Water in the Landscape: A Review of Contemporary Flow Routing Algorithms

Abstract

This chapter reviews the various flow routing algorithms that simulate the distribution and flow of water across landscapes. The distinguishing characteristics of nine such algorithms and the experiments that have been conducted to evaluate their performance over the past 15 years are discussed. From there, we consider three sets of enduring challenges: (1) the role of scale and feedback between soil and water, and the need to consider these issues when characterizing the properties of both; (2) the need for dynamic flow routing algorithms and related indices in many landscapes; and (3) some of the as yet unrealized opportunities for treating space and time as continuous variables in the representation of soil water properties. The chapter concludes by noting the current state-of-the-art and where we might go from here.

Digital terrain modeling

“…Fisher, 1998;Wilson et al, 2008). Reuter et al (2009, p. 90) suggested that the true applicability of DEMs for geomorphometric analysis can only be assessed by providing answers to the following questions: (1) how accurately is the surface roughness represented?, (2) how accurately is the shape of the land surface represented (i.e.…”

Fuzzy Similarity and Fuzzy Inclusion Measures in Polyline Matching: A Case Study of Potential Streams Identification for Archaeological Modelling in GIS

“…The positions of the streams could also be derived from the landscape morphology using a digital terrain model (DTM) and applying M8, MD, or MD8 algorithm, see e.g. (Wilson et al, 2008), which are commonly implemented in GIS software packages. Although the (natural) stream network can be derived from contemporary DTM, these results may not represent the streams in the original position in the past due to the landscape changes.…”

Influence of pit removal algorithms on surface runoff simulation

“…Depending on whether the algorithm concentrates all water in the centre of the pixel, or distributes water across the entire pixel, we can differentiate one-dimensional algorithms such as D8, Rho8 and KRA algorithms [9][10][11] and twodimensional ones, i.e. D∞, FD8 and DEMON algorithms [12][13][14] of the CRRS surface runoff model. Regardless of whether we are dealing with one-dimensional or multi-dimensional surface runoff models, the key assumption is that each analysed DTM pixel has at least one pixel in its immediate neighbourhood that is lower than it, and toward which it directs its flow.…”