proportion of weed-positive parcels nears 50 percent, the spread between our lower confidence limit and our higher one widens. Check out the numbers for the South Fork of the Trinity. We can still say with 95 percent certainty that there are between 14 and 45 weed-positive parcels in that watershed, but it’s a pretty wide target to hit — the difference between 4 percent of the total and 14 percent.

But another way to look at that is by comparing the watersheds one to another. Let’s set aside the “sampled” portion of our dataset and look at just that portion that the crowd marked as either “weed” or “no weed” by at least a three-vote margin. Then let’s compare that to the total hard count of parcels marked as positive.

Watershed Hard-count “weed” % of total “weed” hard count
Lower Klamath 95 6.0%
Trinity 94 6.0%
South Fork Trinity 12 0.8%
Mad-Redwood 232 14.8%
Mattole 324 20.6%
Lower Eel 429 27.6%
South Fork Eel 385 24.5%

The Mattole, the Lower Eel and the South Fork of the Eel each have over 20 percent of the county’s weed-positive parcels. 

Now, remembering that “Weed or No Weed” tells us nothing about the size of the grows on Humboldt County parcels — merely the number of parcels — and still less about the ecofriendliness of those grows, might we still attempt to come up with a rough estimate of the industry’s impact in each of these watersheds?

Let’s compare each watershed’s percentage of weed-positive parcels with their percentage of the county’s total area. This will give us an index ratio that might hint, in the most speculative way, at the impact of Humboldt County’s industry by sub-basin.

Watershed % of county weed % of county area % weed / % area
Lower Klamath 6.0% 14.65% 0.41
Trinity 6.0% 8.43% 0.71
South Fork Trinity 0.8% 3.20% 0.25
Mad-Redwood 14.8% 27.01% 0.55
Mattole 20.6% 13.51% 1.53
Lower Eel 27.6% 24.53% 1.13
South Fork Eel 24.5% 8.67% 2.83

If these speculative ratios in our fourth column, here, can be shown to have any bearing on water-related political questions surrounding the industry right now, it’s interesting to note that the two most heavily impacted watersheds — the Mattole and especially the South Fork Eel — are undammed rivers with few other industrial users, and that a large percentage of the watershed of each is within Humboldt County’s borders. (The Mattole is almost entirely within our borders.) If you wish to tackle these problems seriously, they suggest, it won’t do to point the finger at agricultural water diversions at Potter Valley or Trinity Lake.

Stay tuned! More STATE OF THE WEED coming!