Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 82 of the 82 references it cites were found in a registry or an index; none of the 76 sources its claims cite is checked for support yet.

Critical Tokunaga model for river networks

Open access
Kovchegov, Yevgeniy, Zaliapin, Ilya, Foufoula-Georgiou, Efi · 2022 · Physical Review E

Abstract

The hierarchical organization and self-similarity in river basins have been topics of extensive research in hydrology and geomorphology starting with the pioneering work of Horton in 1945. Despite significant theoretical and applied advances, however, the mathematical origin of and relation among Horton's laws for different stream attributes remain unsettled. Here we capitalize on a recently developed theory of random self-similar trees to elucidate the origin of Horton's laws, Hack's laws, basin fractal dimensions, power-law distributions of link attributes, and power-law relations between distinct attributes. In particular, we introduce a one-parametric family of self-similar critical Tokunaga trees that includes the celebrated Shreve's random topology model and extends to trees that approximate the observed river networks with realistic exponents. The results offer tools to increase our understanding of landscape organization under different hydroclimatic forcings, and to extend scaling relationships useful for hydrologic prediction to resolutions higher than those observed.

Overview

CoChat Summary

TL;DR

The paper introduces a one-parameter family of self-similar critical Tokunaga trees for river networks, unifying Horton prune-invariance and Tokunaga self-similarity and deriving Horton laws, fractal dimension, Hack’s law, and link-attribute power laws. The model includes Shreve’s random topology model at \(c=2\) and closely fits observed exponents with \(c\approx 2.3\).

Key findings

  • In a critical Tokunaga tree, the Horton exponents reduce to two values: \(R_B=R_M=R_A=2c\) and \(R_S=R_L=R_\Lambda=c\). For \(c=2.3\), Table I gives \(R_B=4.6\), \(R_L=2.3\), fractal dimension \(d=\log_c(2c)=1.832\), Hack exponent \(h=\log_{2c}c=0.546\), \(\beta_A=0.454\), and \(\beta_\Lambda=0.832\).
  • Theorem 1 shows that any self-similar tree with a tamed Tokunaga sequence obeys Horton laws for stream numbers and magnitudes, with \(R_B=R_M=1/w_0\), where \(w_0\) is the unique real root of \(t(z)=-1+2z+T(z)\) in \((0,1/2]\). The ideal exact Horton law holds iff \(T_k=0\) for \(k>1\).
  • The critical Tokunaga model includes Shreve’s random topology model as the special case \(c=2\), and \(c\approx 2.3\) approximates observed river networks. The Beaver Creek basin of order \(K=6\) is fit by the critical Tokunaga model with \(c=2.3\) for seven Horton laws.
  • The model yields fractal dimension \(d=\max\{1,\log R_B/\log R_L\}\) and Hack’s law \(h=\log R_L/\log R_B\); in the critical Tokunaga case these become \(d=\log_c(2c)\) and \(h=\log_{2c}c\).

Methods

  • Theoretical/mathematical modeling: river basins are represented as rooted binary planted trees; self-similarity is defined as coordination plus Horton prune-invariance; Tokunaga coefficients \(T_k\) parameterize self-similar trees. The Random Attachment Model (RAM) constructs trees via Poisson attachment along exponential stream segments and a recursive branching-process algorithm.
  • Data/illustration: the stream network of Beaver Creek, Floyd County, KY, of Horton-Strahler order \(K=6\), extracted with RiverTools, is used to illustrate fits to seven Horton laws. Table I compares critical Tokunaga exponents with the OCN model and observed river networks; Fig. 4 uses nine real river basins from Tarboton et al. (1988).
  • Main analysis: generating functions and singularity analysis are used to prove Horton laws, fractal dimension, Hack’s law, and power-law link-attribute distributions; appendices contain proofs. The simulation algorithm generates trees with up to \(10^6\) edges within seconds.

Limitations

  • Stated: the critical Tokunaga model is “merely a subclass of a much broader family of self-similar trees” and may not accommodate all problem-specific data features; deviations from its stiff parameterization may have physical underpinnings.
  • Stated: the ideal Horton law rarely holds in theoretical trees of arbitrarily large size, and the power relations among attributes and power-law link-attribute frequencies derived from the ideal Horton law “remain heuristic”; a formal analysis based on actual Horton laws is presented elsewhere.
  • Evident from the text: empirical support is mainly one basin example (Beaver Creek) plus comparison to previously published exponents and the OCN model; no new multi-basin statistical validation is reported. The model treats convergent tributary networks, while extension to distributary delta networks is left as future work.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1103/physreve.105.014301

Cites

82 of 82 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

FIG. 1 Horton pruning of the stream network of Beaver creek, Floyd County, KY. Streams of orders K = 2 , . . . , 6 are shown by different colors (see legend on the right). Streams of order 1 (source streams) are not shown for visual convenience. (A) The first Horton pruning, after eliminating streams of order K=1. (B)-(E) Second to fifth consecutive Horton prunings. The sixth pruning completely eliminates the network. The channel extraction is done using RiverTools software ( http://rivix.com ).
FIG. 1 Horton pruning of the stream network of Beaver creek, Floyd County, KY. Streams of orders K = 2 , . . . , 6 are shown by different colors (see legend on the right). Streams of order 1 (source streams) are not shown for visual convenience. (A) The first Horton pruning, after eliminating streams of order K=1. (B)-(E) Second to fifth consecutive Horton prunings. The sixth pruning completely eliminates the network. The channel extraction is done using RiverTools software ( http://rivix.com ).CC-BY · From the PDF
FIG. 2 Critical Tokunaga fit to the Horton laws in the stream network of Beaver creek. The stream network is shown in Fig. 1. Symbols correspond to the values of the observed attributes. Lines and dots show the fit by the critical Tokunaga model with c = 2 . 3. (A) Stream numbers N K . The model fit is given by (13); it has asymptotic slope -log 10 (2 c ) ≈ -0 . 66. (B) Average stream magnitude M K (cyan circles) and average number of links S K (magenta squares). The fit for M K is given by (14); it has asymptotic slope log 10 (2 c ) ≈ 0 . 66. The fit for S K is given by c K -1 ; it has theoretical slope log 10 c ≈ 0 . 36. (C) Average total contributing area A K (cyan circles) and average total upstream channel length L tot K (magenta squares). The fitting lines, according to a combination of equations (14) and (50), have theoretical slope log 10 (2 c ) ≈ 0 . 66. (D) Average stream length L K (magenta squares) and average length Λ K of the longest stream to the divide (cyan circles). The fitting lines, according to equations (21) and (23), have theoretical slope log 10 ( c ) ≈ 0 . 36.
FIG. 2 Critical Tokunaga fit to the Horton laws in the stream network of Beaver creek. The stream network is shown in Fig. 1. Symbols correspond to the values of the observed attributes. Lines and dots show the fit by the critical Tokunaga model with c = 2 . 3. (A) Stream numbers N K . The model fit is given by (13); it has asymptotic slope -log 10 (2 c ) ≈ -0 . 66. (B) Average stream magnitude M K (cyan circles) and average number of links S K (magenta squares). The fit for M K is given by (14); it has asymptotic slope log 10 (2 c ) ≈ 0 . 66. The fit for S K is given by c K -1 ; it has theoretical slope log 10 c ≈ 0 . 36. (C) Average total contributing area A K (cyan circles) and average total upstream channel length L tot K (magenta squares). The fitting lines, according to a combination of equations (14) and (50), have theoretical slope log 10 (2 c ) ≈ 0 . 66. (D) Average stream length L K (magenta squares) and average length Λ K of the longest stream to the divide (cyan circles). The fitting lines, according to equations (21) and (23), have theoretical slope log 10 ( c ) ≈ 0 . 36.CC-BY · From the PDF
FIG. 3 Examples of critical Tokunaga trees. The trees are generated by the critical Tokunaga process with c = 2 . 3 and Horton-Strahler order K = 5. The line width is proportional to the contributing area. The figure accurately represents the tree combinatorial structure; the edge lengths are scaled for a better planar embedding.
FIG. 3 Examples of critical Tokunaga trees. The trees are generated by the critical Tokunaga process with c = 2 . 3 and Horton-Strahler order K = 5. The line width is proportional to the contributing area. The figure accurately represents the tree combinatorial structure; the edge lengths are scaled for a better planar embedding.CC-BY · From the PDF
FIG. 4 Fractal dimension d (panel A ) of equation (22) and Hack's exponent h = d -1 (panel B ) of a self-similar RAM tree in the limit of infinite size as a function of the Horton exponents R B and R L . Selected levels of d and h are shown by marked black lines. Green thick lines correspond to the critical Tokunaga process (Definition 2, Eq. (24)) for which R B = 2 R L . Blue dots depict the pairs ( R B , R L ) estimated in nine real river basins by [42], see also [2, Table 2.1].
FIG. 4 Fractal dimension d (panel A ) of equation (22) and Hack's exponent h = d -1 (panel B ) of a self-similar RAM tree in the limit of infinite size as a function of the Horton exponents R B and R L . Selected levels of d and h are shown by marked black lines. Green thick lines correspond to the critical Tokunaga process (Definition 2, Eq. (24)) for which R B = 2 R L . Blue dots depict the pairs ( R B , R L ) estimated in nine real river basins by [42], see also [2, Table 2.1].CC-BY · From the PDF

Verification

Claims the paper makes

0 Valid · 2 Not valid · 49 Not checked · 109 Own results and interpretations

105 more claims

Create a free account to see all 160 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 32
  • Probably not a substantial claim. 21
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

2 of 2

Citations

Cites

Show all 82 references

Research works printed but not linked: 2

  1. V. K. Gupta, 1998, Rainfall, Landforms and Floods
  2. J. T. Hack, 1957, U.S. Government Printing Office, 294

Cited by

Metadata

Authors
Kovchegov, Y., Zaliapin, I., & Foufoula-Georgiou, E.
Affiliations
  • Oregon State University (US)
  • University of Nevada, Reno (US)
  • Department of Civil and Environmental Engineering and Department of Earth System Science, University of California, Irvine, California 92697, USA
Venue
Physical Review E
Published
2022-01-03
Volume
105
Issue
1
Pages
014301-014301
ISSN
  • 2470-0045 (ISSN-L)
  • 2470-0053
OpenAlex
W3166160688
Cited by
7
References
82
Citations by year
2026: 3 · 2025: 1 · 2024: 1 · 2023: 1 · 2022: 1
FWCI
0.30
Type
journal article
Language
en
Open access
Locations
  • Physical review. E (publishedVersion)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University) (cc-by)
Funders
National Science Foundation (DMS-1412557, EAR-1723033, DMS-1839336, EAR-1811909)
Topics
  • Physical Sciences › Environmental Science › Water Science and Technology › Hydrology and Watershed Management Studies
  • Physical Sciences › Earth and Planetary Sciences › Earth-Surface Processes › Hydrology and Sediment Transport Processes
  • Physical Sciences › Physics and Astronomy › Condensed Matter Physics › Theoretical and Computational Physics
Keywords
river networks, Horton's laws, Hack's law, power-law distribution, scaling laws, landscape organization

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-19, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-19 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.