The slippery slope: CAD hachures

The depiction of slope by hachure shading is a cartographic convention that dates back to the Austrian cavalry officer Johann Georg Lehmann who, in 1799, devised a system for the rapid military mapping of terrain by field drawing. It has proved to be an enduring idea and has persisted even though more precise methods of depicting height information have overtaken the field sketch. It can’t be beaten as a short hand for linear features. Where a solid line could be interpreted as hard detail a hachure has a softness that suits earthworks well, when we see a ditch and bank there is raely a sharp dividing line visible between rise and fall of the bank. What we see is a shadow line and that’s just what the hachure aludes to: stronger than a plain dashed or dotted line the hatched shading indicates a recognisable landform as a scaled symbol. Contouring may be more useful but understanding a contoured plan can take a little time compared to the immediate impact of the ticks and lines that shout: ‘here is a slope’ or ‘there is a ditch’ from the plan.

P1020902The shadow of ditch and bank can tell more than a trace of its maxima and minima.

L1322591A very tight contour interval would be needed to pick up all the fine detail here. The firing step is just visible in the photo but almost impossible to detect on the ground.

Who uses hachures today? Hachured plans were considered the definitive site survey record for the now defunct RCHME. The technique is still recommended by English Heritage (‘Understanding the Archeology of Landscapes: a guide to good recording practice’ English Heritage  2007) as good practice as the landform depiction method of choice to support ‘Level 1’ through ‘Level 3’ archaeological records.

At no point does the EH guidance refer to Lehmanns rules for hachures and the guidance is concerned with graphic interpretation of landform rather than a system to depict precise slope metrics. The cartographic principles of depicting slope angle, extent and height are largely ignored in a clear directive to use hachures as an interpretive tool rather than a mapping convention.

Faced with providing a survey for a ‘Level 3’ record and being qualified in both survey and cartography I’m baffled by the lack of a clear hachure convention, particularly in an age when we have access to accurate height information as never before, why on earth should slope depiction be devoid of measured height information: is this method of landscape mapping valid in the modern age?

Conventions for hacures in landscape archeology. English Heritage stipulate, from Level 1 to Level 3, a hachure depiction of landform. So the question is: how to draw convincing hachres in the 19th cent manner in CAD? All we get from EH is an exhortation that:

‘hachures can be used instead of contours where necessary to emphasise or precisely depict the shape of a particular slope’

and the direction that, natural slopes are:

‘drawn with thin heads and broken, wavy tails to distinguish them from hachures representing artificial slopes’

we are referred back to ‘With Alidade and Tape: Graphical & plane table survey of archeological earthworks’ English Heritage 2002. The size of the head is used to denote steepness and the absence of a head (Item J ‘drawn with thin heads and broken, wavy tails’ below) denotes a natural slope:

Bowden 2002

A clear hachure convention from Archaeological Illustration by Lesley Adkins and Roy A. Adkins (1989, 67) uses the spacing of the lines to indicate steepness and the roundness of the head to show sharpness:

Slope depiction_1989 there is no convention for indication of the Z  value despite the direction to scale these symbols along the plan distance of the slope. The capacity to indicate concave and convex slope is possible using a combination of the ‘type of slope’ and ‘changes in slope’ methods.

Surveying for a Level 3 record. Looking at the case studies provided ‘where necessary’ seems to be in all cases where a survey drawing is needed. Given a ‘Level 3’ record requires 1:1000 scale survey with a precision of c0.10m how much survey do I need to do to get the terrain hachured as required?

As far as I can tell the measurements needed for the site plan are 2D shots for:

  • top of slope
  • change of slope
  • bottom of slope
  • Break lines

I could use a grid of 3D data to derive the 2D result and, given how busy some of these plans can be, capture of  a full DTM  would be a wise move. A 3D surface model will be able to be interrogated in CAD to produce the required lines as and when the experts feed me with opinion on the historic features they want to emphasise.

Hachure gives a very good impression of a site interpretation, making what ‘the eye of faith’ sees more real than a DTM or a contour plot of the terrain:

EH Norrington hachuresAn extract from ‘Case Study 8’ (1) shows intensive use of the round headed ‘gradual slope’ hachure, even the sharp slope depiction uses a blob and streak approach to the line which softens the breaklines:

EH Norrington hachures_02These are hand drawn and the push and pull of the pen lines gives a subtle softness to the work absent from CAD examples.

The same method by CAD is shown in Case Study 9:

Blackbeck_01The CAD version lacks the softness of the hand drawn. The chisel headed ticks are sharp edged and the eye is drawn to the mismatch of the edges which is masked in the round head of the hand drawn example. Close inspection reveals the effect of the sharp headed lines adding together to give a jagged impression of the line of the top of the slope :

Blackbeck_02in this respect CAD is very unforgiving, getting variable line thickness is possible but softening the edges of lines is almost impossible. Objects in CAD just don’t have indistinct boundaries the way a hand drawn mark can.

So how to get CAD to produce the same graphic effect?

I developed a technique of hachure generation in CAD which works well but has the failing of the heads being sharp:

Chartley w hachuresThese were propagated by using a variable width polyline traced at right angles to the slope indicated by the contours:

Chartley w contoursThey are quick to do but lack the round headed softness of the pen and ink version.

A polyline can be produced with a variable thickness to emulate a simple stretch triangle. It works well for embankments.

Chartley outer bailey w hachuresWorking up a CAD hachure plan is a curious case of emulating a somewhat poorly defined draughting convention to achieve a cartographic end which is lacking in objective metrics other than being constrained in plan. I have very little idea of how the line separation  relates to scale, the density and intensity of the hachure line is subjective, almost to the point of being personal,  as it is entirely interpretive. In thematic mapping much use is made of hatch, shade and text to bring emphasis to the desired interpretation: hachured plans for landscape archeology are no exception, we are presented with an interpretation of the terrain rather than a record.

Chartley outer bailey w contoursTo me the contour model tells me so much more than ‘one mans hachure’ wooly worm plan. Sure the contours are interpolated but I can read this terrain with confidence in its consistency and completeness knowing the degree of objectivity and repeatability in its method. I can trust it to relate to the landscape as it feels under my boot rather than as seen in the minds eye of an interpreter!

(1) The case studies are taken from ‘Understanding the Archeology of Landscapes: a guide to good recording practice’ English Heritage  2007 Product Code 51320

The hachure convention for English Heritage Archeological Survey is published in ‘With Alidade and Tape: Graphical & plane table survey of archeological earthworks’ English Heriatge 2002. Product code 50692

About billboyheritagesurvey

Heritage worker
Gallery | This entry was posted in Survey Practice and tagged , , , , , . Bookmark the permalink.

One Response to The slippery slope: CAD hachures

  1. Thanks for nice sharing..!!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s