Internet Archaeol. 23. Ruestes. Higuchi viewsheds

3.3 Higuchi viewsheds

3.3.1 Introduction

Binary viewsheds fail to accommodate the fact that features visible from a particular location are seen less clearly as distance increases (Fisher 1994, 164; Wheatley and Gillings 2000, 11). In an attempt to characterise and quantify this phenomenon, the idea of dividing viewsheds into visual ranges has been put forward. However such approaches have rarely been applied within archaeological studies.

In seeking to address the question of falling visual clarity with distance, Fisher (1994, 165) suggested the fuzzy viewshed where values are graded from 1 near the viewer point to 0 at infinity (Figure 13).

Figure 13: Fuzzy viewshed from a single viewer point (from Turó d'en Boscà hillfort)

On the other hand, Wheatley and Gillings (2000, 15-17) presented the Higuchi viewshed, which has been implemented in one published archaeological study (Trick 2004). This method classifies a visible area into three bands, making use of visual indexes proposed by Higuchi. The first band, or short distance, should equal 60 times the size of the dominant tree species in the landscape, whereas the last band, or long distance, should start at a point equivalent to 1100 times the size of the tree. These bands are given meaning in terms of perception (for detailed discussion see Wheatley and Gillings 2000).

The potential of combining the probable and fuzzy viewsheds has been highlighted (Fisher 1994, 172-72; Wheatley and Gillings 2000, 11) and that possibility was considered within this study. However, the alternative of integrating probable and Higuchi viewsheds has been deemed more suitable to address the research questions of the project. Wide bands representing well-defined distances and established according to relevant criteria can be given meaning in terms of effectiveness of visual control. Consequently, they seem more appropriate than arbitrary graded distance bands for dealing with questions of visual control of the landscape.

3.3.2 Criteria and general procedure

Higuchi viewsheds have been created taking the most probable viewshed as input and then following the procedure described by Wheatley and Gillings (2000, 16-17). (Figure 14. See also the Higuchi viewsheds for every individual hillfort: Figures 19, 23, 27, 31, 35, 39, 43 and 47).

Figure 14: Higuchi viewsheds from the hillforts of the study area

The criteria for determining the Higuchi visual ranges have been adapted to the research needs. In particular the first Higuchi band, which would have resulted in roughly 180m (3m height of a tree x60), would not have served to address visual control or social identity issues, since there are not well-known Iberian sites that close to the hillforts. There are no precise rules for assessing radii of human visibility and different criteria can be found in GIS and archaeological literature (García Sanjuán et al. 2006, 185). Rather than relying on the formula proposed by Higuchi (and developed after all in the forested landscapes of rural Japan), I have followed some of the alternative suggestions. In practice, four visual ranges have been delineated:

The first band represents 1km from a given hillfort. In this region clarity can be considered to be perfect (Fisher 1994, 165). Therefore, visual control could have been excellent.
The limit of the second visual range is 3km from a given hillfort. There seems to be some consensus about considering this distance as middle ground [note 5]. At this radius, visibility reduces but it would still be possible to see features with considerable clarity. For example, individual trees would lose clarity at distances much over 4km (Fisher 1994, 165-68). At this range it would be possible to distinguish, for example, a group of people walking along a pathway and, so, visual control would be effective.
A third band has been established up to 6km, when clarity would begin to decay significantly. This has been regarded as the maximum distance for acceptable visual control of the landscape from Iberian watchtowers (Bernabeu et al. 1987, 148). At this distance, effective visual control would depend on factors such as the individual viewer and atmospheric conditions.
Over 6km it is possible to see but not be sure what is being seen, and it is possible that the inhabitants of hillforts would have only been able to distinguish deliberate signals from other hillforts, particularly smoke signals during the day and fire signals at night (Balil 1977, 835). The maximum sightlines at which this would still be possible can be established at around 20 or 30km (Wheatley and Gillings 2000, 6; Zamora 2006, 49-50). The maximum distance of any of our viewsheds is 15.65km, well within this 20-30km threshold. This ensures that when intervisibility between two hillforts is suggested, it would always have been practically possible through the employment of deliberate signalling.

PREVIOUS NEXT CONTENTS HOME