Say no to maps? How to communicate volcanic hazard effectively on Montserrat



K Haynes, J Barclay and N Pidgeon, Volcanic hazard communication using maps: an evaluation of their effectiveness. Bulletin of Volcanology (2007) 70: 123-138.




When professional earth scientists, physical geographers and the like perceive a volcanic or other hazard (such as likely flooding) to an area, their first instinct is to communicate this information, by means of a map, to the residents who might be affected. This approach is drilled into these sorts of scientists from the first days of their training, as they learn both to interpret and eventually to make their own appropriate maps. Go to a conference about an impending hazard or recent disaster anywhere and you will be deluged with maps of various types. After all, we all know that this is the normal way to tell people essential things like the boundary of the Exclusion Zone in southern Montserrat ……. don’t we?


On the other hand, when scientists move outside the cosy world of “professionals” and into the endlessly varied residents of the real world and its various hazard zones, they all know that they will be dealing with innumerable people who cannot “read” conventional maps fluently, to say the least. Yet professionals persist in using maps to communicate with the public. Furthermore, there is remarkably little formal research into the extent to which this approach succeeds or fails. This is the gap that Haynes, Barclay and Pidgeon began to fill for volcanic hazards, by surveying the success of a sample of Montserrat residents at various tasks, using both conventional topographic maps of the island (Fig. 1) and two modern alternatives – digital 3-D views (Fig. 3A-D) and air photographs (especially oblique ones).


The people to be interviewed were selected to represent the things that vary in all populations: age, gender, education etc. One important restriction was that all of them had lived on the island since before June 1997. This is because many of the questions were concerned with the south of the island and there has been very little access there since that time. The interviewers then asked these long-suffering souls to locate all manner of things like north, south, east, west and various villages, ghauts, hills etc on a conventional topographic map, the 3-D “maps” and the air photos. From the interviewer’s viewpoint, the questions were designed to test people in four slightly overlapping fields:

1.  Orientation. Where are north, south, east and west?

2.  Locating. Where’s my home? Where are we now?

3.  Map skills. How do I get from A to B, if there’s a flood in ***** Ghaut?

4.  “Volcanic information. Where are the dangerous places ……. and why?



The contours problem


“Map skills” hits the very raw nerve of contours in many people. When trying to teach a group of people (young or adult) about maps and how to use them, it is notorious that some in the group will be completely defeated by contours. Patience, with a bit of string and a spirit level, will usually get the basic idea of a contour into most people. But the business of zigzags in contours on a hillside denoting ridges and gullies is extremely difficult to get across to those who happen to lack the right sort of “3-D brain”. Thus contours allow effortless 3-D visualisation of a map by those with the right aptitude and training, whilst remaining an impenetrable mystery to many others (including of course some Montserrat residents).


This problem of “reading” contours greatly affected the results of tests set during the interviews. When using a contour map, a few people could accurately place such locations as Plymouth, Harris, Belham Valley, Tar River Valley and St George’s Hill. But the majority scattered their guesses about these locations like confetti across most of the island (Fig. 2).



Fig 1  The contour map of Montserrat used in the interviews. Contours are dark grey; roads are purple; exclusion zone boundaries are thick black; pyroclastic flow deposits (and Boxing Day 1997 debris avalanche) are red; surge deposits are yellow; lahar deposits are brown.

Fig. 2  Interviewees’ attempts at locating places using the contour map.


Alternatives to contour maps


Two alternative types of “map” reduced the location problem substantially; a point tested by statistical analysis:

1.  3-D perspective “views” of the island from various angles, created from a digital model with features such as the new volcanic deposits and the current Exclusion Zone added (Fig. 3A-D).

2.  Both vertical and (especially) oblique air photographs (Fig. 4).


Fig. 3  The 3-D perspective views used in the interviews.


A  Up Belham Valley from the coast.



B  Hovering offshore from Plymouth and looking inland.



C  Hovering offshore from the former W H Bramble Airport and looking inland.



D  Hovering offshore from Tar River Valley and looking inland.


Fig. 4A  Oblique aerial photo of eastern Montserrat used in the interviews.

Fig. 4B Oblique aerial photo of Belham Valley used in the interviews.


Fig. 5  Interviewees’ attempts at locating places using aerial photos



In contradiction to some earlier published research, Haynes, Barclay and Pidgeon could find no gender-linked differences in the success of people with the location tests. The only difference found between men and women was that some of the men, but no women, tended to “lock” onto some distinctive feature of the island, such as Bransby Point or the volcano, and use this to orient themselves and thus find target places. Actually Montserrat is a dream place to orient yourself because the long axis of the pear/mango-shaped island is nearly north-south and the sharp wedge of Bransby Point almost exactly points west to the setting sun.


This research reaches one very clear negative conclusion. Traditional coloured contoured maps are not a very smart way to communicate hazard concepts, exclusion zones etc to the residents of Montserrat, although other published research shows that the residents in other parts of the world prefer maps to alternatives. Although law-abiding people will obey the line marking an exclusion zone on a map, if they lack the ability to “read” contours, they will see no logical reason why the line is drawn as it is. Either 3-D digital “landscapes” of various types (and degrees of sophistication) should be used or (maybe simpler and therefore better) aerial photographs from either satellites or aircraft. Non-vertical photos are particularly useful.



Montserrat Volcano Observatory

Click here to reach page 2 of this topic