A classification approach to reconstruct local daily drying dynamics at headwater streams

Aurélien Beaufort,Eric Sauquet,Julie Carreau

doi:10.1002/hyp.13445

Abstract

AbstractHeadwater streams (HSs) are generally naturally prone to flow intermittence. These intermittent rivers and ephemeral streams have recently seen a marked increase in interest, especially to assess the impact of drying on aquatic ecosystems. The two objectives of this work are (a) to identify the main drivers of flow intermittence dynamics in HS and (b) to reconstruct local daily drying dynamics. Discrete flow states—“flowing” versus “drying”—are modelled as functions of covariates that include information on climate, hydrology, groundwater, and basin descriptors. Three classifiers to estimate flow states using covariates are tested on four contrasted regions in France: (a) a linear classifier with regularization (LASSO for least absolute shrinkage and selection operator) and two non‐linear non‐parametric classifiers, (b) a one‐hidden‐layer feedforward artificial neural network (ANN) classifier, and (c) a random forest (RF) classifier. The three classifiers are compared with a benchmark classifier (BC) that simply estimates dominant flow state for each month based on observations (without using covariates). The performance assessment over the period 2012–2016 carried out by cross‐validation shows that the three classifiers for flow state based on covariates outperformed the BC. This demonstrates the predictive power of the covariates. ANN is the classifier that globally achieves the best performance to predict the daily drying dynamics whereas both RF and LASSO tend to underestimate the proportion of drying states. The covariates are ranked in terms of relevance for each classifier. The monthly proportion of drying states provided by the discrete observation network has a major importance for the three classifiers ANN, LASSO, and RF. This may reflect the proclivity of a site to flow intermittence. ANN gives higher importance to climatic and hydrological covariates and its non‐linearity allows a greater degree of freedom.

Highlights

Headwater streams (HSs) are generally defined as the uppermost streams in a watershed and represent a large part of hydrographical networks (Leopold, Wolman, & Miller, 1964; Nadeau & Rains, 2007)
A a þ b: a Recall 1⁄4 a þ c: (6) In this evaluation, the covariates are grouped according to their type defined in Table 1 except for MPD because this covariate is directly related to local observations
MPD is considered as very important for the three classifiers. This covariate reflects the level of flow intermittence of each Observatoire National des Etiages Network (ONDE) site

Summary

Introduction

Headwater streams (HSs) are generally defined as the uppermost streams in a watershed and represent a large part of hydrographical networks (Leopold, Wolman, & Miller, 1964; Nadeau & Rains, 2007). HS can be fed by groundwater, precipitation, and run‐off from small drainage areas They contribute to the good functioning of rivers (sediment flux, inputs of particulate organic matter and nutrients), provide primordial ecosystem services (biogeochemical cycling, sources of aquatic organisms, aquatic habitat, thermal refuge; Finn, Bonada, Mùrria, & Hughes, 2011; Larned, Datry, Arscott, & Tockner, 2010; Meyer et al, 2007), and constitute reference areas to be preserved (Lowe & Likens, 2005). IRES have seen a marked increase in interest stimulated by the challenges of water management facing the global change context (Acuña et al, 2014; Datry et al, 2016) and by the need to improve existing knowledge on aquatic ecosystems in IRES (Larned et al, 2010; Leigh & Datry, 2017; Sarremejane et al, 2017; Stubbington, England, Wood, & Sefton, 2017)

Objectives

Results

Conclusion