4.3 Forest management using DEMs

Elevation is an integral part of many environmental management applications. To illustrate this with practical examples, we will investigate here how Digital Elevation Models (DEMs) can be used to plan forestry operations. There are many operations undertaken in managing forest stands. Here, we will focus on planning tree planting, managing fire risks, and felling.

How suitable is land for commercial forestry?

One of the main constraints on land suitability for forestry in western Europe, western Canada, and New Zealand is the windiness of the climate. In windier sites, there is a greater chance of trees being blown over in storms. These wind-damaged trees generally have less value as timber than trees that have been felled using specialist machinery. Trees also tend to grow less quickly in windy sites than in sheltered sites, so windiness also reduces the commercial potential of a site for plantation forestry by inhibiting tree growth.

Wind speeds are expensive to measure directly. The equipment for measuring wind, known as an anemometer, is expensive relative to (say) a rain-gauge. An anemometer runs the risk of being stolen or hit by lightning strikes in the remote, upland areas where forestry normally takes place. For this reason, instead of measuring windiness directly, an indirect measure of windiness known as topographic exposure (see Box 1) is often used in forest management instead. Topographic exposure is a measure that can be derived from a DEM that distinguishes sheltered valley bottoms from open plains and exposed ridges.

A DEM can therefore be used to estimate the potential of a site for commercial forestry by calculating the risk of wind damage from altitude and topographic exposure (Blennow and Sallnas, 2004).

Aside from this, LiDAR technology enables the generation of Digital Surface Models (DSMs, which include canopy heights from ‘first return’ LiDAR pulses) as well as ‘bare earth’ Digital Terrain Models. Such DSMs enable individual tree heights to be estimated (Suarez et al, 2005). Previously, these would have been estimated through fieldwork at sample plots, with tree height being measured in the field alongside other timber metrics, such as diameters of tree stems. LiDAR provides a means of monitoring tree growth following planting. However, GIS has also been used to predict potential tree growth prior to planting, drawing on DEM-derived site variables such as aspect, topographic exposure, and altitude itself (Bateman and Lovett, 1998).

How suitable is land for replanting native species?

In many countries, the nature of forestry is changing and the management of forests for ecological and conservation purposes, rather than for commercial timber production, is becoming increasingly important. In the UK , most of the indigenous forest cover has been felled and government schemes have been introduced to help replant such deforested areas with native species. In the absence of human disturbance, the different native tree species occur on different types of site. When planting seedlings of native tree species, seedlings of different species need to be matched to the type of sites on which they would occur naturally. DEMs can be used to help decide which native tree species should be replanted on which sites. One system which matches native tree species to site conditions in this way is the Ecological Site Classification (Ray et al, 1998). This system uses a DEM to calculate windiness, which is then used to decide which species should be planted on high altitude ridges and which species in low altitude valley bottoms.

Planning felling operations

Foresters not only need to plant trees, but in the case of commercial plantations, they also need to fell trees for timber too. DEMs are also useful for planning felling operations:

  • Choice of machinery: The choice of machinery used to extract timber often depends on terrain. For example, on very steep slopes, a cable crane is often used to move timber. The cable crane is a vehicle to which an overhead cable is attached. This cable can then be used to move timber downhill without excessive soil erosion occurring. In contrast, on flatter ground, a forwarder (a tractor with a grab for seizing logs) is typically used. DEMs can be used to calculate gradients, which can then be used to choose machinery for harvesting timber.
  • Visual impact of felling: Large-scale felling of plantation forests has a major visual impact on the landscape, particularly where nearby residents have become accustomed to forest stands as part of their environment. The visualisation capabilities of GIS can be used in conjunction with DEMs to assess this visual impact before any felling operations take place (Wing and Johnson, 2001).

Managing the fire risk to forestry

In many parts of the world, another risk to timber stands comes from fire. Terrain affects fire dynamics in a complex manner, since slope may potentially affect the speeds with which a fire passes through an area. Slope and aspect affect the moisture content of leaf litter available as fuel for the fire and steeper slopes are likely to be less accessible to firefighters in the event that a fire does take place. As a consequence, several software tools (e.g. FARSITE – see below) have emerged for assessing fire risk and modelling the spread of specific, individual fires.


If you are using ArcGIS Desktop, download this zip file and undertake the GIS exercise described in the pdf file. This exercise involves assessing the fire risk for forest stands in the USA, based on slopes derived from a DEM.  If you are using ArcGIS Pro, download this zip file and undertake the exercise described in the pdf file.  This involves modelling hydrological flows across a landscape.

References (Essential reading for this learning object indicated by *)

Bateman, I., and Lovett, A. (1998) Using geographical information systems (GIS) and large area databases to predict Yield Class: a study of Sitka spruce in Wales. Forestry 71 (2), 147-168.

Blennow, K., and Sallnas, O. (2004) WINDA – a system of models for assessing the probability of wind damage to forest stands within a landscape. Ecological Modelling 175 (1), 87-99

Ray, D., Reynolds, K., Slade, J., and Hodge, S. (1998) A spatial solution to Ecological Site Classification for British Forestry using Ecosystem Management Decision Support. Proceedings of the GeoComputation 1998 Conference, Otago, New Zealand http://www.geocomputation.org/1998/37/gc_37.htm

Wing, M. G., and Johnson, R. (2001) Quantifying forest visibility with spatial data. Environmental Management 27 (3), 411-420.

For an example of GIS being used to assess the visual impact of forestry, see this link on the ESRI web site: http://gis.esri.com/library/userconf/proc97/proc97/to250/pap202/p202.htm

The ArcGIS Desktop practical is based (with permission) on the following references:

Price, M. (2003) Modelling the wild land/urban interface. ArcUser Magazine (April – June), 46 – 50. http://www.esri.com/news/arcuser/

United States Department of Agriculture – Rocky Mountain Research Station (2001) Assessing Crown Fire Potential by Linking Models of Surface and Crown Fire behaviour. Available online at: http://www.fs.fed.us/rm/pubs/rmrs_rp029.pdf

For details of a more sophisticated spatial fire model than that covered here in the practical, see the FARSITE software: https://www.firelab.org/project/farsite

Suarez, J., Ontiveros, C., Smith, S., and Snape, S. (2005) Use of airborne LiDAR and aerial photography in the estimation of individual tree heights in forestry. Computers and GeoSciences 31 (2), 253-262.

The ArcGIS Pro practical makes use of Environment Agency lidar data: https://environment.data.gov.uk/DefraDataDownload/?Mode=survey.  It examines how we can use such data to manage chalk streams (see https://www.wwf.org.uk/where-we-work/uk-rivers-and-chalk-streams). Some of the tools in it can be used to produce input parameters for the Universal Soil Loss Equation: https://milford.nserl.purdue.edu/weppdocs/overview/usle.html.

Comments are closed.