On the afternoon of March 29, 2025, the pilot of a Socata TBM-700 (N721MB) was fatally injured in an accident while on approach to Anoka County/Blaine Airport (Janes Field)(KANE). According to FlightAware (below), the flight departed Des Moines International Airport (KDSM) at 11:12 a.m. CDT and it was headed to Minneapolis. Local news reported that the crash occurred near Brooklyn Park, Minnesota around 12:22 p.m. CDT (1722Z). The aircraft crashed into a home and all occupants of the home were able to escape. A news video of the crash shows the aircraft plummeting to Earth in almost a vertical orientation suggesting a loss of control (screenshot shown below). This was similar to a King Air E90 fatal accident in 2022 that is described here.

The NTSB has visited the site and has issued the preliminary report that can be found here. They will likely issue their final report in about 18 months from the date of the accident. The only weather information offered in the preliminary report was the following:

ANE weather reported at 1149, wind from 030° at 12 knots, visibility 10 miles or greater, overcast ceiling 900 ft above ground level, temperature of 3°C, dewpoint of 2°C, and an altimeter setting of 29.92 inches of mercury. An Aviation Weather Center icing probability chart indicated that at 1230, the icing probability at 3,000 ft msl was at or above 50-60% in the area over the accident flight path and that a Graphical Airmen’s Meteorological Information for moderate icing and instrument flight rule conditions were active.

There are a lot of potential factors to cause such an accident that may or may not include weather. However, as will be discussed below, potential icing conditions were indeed present in the area and this accident may have been associated with those icing conditions. Please understand that this is pure speculation and may not be the probable cause of this crash.

While the TBM 700 is generally thought not to be an airplane susceptible to tailplane icing, it is still important to discuss this. Kurt Blankenship who is a research pilot and deputy of aircraft operations at the Glenn Research Center in Cleveland has stated, "that when the tailplane is accreting ice with flaps extended, even a small change to the angle of attack can initiate the stall on some aircraft." He goes on to say, "something as simple as changing the power can induce this stall especially on the high end of the flap speed." Recovery from a tailplane stall is completely opposite a normal wing stall. A tailplane stall can be very violent with the controls quite literally snatched from your hands. Often it takes an enormous amount of back pressure on the controls to recover...sometimes even requiring that the copilot assist in this recovery. Most instances of ice contaminated tailplane stalls occur on approach to land while low to the ground where recovery is near impossible.

Please note that this discussion is for educational and entertainment purposes only and may contain errors and omissions. If you like what you read here, please visit https://ezwxbrief.com for the best source online source of aviation weather and subscribe free for 30 days. If you are looking to learn more about reading a Skew-T log (p) diagram, please considering purchasing the most comprehensive text available to pilots, Weather Essentials for Pilots: The Skew-T Edition available in eBook only. Also, you can purchase the recordings to our live Weather Essentials and Advanced Weather Essentials for Pilots and the popular Skew-T Weather Essentials classes to learn how to minimize your exposure to adverse weather.

According to the FlightAware log, the pilot climbed to the cruising altitude of 25,000 feet MSL after departing Des Moines. The aircraft descended to 3,000 feet and remained at that altitude until established on final approach for a total of 15 minutes at 3,000 feet based on the FlightAware track log. The routine observation issued at 1800Z reported northeast winds at 11 knots, 10 statute miles surface visibility with an overcast ceiling of 1,100 feet as shown below.

KANE 291800Z 03011KT 10SM OVC011 04/02 A2992=

Given the marginal VFR (MVFR) and IFR conditions around the Minneapolis-St. Paul metropolitan area, the aircraft was vectored to the RNAV to Runway 09 approach at KANE based on the LiveATC.net archive and the NTSB preliminary report. The pilot calmly checked in with Anoka tower and was cleared to land by the tower controller. The pilot acknowledged the landing clearance. Three minutes and 46 seconds later, the tower controller issued a low-altitude alert. There were no further transmissions received from the pilot.

As shown below from the 1800Z surface analysis, a stationary front was analyzed to extend from eastern Michigan through central Iowa (red circle represents the approximate location of the accident site). The flight occurred entirely on the cold side of this stationary front. Surface winds were from the east and northeast over most of the area north of the front. This brings in cold, dense air near the surface with surface temperatures that varied from the low 30s to 40°F.

There was light rain and drizzle reported throughout the Minneapolis area based on nearby surface observations and visible on the 1725Z NEXRAD mosaic below. Therefore, an approach from the south likely exposed the aircraft to some of this light precipitation.

Given the widespread IFR and marginal VFR conditions throughout the area, the Aviation Weather Center issued this IFR G-AIRMET (below) valid at 18Z.

On this particular afternoon, there were multiple freezing levels in the area. Near the destination there were three freezing levels that included 720 mb (~9,500 ft MSL), 875 mb (~4,500 ft MSL) and 960 mb (~2,000 ft MSL). This can be seen on the forecast sounding from the 3-km North American Mesoscale (NAM) 12Z model run valid at 17Z.

A closer examination below shows a fairly deep elevated temperature inversion from 925 mb (~2,500 feet MSL) to 800 mb (~6,000 feet MSL). The air is saturated from the base of this elevated inversion to 825 mb (~5,500 feet MSL). This creates a cloud depth of ~3,200 feet. This is not uncommon for a deep stratocumulus cloud deck. The lapse rate from the surface to the base of the saturated layer is dry adiabatic. This gives rise to some low level instability as evidence by the surface-based parcel lapse rate (dashed purple line) which is to the right of the environmental temperature in the lower part of this shallow layer.

The base of the saturated layer is 840 mb or approximately 2,300 feet MSL. Given a field elevation of 912 feet, this equates to 1,400 AGL. This is consistent with the 1800Z automated surface observation taken at KANE that shows a ceiling of 1,100 feet.

Low level winds shown in the forecast sounding are from the east to northeast bringing in cold, dense air near the surface. Above 700 mb (~10,000 feet) the winds shift to a southwesterly and westerly flow which provides a source of warmer air aloft. This gives rise to the elevated temperature inversion creating a "warm nose" that describes warm air over cold air and creates multiple freezing levels. The G-AIRMET below valid at 1800Z was issued by the Aviation Weather Center and depicts multiple freezing levels from the surface to 12,000 feet MSL.

This creates a saturated layer with environmental temperatures below 0°C extending from 960 mb (~2,000 ft MSL) to 875 mb (~4,500 ft MSL) for a depth of 2,500 feet. The coldest temperature in this layer was shown to be -3°C at 3,000 feet MSL. As mentioned above, the aircraft leveled off at 3,000 feet MSL for a total of 15 minutes before beginning a descent on the approach.

The Aviation Weather Center also issued this G-AIRMET below for moderate ice from the freezing level to 22,000 feet MSL.

From 14Z to 19Z there were only a few pilot weather reports (PIREPs) for icing in the area. Most of the reports were negative for ice or light or trace rime ice. Many of these PIREPs reported the lower cloud layer between 1,800 feet MSL to 5,500 MSL which is consistent with the forecast sounding above.

UA /OV MSP255025/TM 1442/FL052/TP S22T/SK TOP052 LYR ABV/IC NEG ICE/RM DURC

UA /OV MSP270005/TM 1420/FLDURC/TP B738/SK OVC016-TOP047/IC NEG ICE

UA /OV FOD/TM 1405/FL025/TP CRJ2/SK BASE025/TB NEG/IC NEG

UA /OV FSD190004/TM 1423/FL026/TP A319/SK OVC026-TOP046 LYRS ABV/RM DURC

UA /OV MSP120015/TM 1553/FL030/TP PAY1/TA 10 @ 070 01 @ 030/IC NEG/RM DURD

UA /OV CFE/TM 1526/FL055/TP SR22/SK BASE018-TOP055/TA M05/IC TRACE RIME

UA /OV MSP310025/TM 1804/FL030/TP SR22/IC LGT RME OAT M02

Even though there are not many reports of airframe ice, understand that icing is very transitory. There was an accident of an EMB-120RT near Monroe, Michigan that illustrates this very well. The following can be found on this NASA site.

MONROE, MICHIGAN JANUARY 9, 1997 EMB-120RT Comair Flight 3272

The aircraft was being vectored for an ILS approach to Detroit (DTW). The pilots slowed the aircraft to 150 knots while the aircraft leveled at 4,000 feet. While turning to the assigned heading the aircraft experienced an uncommanded roll and the autopilot disconnected.

Less than two seconds after the autopilot disconnected the flight data recorder indicated roll attitude increased from about 45 degrees left bank to about 140 degrees left bank and the pitch attitude decreased from nearly 2 degrees nose up to about 17 degrees nose down.

The control wheel position moved from about 18 degrees right to 19 degrees left. The aircraft struck the ground in a steep nose-down attitude about 19 nautical miles southwest of the airport.

Evidence indicates that the aircraft was picking up ice at a significant rate for less than a minute and the total accumulation was between 1/4 to 1/2 inch of ice.

Examine the following two PIREPs...

DTW UA/OV DTW 230025/TM 1610/FL050/TP DC3/IC MDT MIXED (Local time - 16:10 UTC - 21:10)

DTW UA/OV DTW 240021/TM 1738/FL040/TPC310/TA-10/TB LGT OCNL MDT CHOP/IC NEG/RM BTWN LYRS (Local time - 17:38 UTC - 22:38)

These two PIREPs illustrate the transitory nature of icing.

At approximately 20:54 only a few miles away over Monroe, Michigan an EMB-120RT crashed during a rapid descent after an uncommanded roll. It is likely that the aircraft accumulated clear-mixed ice on the leading edge deicing boots, possibly with an ice ridge forming on the upper surface leading edge, as the airplane descended from 7,000 feet to 4,000 feet in icing conditions.

The NTSB developed a questionnaire to gather more information about the weather conditions at the time of the Monroe, Michigan accident. Pilots operating in the area at the time of the accident reported the following:

About 10 minutes ahead of flight 3272, a pilot reported light rime icing while on approach.

About 5 minutes ahead of flight 3272, a pilot observed no icing on the approach.

About 1 minute ahead of flight 3272, a pilot reported moderate rime icing.

About 2 minutes behind flight 3272 a pilot reported moderate-to-severe icing.

According to icing expert, Ben Bernstein who did the detailed weather analysis for this accident...

"The pilots reported that the icing was “not present,” “light,” “moderate,” and “extremely heavy to severe.” The apparent disagreement was attributed to small-scale variations in the clouds and how the aircraft traversed them. This example demonstrates that small variations in location, time, and approach can be the difference between no icing and icing."

At 6:39 a.m. CDT, the weather forecast office out of the NWS Twin Cities/Chanhassen, Minnesota issued the following area forecast discussion (AFD). This was in support of the issuance of the 12Z TAFs and appeared in the aviation section.

"Biggest change with the 12z TAFs is that we decided to lean

much harder into the GFS/RAP/HRRR for p-types. This means we

backed off pretty hard on the fzra mention, instead going with

a snow/sleet mix for when mixed p-types are expected."

This precipitation type discussion above was describing the weather expected in the overnight hours of Saturday into Sunday (well past the time the TBM 700 was en route). The forecast below was the one issued for MSP prior to the 12Z issuance. Notice the mention of mixed precipitation of -FZRAPL beginning at 06Z on March 30th.

KMSP 290901Z 2909/3012 01012KT P6SM OVC014

FM291500 03010KT P6SM OVC011

PROB30 2917/2920 4SM -RA OVC007

FM292000 04012G22KT 4SM -RA BR OVC010

FM292300 04012G24KT 3SM -RA BR OVC008

FM300600 04014G23KT 4SM -FZRAPL BR OVC007

FM301000 03013G22KT 2SM -SN BR OVC008=

As described in the AFD, the 12Z issuance of the TAF for MSP no longer included a forecast for FZRA.

KMSP 291138Z 2912/3018 02009KT 4SM BR OVC008

FM291600 04011G20KT P6SM OVC009

FM291900 05012G21KT 6SM -RA BR OVC010

FM292200 05013G22KT 3SM -RA BR OVC008

FM300000 04013G22KT 2SM -RA BR OVC007

FM301100 03012G20KT 3SM -SNPL BR OVC007

FM301500 02013G23KT 3SM -SN BR OVC009=

The AFD continued with...

"We didn`t change precip timing much, though the 12z MPX sounding

coming in does support what we`re seeing in the forecast soundings

with a deep low level saturated atmosphere below a strong inversion

that peaks at 825mb. This type of sounding can support

drizzle/mist generation and that is in fact what we`re seeing

this morning. As we warm some this morning, we expected the

drizzle to end before steady rain arrives this afternoon."

This part of the discussion is highly relevant at the time of the flight. That is, a deep, low-level saturated layer with a strong (elevated) inversion was present. They explain that soundings such as this are conducive for the generation of drizzle/mist. And they further emphasize that's exactly what they were seeing in the observations.

There were plenty of surface observations in the area that reported rain or drizzle. Specifically, 13 minutes prior to the accident the automated weather at KFBL shown below reported moderate drizzle. This airport is south of Minneapolis and just to the east of the flight path. The aircraft was descending through 5,800 feet upon passing by this airport and the temperature was likely +1°C. The aircraft was in a descent to 4,000 feet based on the FlightAware track. Drizzle-sized drops can produce areas of supercooled LARGE drop (SLD) icing.

KFBL 291735Z AUTO 03007KT 10SM DZ OVC006 03/02 A2987 RMK AO2 LTG DSNT SW

The 12Z radiosonde observation (RAOB) they are referencing is shown below. This depicts a very similar temperature, dewpoint and wind profile to the NAM forecast sounding shown earlier. This implies that the cold air near the surface was locked in place and did not vary much throughout the morning. It also verifies that the forecast sounding was likely very representative of the environment.

The icing guidance found on the Aviation Weather Center (AWC) website shown below for 3,000 feet MSL does show light to moderate ice (blue) with areas of SLD (red). This is likely driven by the liquid precipitation falling throughout the area into a region of air that is below 0°C.

Given the analysis above, it's reasonable to assume that airframe icing may have played a role in this accident. In this situation the pilot descended from very cold conditions at FL250 to temperatures well above freezing (+6°C) due to the elevated temperature inversion as evidenced by the forecast sounding and nearby RAOB. With a surface temperature of +3°C based on the surface observation, many pilots assume that the entire temperature profile down to the surface is ABOVE freezing and therefore there is no risk of icing. At this point, the pilot may have not been looking for ice accretions while on approach to land.

However, it's also important to recognize that with environmental temperatures only a couple degrees below freezing, kinetic heating will most certainly keep ice from accreting on the immediate leading edges of the wings and horizontal stabilizers. If the atmosphere was a few degrees colder than forecast, it's conceivable that some ice may have accreted on the wings and/or tail and could lead to a stall especially if there was a failure of the ice protection system or the ice protection system was not used properly. Additionally, even if the boots were cycled properly, runback ice (clear ice accreting behind the protected area) can lead to a sharp reduction in airspeed and increase the stall speed of the aircraft. This would be common if the aircraft is flown into freezing rain/drizzle. Based on the observations, rain and drizzle was falling throughout the area. This would give rise to a freezing rain/drizzle potential at 3,000 feet MSL where the pilot spent about 15 minutes of flight time at a temperature below 0°C. Then if flaps were extended, this would lead to an increase in the angle of attack and lead to an abrupt stall.

Most pilots are weatherwise, but some are otherwise™

Dr. Scott Dennstaedt

Weather Systems Engineer

Founder, EZWxBrief™

CFI & former NWS meteorologist