Analysing forest transpiration model errors with artificial neural networks

Stefan C. Dekker, Willem Bouten, Marcel Schaap

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

A Single Big Leaf (SBL) forest transpiration model was calibrated on half-hourly eddy correlation measurements. The SBL model is based on the Penman-Monteith equation with a canopy conductance controlled by environmental variables. The model has eight calibration parameters, which determine the shape of the response functions. After calibration, residuals between measurements and model results exhibit complex patterns and contain random and systematic errors. Artificial Neural Networks (ANNs) were used to analyse these residuals for any systematic relations with environmental variables that may improve the SBL model. Different sub-sets of data were used to calibrate and validate the ANNs. Both wind direction and wind speed turned out to improve the model results. ANNs were able to find the source area of the fluxes of the Douglas fir stand within a larger heterogeneous forest without using a priori knowledge of the forest structure. With ANNs, improvements were also found in the shape and parameterisation of the response functions. Systematic errors in the original SBL model, caused by interdependencies between environmental variables, were not found anymore with the new parameterisation. After the ANNs analyses, about 80% of the residuals can be attributed to random errors of eddy correlation measurements. It is finally concluded that ANNs are able to find systematic trends even in very noisy residuals if applied properly.

Original languageEnglish (US)
Pages (from-to)197-208
Number of pages12
JournalJournal of Hydrology
Volume246
Issue number1-4
DOIs
StatePublished - Jun 1 2001
Externally publishedYes

Fingerprint

transpiration
artificial neural network
neural networks
eddy covariance
environmental factors
leaves
parameterization
eddy
calibration
Penman-Monteith equation
wind direction
Pseudotsuga menziesii
wind speed
wind velocity
canopy

Keywords

  • Artificial Neural Networks
  • Forest transpiration
  • Model errors
  • Penman-Monteith

ASJC Scopus subject areas

  • Soil Science
  • Earth-Surface Processes

Cite this

Analysing forest transpiration model errors with artificial neural networks. / Dekker, Stefan C.; Bouten, Willem; Schaap, Marcel.

In: Journal of Hydrology, Vol. 246, No. 1-4, 01.06.2001, p. 197-208.

Research output: Contribution to journalArticle

Dekker, Stefan C. ; Bouten, Willem ; Schaap, Marcel. / Analysing forest transpiration model errors with artificial neural networks. In: Journal of Hydrology. 2001 ; Vol. 246, No. 1-4. pp. 197-208.
@article{ec3874ca39154484a7b17e5eb087cdbc,
title = "Analysing forest transpiration model errors with artificial neural networks",
abstract = "A Single Big Leaf (SBL) forest transpiration model was calibrated on half-hourly eddy correlation measurements. The SBL model is based on the Penman-Monteith equation with a canopy conductance controlled by environmental variables. The model has eight calibration parameters, which determine the shape of the response functions. After calibration, residuals between measurements and model results exhibit complex patterns and contain random and systematic errors. Artificial Neural Networks (ANNs) were used to analyse these residuals for any systematic relations with environmental variables that may improve the SBL model. Different sub-sets of data were used to calibrate and validate the ANNs. Both wind direction and wind speed turned out to improve the model results. ANNs were able to find the source area of the fluxes of the Douglas fir stand within a larger heterogeneous forest without using a priori knowledge of the forest structure. With ANNs, improvements were also found in the shape and parameterisation of the response functions. Systematic errors in the original SBL model, caused by interdependencies between environmental variables, were not found anymore with the new parameterisation. After the ANNs analyses, about 80{\%} of the residuals can be attributed to random errors of eddy correlation measurements. It is finally concluded that ANNs are able to find systematic trends even in very noisy residuals if applied properly.",
keywords = "Artificial Neural Networks, Forest transpiration, Model errors, Penman-Monteith",
author = "Dekker, {Stefan C.} and Willem Bouten and Marcel Schaap",
year = "2001",
month = "6",
day = "1",
doi = "10.1016/S0022-1694(01)00368-7",
language = "English (US)",
volume = "246",
pages = "197--208",
journal = "Journal of Hydrology",
issn = "0022-1694",
publisher = "Elsevier",
number = "1-4",

}

TY - JOUR

T1 - Analysing forest transpiration model errors with artificial neural networks

AU - Dekker, Stefan C.

AU - Bouten, Willem

AU - Schaap, Marcel

PY - 2001/6/1

Y1 - 2001/6/1

N2 - A Single Big Leaf (SBL) forest transpiration model was calibrated on half-hourly eddy correlation measurements. The SBL model is based on the Penman-Monteith equation with a canopy conductance controlled by environmental variables. The model has eight calibration parameters, which determine the shape of the response functions. After calibration, residuals between measurements and model results exhibit complex patterns and contain random and systematic errors. Artificial Neural Networks (ANNs) were used to analyse these residuals for any systematic relations with environmental variables that may improve the SBL model. Different sub-sets of data were used to calibrate and validate the ANNs. Both wind direction and wind speed turned out to improve the model results. ANNs were able to find the source area of the fluxes of the Douglas fir stand within a larger heterogeneous forest without using a priori knowledge of the forest structure. With ANNs, improvements were also found in the shape and parameterisation of the response functions. Systematic errors in the original SBL model, caused by interdependencies between environmental variables, were not found anymore with the new parameterisation. After the ANNs analyses, about 80% of the residuals can be attributed to random errors of eddy correlation measurements. It is finally concluded that ANNs are able to find systematic trends even in very noisy residuals if applied properly.

AB - A Single Big Leaf (SBL) forest transpiration model was calibrated on half-hourly eddy correlation measurements. The SBL model is based on the Penman-Monteith equation with a canopy conductance controlled by environmental variables. The model has eight calibration parameters, which determine the shape of the response functions. After calibration, residuals between measurements and model results exhibit complex patterns and contain random and systematic errors. Artificial Neural Networks (ANNs) were used to analyse these residuals for any systematic relations with environmental variables that may improve the SBL model. Different sub-sets of data were used to calibrate and validate the ANNs. Both wind direction and wind speed turned out to improve the model results. ANNs were able to find the source area of the fluxes of the Douglas fir stand within a larger heterogeneous forest without using a priori knowledge of the forest structure. With ANNs, improvements were also found in the shape and parameterisation of the response functions. Systematic errors in the original SBL model, caused by interdependencies between environmental variables, were not found anymore with the new parameterisation. After the ANNs analyses, about 80% of the residuals can be attributed to random errors of eddy correlation measurements. It is finally concluded that ANNs are able to find systematic trends even in very noisy residuals if applied properly.

KW - Artificial Neural Networks

KW - Forest transpiration

KW - Model errors

KW - Penman-Monteith

UR - http://www.scopus.com/inward/record.url?scp=0035370803&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035370803&partnerID=8YFLogxK

U2 - 10.1016/S0022-1694(01)00368-7

DO - 10.1016/S0022-1694(01)00368-7

M3 - Article

AN - SCOPUS:0035370803

VL - 246

SP - 197

EP - 208

JO - Journal of Hydrology

JF - Journal of Hydrology

SN - 0022-1694

IS - 1-4

ER -