Affordable Access

WRRCTR No.148 Soil and Sampling Scheme for Characterizing Soil Hydraulic Properties of a Watershed

Water Resources Research Center, University of Hawaii at Manoa
Publication Date
  • Hydraulic Conductivity
  • Hydrologic Models
  • Sampling Statistics
  • Surface-Groundwater Relationships
  • Soil Water
  • Watershed (Basins)
  • Monte Carlo Method
  • Hawaii
  • Soil Hydraulic Indices
  • Geostatistics
  • Saturated Conductivity
  • Drainage-Flux Method
  • Kriging
  • Autocorrelation
  • Variogram
  • Spatial Variability
  • Oxisols
  • Pearl Harbor
  • Oahu
  • Hydrologic Cycle -- Mathematical Models.
  • Hydrologic Models -- Hawaii -- Oahu.
  • Oxisols -- Hawaii -- Oahu.
  • Pearl Harbor (Hawaii)
  • Watersheds -- Hawaii -- Oahu -- Mathematical Models.
  • Design
  • Earth Science
  • Geography
  • Mathematics


Watershed modeling, which incorporates the stochastic nature of the hydraulic properties of the land surface and rainfall, requires a mathematical description of watershed variability, including the frequency distribution of key hydrologic parameters and the spatial structure of variances. Heterogenous watersheds require extensive sampling to characterize the spatial distribution of a property, such as hydraulic conductivity, which is frequently required as input to model calculations of infiltration and runoff. Since hydraulic conductivity, K, varies with water content, Ө, and soil water pressure, h, the K(Ө) and X(h) relationships can be conveniently represented by parameters in mathematical expressions relating these variables. The parameters of three different equations are examined as indices of the hydraulic properties of Oxisol soils in Hawaii's Pearl Harbor watershed. When what to measure and how to mathematically express the results is decided, the number and location of field measurement sites in a particular watershed are determined. Geostatistical concepts are applied to design a sampling scheme for a specific watershed in which the measured value of a hydrologic property or index at a given point is correlated with other measured values of the property that is dependent on the distance between sampling points. Required statistical parameters for the geostatistical approach are the mean, variance, and autocorrelation function or variogram. Criteria are specified for selecting the location and smallest possible number of observation points to best estimate the statistical parameters. The results suggest that 30 measurement sites are the minimum sample size for estimating the parameters required for stochastic modeling. The proposed sampling procedure is illustrated with a sampling strategy for a portion of the Pearl Harbor watershed.

There are no comments yet on this publication. Be the first to share your thoughts.


Seen <100 times