Two Derailments in 36 Hours Choke Major U.S. Freight Corridors

New sensor data from RailState reveals a 73.1% container volume drop on a key transcontinental route.

Overview

In the span of 36 hours this week, two Union Pacific freight trains derailed on major corridors: one hauling hazmat through suburban Houston, the other scattering intermodal containers across the Southern California desert. RailState can now measure the current impact of these derailments on freight flows at both locations.

What Happened

On the morning of March 18, 23 cars of a manifest train carrying ethanol, LPG, and other hazardous materials derailed in Richmond, Texas, about 30 miles southwest of Houston. The next evening, roughly 20 cars of an intermodal train derailed in Mecca, California, on UP's Sunset Route, one of the railroad's primary transcontinental corridors, sending approximately 40 shipping containers tumbling across the desert floor. No injuries were reported at either site.

Both incidents shut down critical freight arteries. But measuring exactly how much traffic was disrupted, and how quickly, has traditionally been impossible without waiting weeks for railroad self-reported data. RailState's sensor network makes this data available in hours instead of weeks.

RailState operates a growing network of wayside sensors positioned along rail corridors across North America. These sensors automatically detect and classify every passing train, recording train type, individual car types, hazmat placards, container details, speed, and direction of travel. The result is a continuous, train-by-train, car-by-car record of freight movement at specific geographic points.


Richmond, TX
Union Pacific • UP • Disruption from 2026-03-18
Richmond, TX

RailState Capturing Train Details During the Derailment

At Richmond, the RailState sensor sits outside of the rail right of way, just 0.6 miles from the derailment site. It was actively capturing data on this very train at the moment the cars left the rails.

The sensor began reading the westbound manifest train at 4:49 AM Central Time on March 18. The reading ended abruptly at 4:51 AM, just under two minutes in, when the derailment interrupted the train's passage. Before the disruption cut the reading short, RailState had already captured a detailed car-by-car inventory of the train's consist — a real-time manifest of exactly what was on the tracks when the incident occurred.

The analysis for this location splits March 18 at the approximate derailment time, comparing post-derailment traffic (March 18 after 4 AM CT through March 20) against the pre-disruption 7-day baseline (March 10-16).

The Derailed Train — Captured in Real Time

RailState's Richmond sensor was actively capturing this train at the exact time the derailment occurred. The sensor began reading the westbound manifest at March 18, 2026 at 4:49 AM Central Time and recorded the train traveling at 27.5 mph. The reading was cut short at 4:51 AM CT when the derailment interrupted the train's passage past the sensor. The derailment was reported to authorities at approximately 5:00 AM — minutes later.

Despite the interrupted reading, RailState captured detailed data on the majority of the consist, providing a car-by-car inventory of what was on the tracks at the moment of the incident: 3 locomotives and 85 freight cars, including 64 tank cars, 9 covered hoppers, 7 vehicular flatcars, and 5 box cars.

Train Consist Diagram

Each block represents one car, colored by type and hazmat placard. Scroll horizontally to see the full train.

L
L
L
B
B
H
H
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
B
F
F
F
F
F
F
F
H
H
B
B
H
H
H
H
H
Locomotive Ethanol (UN1170) LPG (UN1075) Phosphoric Acid (UN1805) Hexane (UN1208) Other Hazmat Tank Car (no placard) Covered Hopper Box Car Vehicular Flatcar

Hazardous Materials Placards

Of the 64 tank cars captured, 54 displayed hazardous materials placards — a significant proportion of the train was carrying regulated commodities. The placards recorded by RailState's sensors detail exactly what was on the rails at the time of the incident:

UN PlacardCommodityCarsShare of Hazmat
UN1075Liquefied Petroleum Gas (LPG)3056%
UN1170Ethanol1528%
UN1805Phosphoric Acid59%
UN1208Hexane24%
UN2078Toluene Diisocyanate12%
UN3475Ethanol/Gasoline Blend12%
Total Hazmat Cars54100%
Volume Impact

Measured Disruption — Severe

Comparing the post-disruption period (Post-4AM Mar 18) against the 7-day pre-disruption baseline, RailState sensors recorded steep declines across all metrics at Richmond, TX:

7.3
Trains/Day (Disruption)
-79.1%
Baseline: 35.1/day
967
Carloads/Day (Disruption)
-77.2%
Baseline: 4247/day
Daily Volume

Traffic Trends — Last 30 Days

Red bars indicate post-disruption days. Green line is the 7-day moving average.

Breakdown

Volume by Category

Car TypeDisruption AvgBaseline AvgChange
Box Car 74.0 325.6 -77.3%
Flat Car 121.0 319.1 -62.1%
Gondola 170.3 479.0 -64.4%
Hopper 392.7 1802.0 -78.2%
Intermodal 55.0 416.9 -86.8%
Other 0.0 8.6 -100.0%
Tank Car 154.0 896.3 -82.8%
Train TypeDisruption AvgBaseline AvgChange
Automotive 0.3 1.9 -82.1%
Coal Unit 0.0 0.7 -100.0%
Grain Unit 0.7 6.3 -89.4%
Intermodal 0.7 4.0 -83.3%
Manifest 5.3 19.1 -72.1%
Passenger 0.0 0.6 -100.0%
Petroleum Unit 0.3 2.6 -87.0%
DirectionDisruption AvgBaseline AvgChange
Eastbound 2.7 15.7 -83.0%
Westbound 4.7 19.4 -76.0%

Mecca, CA
Union Pacific • UP • Disruption from 2026-03-19
Mecca, CA

Derailment on UP's Major Intermodal Route from Southern California

At Mecca, the RailState sensor sits directly on the Sunset Route mainline, 7.3 miles from the derailment site. It had captured the full profile of the intermodal train the day before the incident, providing a complete record of the containers and cars involved.

Using 90 days of continuous baseline data, RailState quantified the disruption almost immediately. The impact was stark: eastbound container volume on the Sunset Route dropped sharply from baseline levels. That's a critical artery for intermodal freight moving from Southern California ports toward destinations across the Southwest, and the derailment choked it overnight.

This kind of granular, location-specific impact measurement, available within hours of an event, simply doesn't exist anywhere else. Traditional rail data sources rely on railroad self-reporting released with multi-week delays.

The Derailed Train — RailState Profile

RailState's sensor at Mecca captured the train that would later derail as it passed the sensor location on March 18, 2026 at 3:47 PM Pacific Time — approximately 26 hours before the derailment occurred roughly 7 miles to the east. The sensor recorded the following details:

Train Type
Intermodal
UP • Eastbound
Speed at Sensor
45.7 mph
Instantaneous reading
Estimated Length
11,063 ft
2.1 miles · 2.8 min to pass

Consist Breakdown

The train consisted of 4 locomotives and 66 well cars (185 platforms), carrying a total of 398 intermodal containers:

Container TypeCountShare
40-foot Containers32982.7%
20-foot Containers6315.8%
45-foot Containers51.3%
20-foot Tank Containers10.3%
Total398100%

The train was predominantly loaded with standard 40-foot international shipping containers (83% of the consist), consistent with an import-loaded intermodal service moving eastbound from the Southern California port complex. All containers were reported to contain non-hazardous products.

Eastbound Container Volume Impact

The Mecca sensor sits on UP's Sunset Route, a key corridor for intermodal containers moving eastbound from Southern California ports and the LA Basin toward destinations across the Southwest and beyond. This derailment — which involved an intermodal train itself — has directly disrupted this critical container flow.

Eastbound Container Volume Change: -73.1% (Disruption: 1065.5/day vs Baseline: 3964.0/day)

Eastbound Containers by Type

Container TypeDisruption Daily AvgBaseline Daily AvgChange
Container 20 Feet 23.5 269.7 -91.3%
Container 40 Feet 101.0 1252.6 -91.9%
Container 45 Feet 0.0 9.1 -100.0%
Container 53 Feet 923.0 2350.6 -60.7%
Tank Container 20 Feet 0.0 2.9 -100.0%
Trailer 20 Feet 3.0 8.0 -62.5%
Trailer 53 Feet 15.0 71.1 -78.9%
Volume Impact

Measured Disruption — Severe

Comparing the post-disruption period (Since Mar 19) against the 7-day pre-disruption baseline, RailState sensors recorded steep declines across all metrics at Mecca, CA:

7.0
Trains/Day (Disruption)
-76.7%
Baseline: 30.0/day
1100
Carloads/Day (Disruption)
-79.5%
Baseline: 5354/day
1790
Containers/Day (Disruption)
-75.3%
Baseline: 7251/day
Daily Volume

Traffic Trends — Last 30 Days

Red bars indicate post-disruption days. Green line is the 7-day moving average.

Breakdown

Volume by Category

Car TypeDisruption AvgBaseline AvgChange
Box Car 11.0 259.7 -95.8%
Flat Car 34.5 486.3 -92.9%
Gondola 15.5 103.0 -85.0%
Hopper 28.5 311.6 -90.9%
Intermodal 966.5 3850.6 -74.9%
Other 3.0 10.9 -72.4%
Tank Car 41.0 331.6 -87.6%
Train TypeDisruption AvgBaseline AvgChange
Automotive 0.0 2.7 -100.0%
Intermodal 6.0 18.1 -66.9%
Manifest 1.0 8.4 -88.1%
Passenger 0.0 0.7 -100.0%
DirectionDisruption AvgBaseline AvgChange
Eastbound 4.0 13.9 -71.1%
Westbound 3.0 16.1 -81.4%

Significance

Why This Matters

For government agencies and emergency responders, RailState data provides an immediate, independent picture of what's on the tracks at a derailment site — including hazmat details — independently from railroad-provided data. When a train carrying ethanol and LPG derails near a medical center in a residential community, knowing the full consist in minutes rather than days can shape evacuation decisions and resource deployment.

For shippers and logistics operators, the ability to see a volume drop on a major corridor in near-real time means faster rerouting decisions, better customer communication, and reduced exposure to cascading delays.

For commodity and freight market analysts, RailState turns rail disruptions from anecdotal news events into quantified, data-driven signals. A sharp drop in container volume on a transcontinental corridor isn't a headline — it's a tradeable data point, available the same day.