The Setup
Port-adjacent industrial real estate demand is commonly modeled from cargo throughput: more volume through the port means more warehouse, distribution, and transshipment demand. The estimated_cargo_hydrostatic field is the primary machine-derived throughput proxy at vessel-visit resolution. Of 41,377 vessel visits in 60 days:
- 25.9% produced a cargo estimate of exactly 0 tons — the
no_draft_dataplaceholder. This is not a small load; it is a method flag indicating no draft data was available. The field value is 0, but the cargo estimate is absent. - 16.7% produced no estimate at all — null hydrostatic value.
Together, 42.6% of vessel visits contribute nothing meaningful to throughput aggregation. The remaining 57.4% — those with a method-derived figure — split across a hierarchy where method quality varies significantly by port and vessel class.
The Chain
The implication for CRE modeling depends on which port. Ports where bulk carriers dominate and where draft data is consistently available — Houston's ship channel, Norfolk Lambert's Point, Paranaguá — will have higher proportions of draft-based estimates and lower fallback rates. Ports with high vessel turnover, transient traffic, or poor AIS draft reporting will accumulate no_draft_data visits at higher rates, compressing throughput aggregates toward zero even when actual cargo is moving.
That asymmetry means port-adjacent cell scores for markets like Houston or Seattle carry different throughput signal fidelity than comparable cells elsewhere. A cell adjacent to a port with 40% no_draft_data visits and a cell adjacent to a port with 5% no_draft_data visits will both read as "port-adjacent" in a scoring model, but the underlying throughput data quality differs structurally.
Bulk carrier cargo — the primary driver of warehouse and logistics real estate demand — introduces a second complication. The fallback_tons_per_meter model fires on 36.6% of all visits and estimates bulk carrier cargo at an average 7,641 tons per visit. The trim_saline_corrected method — which uses actual measured draft and port salinity — produces 5,557 tons per visit for the same vessel class. That is a 38% gap. A port that appears to handle heavy bulk loads may be reflecting fallback overestimation, not actual cargo volume.
The Implication
Industrial CRE site selection that treats port throughput as an input signal without adjusting for method tier is ingesting a composite that mixes draft-survey measurement, population fallback, and zero placeholders in variable proportions by port. The throughput "number" is not missing — a calculation ran and returned a value — but that value is not uniformly grounded in what physically passed through the port.
The practical consequence: markets where the fallback rate is high will appear to have consistent throughput signal even when draft data coverage is thin. Markets where no_draft_data dominates will appear to have lighter cargo flows than occurred, because the placeholder returned 0 tons rather than an absence of data. Neither of these errors is random; they correlate with port infrastructure and vessel type composition, which in turn correlates with the industrial property types adjacent to those ports.
A warehouse acquisition model that leans on throughput delta as a demand predictor needs to know whether that delta reflects actual cargo change or a shift in which hydrostatic method fired.
What to Watch
The cleanest leading indicator of throughput signal quality is the trim_saline_corrected penetration rate by port. At the current global average of 1.24%, most aggregated throughput is fallback-derived or zeroed. Ports where that rate is above 5% are producing measurement-backed cargo data; those below 1% are almost entirely running on population models.
A port-level hydrostatic method distribution has not been surfaced as a standard data quality flag in cell-score output. Adding that flag — even as a confidence band on development_pipeline for port-adjacent cells — would let market selection models adjust for what the underlying throughput data actually represents.
Limitations
The connection between hydrostatic method quality and CRE demand-signal error is indirect. The fallback overestimates bulk carrier cargo relative to the saline-corrected measurement, but that does not mean every port with high fallback rates is misrepresenting its actual throughput. The fallback fires on vessels without complete draft records; those vessels may have genuinely different load profiles from vessels with complete records. Market-level throughput aggregation across many visits may partially cancel systematic biases. None of the cell-score development_pipeline values have been traced directly to hydrostatic method composition as a confirmed cause of scoring error.
Data as of 2026-06-09. Sources: vessel_visits (60-day window, n=41,377), cell_scores, port_water_density_cells.