There are many factors that determine how we perceive and ultimately classify the climate into distinct regions, or biomes. For example, precipitation, temperature, flora and fauna are key ingredients in our delineation of biomes. Classifying biomes is akin to understanding the homogeneity of climate and resources on a local scale. It is not only important that we can classify climate biomes, but also their evolution with the changing climate. For emphasis, it is not possible to perform any meaningful environmental preservation without predetermined knowledge of what the environment is at a coarse scale.
Historically, the standard used to classify climate biomes has been the Köppen-Geiger (KG) model [1]. The KG model is an expert based judgment that describes climate zones based off of temperature and precipitation measurements. The model utilizes a fixed decision tree, where each branch uses various information about temperature and precipitation. This heuristic allows one to broadly assess climate regions. While KG is interpretable, it is overly simplistic and somewhat arbitrary.
A solution to this problem is to move towards data-driven methods of classification. Various authors have noted this, and have applied unsupervised models to classify the climate (see for example, [2], [3]). These data driven approaches to climate clustering are an epistemological improvement over the use chosen heuristics of KG. However, they still suffer from two key problems. First, the algorithms themselves are user chosen, and therefore somewhat arbitrary themselves. There is no external measurement to measure how much one trusts the clustering result. Second, because the algorithms are dependent on the input data, they are dependent on the scale at which the data is acquired. A priori, it is not clear how this “hidden parameter” of scale affects the overall clustering result. The goal of this project is to address both of these problems.
[1] Markus Kottek, Jürgen Grieser, Christoph Beck, Bruno Rudolf, and Franz Rubel. World map of the köppen-geigerclimate classification updated.Meteorologische Zeitschrift, 15(3):259–263, 2006.
[2] Pawel Netzel and Tomasz Stepinski. On using a clustering approach for global climate classification.Journal ofClimate, 29(9):3387–3401, 2016.
[3] Jakob Zscheischler, Miguel D Mahecha, and Stefan Harmeling. Climate classifications: the value of unsupervisedclustering.Procedia Computer Science, 9:897–906, 2012.