Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Environmental Microbial & Food Safety Laboratory » Research » Publications at this Location » Publication #314634

Title: Modern concepts of scale and scaling in soils

Author
item Pachepsky, Yakov

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 2/25/2015
Publication Date: 6/16/2015
Citation: Pachepsky, Y.A. 2015. Modern concepts of scale and scaling in soils. Meeting Abstract. Coruna, Spain on June 16-19, 2015.

Interpretive Summary:

Technical Abstract: Transferring information across scales is the core activity in environmental research and applications. The objective of this presentation is to provoke discussion on status of scale concepts and techniques in soil systems analysis that operates with data collected at different scales and has to overcome the scale mismatch among components of knowledge acquisition, packaging and use for societal needs. Three major definitions of scale – via hierarchies, measurement metrics, and similitude are discussed, and differences in scaling under each of the definitions are reviewed. Advantages and limitations of traditional scaling methods such as dimensional analysis and inspectional analysis are acknowledged. The power law scaling is reviewed with regard to mechanisms and models leading to it. Support change techniques via aggregation and weighted interpolation including wavelet decomposition are summarized. The fast growing field of scale change in spatio-temporal information is represented with scaling using empirical orthogonal functions, data assimilation with modeling-based interpolation, and cumulative distribution matching. Special cases of scaling based on processes and phenomena, such as fragmentation, temporal stability, and Buckingham flow, are briefly summarized. The importance of scaling variability metrics and parameters of flux models is underscored. The role of scaling increases in the era of ‘big data’. The arsenal of scaling methods is expected to grow as it performs the important function of obtaining data we need from data we have.