Elsevier

Journal of Hydrology

Volume 222, Issues 1–4, 13 September 1999, Pages 18-36
Journal of Hydrology

Runoff characteristics of pipeflow and effects of pipeflow on rainfall-runoff phenomena in a mountainous watershed

https://doi.org/10.1016/S0022-1694(99)00090-6Get rights and content

Abstract

In order to assess the runoff characteristics of pipeflow in mountainous watersheds, precipitation and amount of pipeflow were measured for 3 years at a forested mountainous watershed in the central part of Japan. Similar changes in runoff characteristics were observed for two pipes over the study period; the maximum discharge rate of pipeflow increased, the recession hydrograph of pipeflow became steep, and the duration of pipeflow discharge became shorter. However, the changes of runoff characteristics had no significant effect on relationships between total rainfall and total discharge from pipes. Based on observations of sediment discharge from soil pipes, it was suggested that the sediment discharge from these soil pipes caused the changes of runoff characteristics of pipeflow. Moreover, in order to clarify the effects of pipeflow on streamflow, simultaneous measurements of pipeflow and streamflow were conducted at the same watershed. Results showed that the water discharge phenomena from this watershed can be classified into three stages based on the soil pipes which contribute the water discharge: (1) under small amounts of rainfall, water discharge occurs only at the spring situated at the watershed exit; (2) as the magnitude of rainfall increases, water comes out from a pipe located at 6 m upstream from the watershed exit; and (3) under the heaviest rainfall, additional discharge occurs from a group of pipes located 10 m upstream from the watershed exit.

Introduction

A chain of connected macropores, developed nearly parallel to the soil surface, is commonly found in slopes, ranging from subarctic wetland (Woo and diCenzo, 1988) to tropical rain forest (Walsh and Howells, 1988, Elsenbeer et al., 1995, Elsenbeer and Lack, 1996). These macropores are commonly referred to as ‘soil pipes’, and the water flowing through them is called ‘pipeflow’ (e.g. Jones, 1971, Beven and Germann, 1982). Numerous studies have demonstrated that a great quantity of pipeflow from soil pipes can be produced during rainfall (McCaig, 1983, Tsukamoto and Ohta, 1988, Kitahara and Nakai, 1992, Onda, 1994). Tracer experiments have indicated a velocity higher than Darcian groundwater flow (Mosely, 1982, Tanaka et al., 1988). The high velocity of pipeflow has also been indicated in studies that investigated the relationship between the runoff and soil water content or the groundwater level (McDonnell, 1990, Tani, 1997). In studies involving the chemistry of pipeflow, the role of pipeflow contents has been shown to be an important concern in stream water quality (Muscutt et al., 1990, Elsenbeer et al., 1995, Elsenbeer and Lack, 1996).

As suggested by Kirkby (1988), investigations such as those mentioned above indicate that pipeflow plays an important role in the lateral movement of water in forested hillslopes. However, Selby, 1993, Bryan and Jones, 1997, Jones, 1997 indicated the general lack of long term pipeflow measurements from a variety of catchments and climatic environments. Only in the Maesnant basin (mid-Wales), over 200 storms has been monitored (Jones and Crane, 1984, Jones, 1987, Jones, 1997). Moreover, very few studies have simultaneously measured the discharge rates of pipeflow and streamflow in natural mountainous watersheds. Accordingly, adequate information is lacking with regard to the effects of pipeflow on runoff from such watersheds.

The objectives of this study are to obtain an insight into the runoff characteristics of pipeflow and to evaluate the effects of pipeflow on water discharge from mountainous watershed through field measurements.

Section snippets

Site description

The observations were conducted at the Toinotani 0-order valley watershed of Kyoto University Forest in Ashiu. The watershed is located in the northeastern part of Kyoto Prefecture in the central part of Japan (Fig. 1). With an area of 0.64 ha and a mean hollow gradient of 35.8° (Fig. 2), the altitude of the watershed ranges from 730 to 830 m above sea level. The watershed is covered with a closed forest of predominately Cryptomeria japonica and Betula grossa, and is underlain by Paleozoic

General trends of observed hydrographs

Table 1 shows the number of monitored storm events and the number of storm events in which flow occurred in Pipes A and B1. Fig. 6, Fig. 7 show hyetographs and hydrographs of the flow from Pipe A from 1 July to 31 October 1993 and 1996, respectively. Fig. 6 also shows the hydrograph of Pipe B1 flow and Fig. 7 shows the flow duration of Pipe B1. Moreover, the streamflow hydrograph is also shown in Fig. 7. Water runoff from Pipe A occurred only during heavy rainfall storms above approximately 12 mm

Conclusions

Over a three-year period, the following measurements were made: precipitation and the amount of pipeflow and stream discharge, in order to assess the runoff characteristics of pipeflow, and the effects of pipeflow on the water discharge. The following conclusions were drawn:

  • 1.

    The maximum pipeflow was 0.1 mm h−1 (0.18 l s−1), mean lags in the start and peak of pipeflow in comparison with the start and end of rainfall were 11.0–12.1 and 1.6–3.7 h, respectively.

  • 2.

    The runoff characteristics of the ephemeral

Acknowledgements

The authors express their deep appreciation to Prof Sumiji Kobashi., Drs Masaharu Fujita and Nobuhito Ohte (all from Kyoto University) for their valuable and encouraging advice and suggestions.

References (36)

  • K.J. Beven et al.

    Macropores and water flow in soils

    Water Resour. Res.

    (1982)
  • W.E. Dietrich et al.

    Channel head

  • T. Dunne

    Field studies of hillslope flow process

  • H. Elsenbeer et al.

    Chemical fingerprints of hydrological compartments and flow paths at La Cuenca, western Amazonia

    Water Resour. Res.

    (1995)
  • J.A.A. Jones

    Soil piping and stream channel initiation

    Water Resour. Res.

    (1971)
  • J.A.A. Jones et al.

    Pipeflow and pipe erosion in the Maesnant experimental catchment

  • J.A.A. Jones

    The effects of soil piping on contributing area and erosion pattern

    Earth Surf. Proc. Landf.

    (1987)
  • J.A.A. Jones

    Pipeflow contribution areas and runoff response

    Hydro. Proc.

    (1997)
  • Cited by (108)

    • Piping formation and distribution in the semi-arid Northern Negev environment: A new conceptual model

      2022, Catena
      Citation Excerpt :

      Piping is a water-erosion process in hillslopes that leads to subsurface macropore formation, creating discrete preferential flow paths parallel to the hillslope (Sharma et al., 2010; Uchida et al., 1999; Weiler and McDonnell, 2007).

    • Geophysical for granitic joint patern and subsurface hydrology related to slope instability

      2022, Journal of Applied Geophysics
      Citation Excerpt :

      In forest area, where the roots reach up to 200 cm in the soil, they are responsible for the predominance of lateral flows. In this way, from the results of Marques (2016) and Borges (2016), we can suggest the formation of lateral preferential paths in depth, or in the soil-rock contact on the slope, formed mainly by root activity, as pointed out by Uchida et al. (1999) and McDonnell et al. (2007). The existence of subsurface flows on the hillside suggests the high occurrence of landslides in secondary forest and grassland coverages in the extreme rain event in January 2011 in the mountainous region of Rio de Janeiro and specifically in the Córrego Dantas watershed.

    View all citing articles on Scopus
    View full text