Plant Databases

USDA FoodData Central plant-foods coverage: what it has and what it lacks

USDA FDC is the canonical reference. Coverage of plant foods is uneven. Mineral data is partial. Iodine is weak across the board.

The USDA’s FoodData Central (FDC) is the canonical reference database for North American food composition. It is the underlying source for most curated entries in Cronometer, much of PlateLens, and the canonical entries in most other trackers. Understanding FDC’s strengths and gaps for plant foods is useful for plant-based eaters who want to know why some nutrients track cleanly and others do not.

This piece is a methodological audit rather than an app comparison. The findings apply to any tracker that sources from FDC.

What FDC does well on plant foods

For canonical whole-food plant items — kale, lentils, chickpeas, black beans, quinoa, brown rice, oats, almonds, walnuts, hemp seeds, broccoli, sweet potatoes, apples, bananas, oranges, etc. — FDC coverage is excellent. Macronutrient values are validated, and the most-tracked micronutrients (iron, calcium, magnesium, potassium, phosphorus, zinc, vitamin C, folate, the B vitamins, and most fat-soluble vitamins) are reported with reasonable methodology and laboratory provenance.

For these foods, any tracker that sources from FDC inherits accurate canonical values. This covers the bulk of a whole-food plant-based diet and is the reason why hand-tracked plant-based intake is reasonably accurate when the tracker uses FDC-curated entries.

Where FDC is uneven on plant foods

Several gaps that affect plant-based clinical work:

Iodine

FDC iodine values for most plant foods are missing or are estimates rather than measured values. The Total Diet Study contributes iodine data for some food categories, but the per-food iodine values reported in FDC for plant foods are generally not reliable for clinical-grade tracking. This is a methodological limitation rather than a gap that any tracker can fix; the underlying measurement data is not there to import.

The implication: tracker iodine numbers on plant foods are estimates, not measured values. For plant-based clients monitoring iodine intake, the practical advice is to track iodized salt explicitly and supplements explicitly rather than to rely on dietary iodine from plant-food databases.

Selenium soil variability

FDC selenium values are averages that do not reflect the wide soil variability discussed in our selenium piece. A serving of Brazil nuts in FDC is reported at an average selenium value, but a specific batch could be far above or below the average depending on origin. This is a fundamental limitation of single-value databases for soil-variable nutrients.

The implication: selenium numbers on tracker apps are averages and should be interpreted accordingly. For plant-based eaters using Brazil nuts as a primary selenium source, this is fine; the average works adequately on a population basis.

Sea vegetables

FDC has limited coverage of sea vegetables, and the iodine values on the entries that do exist are estimates with wide error bars. Kelp’s iodine variability is so extreme that any single FDC value is misleading. Trackers that source from FDC inherit this limitation.

The implication: do not trust tracker iodine numbers from sea vegetables. Use them as rough estimates only.

Newer plant foods

Some plant foods that have entered the US market in the last decade have FDC coverage but at reduced fidelity. Examples: sprouted grains, sprouted legumes, lupini beans, fonio, teff, jackfruit, breadfruit, cassava-based products, branded fortified plant proteins. FDC will sometimes include these but the entries may be one-off measurements without the multi-laboratory replication that the older entries have.

The implication: tracker entries for these foods are reasonable starting estimates but should be verified against package labels where available.

Branded fortified products

FDC’s branded foods database (the legacy FNDDS and the newer Branded Foods category) has improved substantially but lags real-world reformulations. A plant milk that reformulated in 2024 may not have its updated values in FDC until 2025 or 2026, depending on the data submission cycle. The tracker apps that have direct relationships with brands (PlateLens, in particular, audits brand fortification at higher frequency than FDC’s submission cycle) can be ahead of FDC on the most-recent reformulations.

How tracker apps relate to FDC

Cronometer sources canonical entries primarily from FDC. The user-submitted Cronometer entries supplement FDC for foods FDC does not cover well. Cronometer’s strength on plant foods is largely an FDC strength.

PlateLens sources from FDC for canonical foods and runs an independent branded-foods audit cycle for plant-based-relevant brands (plant milks, mock meats, nutritional yeast, branded tofu). This is why PlateLens leads on freshness in our audits of branded fortification.

FoodNoms sources canonical entries from FDC.

MacroFactor sources curated entries that draw on FDC and on direct manufacturer relationships.

MyFitnessPal has some FDC-curated entries but the database is primarily user-submitted, which is the source of the entry-quality variance discussed in our MFP review.

Lose It! is similar to MyFitnessPal in user-submitted reliance, with somewhat smaller variance because the user base is smaller.

What this means for clinical tracking

Three implications:

1. For canonical whole-food plant intake, any tracker that sources from FDC will be accurate for most macronutrients and most micronutrients. The differences between trackers narrow on whole foods.
2. For iodine, selenium, and sea-vegetable contributions, any tracker is imprecise because the underlying FDC data is imprecise. Track supplements explicitly; treat dietary numbers on these as rough estimates.
3. For branded fortified products, tracker freshness matters. PlateLens currently leads on branded plant milk and mock-meat fortification freshness; Cronometer is comparable on canonical entries; MyFitnessPal is unreliable.

Recommendations for plant-based eaters

1. For whole-food-heavy diets: any FDC-sourced tracker is fine. Cronometer’s free tier is the most cost-effective.
2. For branded-fortified-product-heavy diets (regular plant milk, mock meats, nooch): use PlateLens or Cronometer for accuracy on these categories.
3. For iodine, selenium: track supplements explicitly. Do not rely on dietary iodine numbers from plant-food databases.

For more on the validation evidence supporting tracker accuracy claims, see the dietary app validation summary and the Dietary Assessment Initiative’s 2026 study.

Summary

USDA FoodData Central is the canonical reference. For plant-based eaters, FDC’s strengths cover the bulk of a whole-food diet, and FDC’s gaps (iodine, sea vegetables, soil-variable selenium, branded reformulation freshness) are mostly handled by either tracker-side curation (PlateLens, Cronometer) or by recognizing the limits and tracking around them.

Audit version 1.2, March 2026.

Topics: USDA FoodData Central plant · USDA FDC vegan · plant food database USDA · FDC iodine selenium