This study uses the concept of Potential Natural Vegetation (PNV), developed in the mid-1900s by German botanist Reinhold Tüxen. Described by the authors as “one of the most successful novelties in vegetation science over the last decades” [Liu 2009: 1313], PNV can be defined as a projection of the natural vegetation that would exist in a given area in the absence of human interference.
“By showing the relationships between environmental variables and vegetation types, maps of the PNV are an important instrument in the study and planning of the environment, and act as decision-support tools for the solutions to environmental issues” [Liu 2009: 1313]. Such maps are informed by studying remnant natural (old growth) vegetation in the area and site observations of the area to be mapped.
Computer modeling can be used to predict “the geographic distribution of vegetation composition across a landscape from mapped environmental variables, such as climate, soils, and geology. When a predictive vegetation modeling is calibrated using observation of vegetation composition taken from mature or ‘climax’ vegetation stands, then potential natural vegetation is portrayed in a predictive map” [Liu 209: 1314].
Focusing on northeastern China, the study identified 16 vegetation types in the region, along with the environmental factors influencing their distribution. Climatic factors included: mean annual temperature, mean temperature of the coldest month, relative humidity, and potential evapotranspiration rate. Topographical factors were elevation and slope.
“Generally, as the elevation increases, the change of temperature and moisture leads to the obvious differentiation phenomenon in vegetation vertical zones. Slope is related to the hydrology (overland and subsurface flow velocity and runoff rate) and potential soil moisture and soil development of a habitat” [Liu 2009: 1315].
They compared the map created by their model to existing vegetation maps of the region. “Visual comparison of the predicted PNV distributions with their actual equivalents indicates a good agreement” [Liu 2009: 1317]. Some modeled vegetation types did not agree with existing maps, however, meaning that “some more important environmental factors may have been missing in the model” [Liu 2009: 1318]. The authors also state that calibrating their model with additional field data on what is currently growing, collected from throughout the region, would improve the model’s accuracy.
The article concludes by stating that a ‘vegetation-environment’ model can help to determine PNV under not only current, but also predicted future environmental conditions.
Liu, Huamin, Lixin Wang, et al., 2009, Predictive modeling of the potential natural vegetation pattern in northeastern China, Ecol Res 24, https://esj-journals.onlinelibrary.wiley.com/doi/10.1007/s11284-009-0616-3