Premium
A model to consider the spatial variability of rainfall partitioning within deciduous canopy. II. Model parameterisation and testing
Author(s) -
DAVIE T. J. A.,
DUROCHER M. G.
Publication year - 1997
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/(sici)1099-1085(199709)11:11<1525::aid-hyp483>3.0.co;2-s
Subject(s) - throughfall , stemflow , interception , environmental science , canopy , spatial variability , canopy interception , hydrology (agriculture) , evapotranspiration , soil science , mathematics , statistics , ecology , soil water , geology , geotechnical engineering , biology
This paper gives details of how the spatially variable rainfall partitioning model, presented in Davie and Durocher, (1997) [Hydrol. Process., 11 , 1509–1523], is parameterized, and the testing of its capabilities. A majority of the input parameters required for running the model can be derived from physical measurements. The remainder can be derived from previous forest interception studies. The measured values used in model testing were from a mixed oak and sweet chestnut plantation on the outskirts of Bristol, UK. A sensitivity analysis of the model has been carried out. The importance to throughfall of potential evaporation rate, storage capacity and gap area within the canopy is highlighted. For stemflow the proportion of tree crown area generating stemflow is the most important parameter. This is largely a reflection of the model structure. The model has been tested with field data to evaluate its ability for predicting: total throughfall and stemflow for individual events; spatial variability in rainfall partitioning; and temporal dynamics of throughfall. The model was able reproduce these data sets; however the field data (from an intensive plot study) was not adequate to allow complete validation of the model. The novel feature of the model (spatial variability of rainfall partitioning) was shown to be quite well simulated. Suggestions are given for using this type of model in future research. © 1997 John Wiley & Sons, Ltd.