A good map is a beautiful thing. It presents complex data and ideas in informative and commutative ways. If done well, the user of a map should be able to gain insight where data alone may not have provided it. Hence, the power of representative and artistic graphic expression.
Frequently lost however in a simple map, intentionally or not, is the intricacy of the data and processes. While subscribing to and representing Saint-Exupery’s principle of perfection coming not from adding more but when there is nothing left to take away, or Japanese iki philosophies of direct and simple designs, it is important to additionally understand and recognize the core or basis of a graphic. Generally, the substance under the surface is what gives value beyond aesthetics.
Take for example the map above. I call it a map because it is geo-spatially explicit, however it may be more correct to refer to it as a diagrammatic or graphic representation, given there are no topographic or physical features shown. The reason I refer to it as a map is that all of those features (topographic, physical, geomorphic, etc.) were used in its creation. They are there, just represented differently.
At first glance, the map shows forest carbon displayed by vegetative covertype on a series of inventory points. The size of the circle indicates the amount of forest carbon in metric tons of carbon dioxide equivalency per acre (MtCO2e/ac) as derived from an inventory, and the color indicates the forest vegetative community. This representation allows the user to make inferences and gain insight regarding amount and distribution of covertypes and carbon sequestration. However, with a little additional knowledge as to how these forest covertypes were derived, further insights can be gained. Hidden in these covertypes is the complexity of the map.
The area was stratified into forest covertypes through geomorphic analysis and applying probable vegetative communities. A geomorphic approach was used given the strong correlation between landscape features and vegetative covertypes. This relationship has been well documented in scientific literature, with some such as Baily stating in 1987, “… landform provides the best means of identifying local ecosystems.” He further explained, “At the micro-scale, such patterns can be divided topographically into slope and aspects units that are relatively consistent to soil moisture regime, soil temperature regime and plat association, i.e. the homogeneous ‘site’.” Others, such as Abella concluded additionally, “Geomorphology is a dominant structuring variable affecting ecosystem distribution…” Therefore, this type of analysis provides an effective tool in ecosystem classification based on physical landscape characteristics.
The variables of interest affecting covertype that were analyzed in developing this map included aspect, slope position, and landform features such as ridge-tops and drains. To analyze these features through a GIS, a Triangulated Irregular Network, or TIN, was derived from a Digital Elevation Model (DEM). This provided a vector-based representation of irregularly distributed nodes containing three-dimensional coordinates arranged in non-overlapping triangles. These nodes were queried and symbolized for the specified variables that influence vegetative covertype. Additionally, a flow accumulation analysis was applied to the TIN to determine other important variables such as slope position and landform features including ridge-tops and lower slopes. Upper slopes and ridge-tops were qualified as those with little or no moisture accretion, while conversely lower slopes and drains were those with higher moisture accretion. Lastly, buffers were created around all perennial and intermittent stream polylines representing those lower slope and drain forest types. These various physical, geomorphic, and site characteristics, plus their surrounding computations, allowed forest covertypes to be inferred.
I bring this all forward not to describe methods of identifying covertypes– nothing here is unique. It is intended to demonstrate that underlying a simple graphic representation, complex data and processes provide the basis for a scientific expression. In tying back to the ideals of Japanese art and aesthetics, here is beauty and the appeal – taking the complex and making it simple, only to see the simple and recognize the complex.
Be mindful of the ‘simple’ and seek to understand its complexity – not just in maps.
UPDATE: This article is featured in the January issue of the Society of American Foresters publication, The Forestry Source. See here.
 See for example: Bolstad, et al. 1998. Predicting Southern Appalachian overstory vegetation with digital terrain data. Landscape Ecology 13: 271-283.;
Bowman, I. 1911. Forest Physiography. New York, NY: John Wiley and Sons;
McNab, WH, Et al. 1999. An unconventional approach to ecosystem unit classification in western North Carolina, USA. Forest Ecology and Management, 114: 405-420.;
Whittaker, RH. 1956. Vegetation of the Great Smoky Mountains. Ecological Monographs, 26: 1-80.;
Whittaker, RH. 1966. Forest dimensions and production in the Great Smoky Mountains. Ecology, 47: 103-121.
 Bailey, RG. 1987. Suggested hierarchy of criteria for multi-scale ecosystem mapping. Landscape Urban Planner, 14: 313-319.
 Abella, SR. 2003. Quantifying ecosystem geomorphology of the southern Appalachian Mountains. Physical Geography, 24 (6): 488-501.