1.  Introduction

Damages surveys are by far the best way to verify and characterize tornadoes. However, they are labour intensive, somewhat subjective, and require thorough knowledge of severe storm meteorology and at least a basic understanding of wind engineering concepts. This guide has been produced in an attempt to ensure a consistent level of knowledge among those who conduct damage surveys and those who wish to learn.

This document is a modified version of a presentation to the Toronto Weather Office in the spring of 2001. Much of the content is derived from A Guide for Conducting Damage Surveys by Bill Bunting and Brian Smith of the US National Weather Service (Bunting and Smith, 1993). This is a great resource so please refer to this publication for more detailed information. Additional ideas have come from Pat McCarthy of the Prairie and Northern Region of MSC and from my own experiences with conducting damage surveys in southern Ontario and (once!) in Sydney, Australia. Happy hunting!
 

2.  Basic Goals

The basic goals when conducting a damage survey are to:

• determine what phenomenon (tornado, downburst, derecho, flash flood) caused the reported damage,
• document the location, length, and width of the damage track as well as the times at which damage occurred,
• estimate the intensity of the phenomenon using the Fujita damage scale (Fujita, 1981) if the damage was due to wind,
• obtain information on costs, injuries and fatalities,
• find out from local residents if warnings were heard, and
• produce a report with all pertinent data and results.

3.  Preparation

So what do you need to take on a damage survey? Here are a few suggestions:

• accurate road maps, a compass, and a GPS unit,
• the relevant storm log and radar images,
• video and still cameras,
• a tape recorder,
• note paper, a clipboard, and several sharpened coloured pencils,
• a 30m+ measuring tape,
• government identification (if EC),
• proper clothing (boots, jacket, hat, sunscreen, etc.), and
• a safety kit (vests, pylons, first aid items, etc.).

4.  The Investigation

The following explains how to go about conducting the field investigation:

• at least one person with experience and training should accompany the survey team to ensure accurate results,
• a survey should be conducted as soon after the event as possible; otherwise, debris may be cleaned up and evidence may be hard to find,
• start in the area of worst damage and work outwards criss-crossing the damage track using available roads,
• use coloured pencils to mark damage vectors (downed trees, corn stalks, etc.), roads used, F-scale estimates, etc.,
• photograph / videotape damage patterns that indicate rotating winds and damage that may be used to justify F-scale ratings,
• talk to eyewitnesses about what they saw, the time that they saw it, and whether they heard warnings,
• obtain photographs or video taken during the event, and
• document any unusual occurrences and airborne missiles (particularly their approximate weight).

5.  Tornadic vs. Straight-line Winds

The following table is useful for determining whether wind damage was caused by a tornado or a downburst.
 

Characteristic	Tornado	Downburst
Aspect ratio	long and thin	short and wide
Damage gradient	high	low
Trajectories of debris	narrow and convergent	broad and divergent
Appearance of damage	chopped up, chaotic	laid out neatly
Visual cues from ground	swirls, mud splattered on walls	no swirls, no mud
Visual cues from aircraft	vortex mark, herringbone tree damage	‘starburst’ pattern with microbursts

 

6.  Pitfalls

Remembering the following may help you determine the cause of the damage:

• a fast-moving tornado may only cause damage on one side, giving the appearance of straight-line winds - click here for more details
• supercell tornadoes are often cyclic in nature resulting in a series of tornadoes along the storm track
• mutli-vortex tornadoes (having suction vortices) may produce patchy or sporadic damage
• damage may be due to a combination of downbursts and tornadoes
• downbursts can change the direction of movement of a tornado
• different types of structures / trees have different damage thresholds i.e. old house vs. new house, wood building vs. brick building, shallow-rooted vs. deep-rooted trees
• trees, signs, etc. can be ‘twisted’ due to their asymmetrical shape and this does not necessarily indicate a tornado
• eyewitnesses may report ‘swirling debris’ when they mean ‘twirling debris’ - there is a significant difference

7.  Tips

Also, keep these tips in mind during the investigation:

• try the local police station and local media outlets for information on damage as well as detailed maps of the area,
• an aerial survey may be warranted if the damage path is well defined and continuous where you are able to see it, or if the path is unreachable by road,
• when taking still pictures, always include something in the picture to indicate scale,
• pay attention to even the smallest details,
• always ask permission before entering private property, and
• the media, if present, may ask you for a verdict based on the field investigation alone - make sure to let them know that the results are preliminary and may change with additional information and analysis.

8.  Putting it all together

So you've done the field work, now what? You need to create a report to document your findings keeping the following in mind:

• it may take several days to get all the necessary evidence together i.e. photos, eyewitness accounts, etc.,
• careful synthesis of all of the available evidence is needed to make sound conclusions - be patient,
• in the report, make sure to state how you came to your conclusions i.e. why do you think it was a tornado, what evidence supports your F-scale rating, etc.?, and
• in some cases, there will not be enough evidence to come to a solid conclusion - don't force a result.

References

Bunting, W.F. and B.E. Smith, 1993: A Guide For Conducting Damage Surveys. NOAA Tech. Memo. NWS SR-146.

Fujita, T.T. 1981. Tornadoes and Downbursts in the Context of Generalized Planetary Scales. J. Atmos. Sci. 38: 1511-1534.
 

Appendix - Tornado motion and its effect on damage

Consider a tornado that is moving toward the northeast at 80 km/h. It also rotating such that the maximum tangential velocity is 80 km/h. As shown in the following diagram, the winds in the area along the right ( southeast) side of the tornado path are a vector sum of the translational and tangential velocities i.e. 80 km.h + 80 km/h = 160 km/h. Conversely, the winds in the area along the left (northwest) side of the tornado are zero since the translational and tangential velocities cancel each other there. Thus, winds on the right (southeast) side of the tornado are capable of causing damage while winds on the right side are not. The resulting damage in such as situation can resemble that from straight-line winds. However, the give-away is the narrow damage path. Also, in cases where the tangential speed is considerably higher than the translational speed, there may be very minor damage on one side of the tornado and major damage on the other. The damage pattern in this case would typically show subtle signs of rotation.