Evaluating the agreement between measurements and models of net ecosystem exchange at different times and timescales using wavelet coherence

An example using data from the North American Carbon Program Site-Level Interim Synthesis

P. C. Stoy, M. C. Dietze, A. D. Richardson, R. Vargas, A. G. Barr, R. S. Anderson, M. A. Arain, I. T. Baker, T. A. Black, J. M. Chen, R. B. Cook, C. M. Gough, R. F. Grant, D. Y. Hollinger, R. C. Izaurralde, C. J. Kucharik, P. Lafleur, B. E. Law, S. Liu, E. Lokupitiya & 13 others Y. Luo, J. W. Munger, C. Peng, B. Poulter, D. T. Price, D. M. Ricciuto, W. J. Riley, A. K. Sahoo, K. Schaefer, Christopher R Schwalm, H. Tian, H. Verbeeck, E. Weng

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Earth system processes exhibit complex patterns across time, as do the models that seek to replicate these processes. Model output may or may not be significantly related to observations at different times and on different frequencies. Conventional model diagnostics provide an aggregate view of model-data agreement, but usually do not identify the time and frequency patterns of model-data disagreement, leaving unclear the steps required to improve model response to environmental drivers that vary on characteristic frequencies. Wavelet coherence can quantify the times and timescales at which two time series, for example time series of models and measurements, are significantly different. We applied wavelet coherence to interpret the predictions of 20 ecosystem models from the North American Carbon Program (NACP) Site-Level Interim Synthesis when confronted with eddy-covariance-measured net ecosystem exchange (NEE) from 10 ecosystems with multiple years of available data. Models were grouped into classes with similar approaches for incorporating phenology, the calculation of NEE, the inclusion of foliar nitrogen (N), and the use of model-data fusion. Models with prescribed, rather than prognostic, phenology often fit NEE observations better on annual to interannual timescales in grassland, wetland and agricultural ecosystems. Models that calculated NEE as net primary productivity (NPP) minus heterotrophic respiration (HR) rather than gross ecosystem productivity (GPP) minus ecosystem respiration (ER) fit better on annual timescales in grassland and wetland ecosystems, but models that calculated NEE as GPP minus ER were superior on monthly to seasonal timescales in two coniferous forests. Models that incorporated foliar nitrogen (N) data were successful at capturing NEE variability on interannual (multiple year) timescales at Howland Forest, Maine. The model that employed a model-data fusion approach often, but not always, resulted in improved fit to data, suggesting that improving model parameterization is important but not the only step for improving model performance. Combined with previous findings, our results suggest that the mechanisms driving daily and annual NEE variability tend to be correctly simulated, but the magnitude of these fluxes is often erroneous, suggesting that model parameterization must be improved. Few NACP models correctly predicted fluxes on seasonal and interannual timescales where spectral energy in NEE observations tends to be low, but where phenological events, multi-year oscillations in climatological drivers, and ecosystem succession are known to be important for determining ecosystem function. Mechanistic improvements to models must be made to replicate observed NEE variability on seasonal and interannual timescales.

Original languageEnglish (US)
Pages (from-to)6893-6909
Number of pages17
JournalBiogeosciences
Volume10
Issue number11
DOIs
StatePublished - 2013

Fingerprint

net ecosystem exchange
wavelet
timescale
synthesis
carbon
ecosystem
programme
ecosystems
ecosystem respiration
respiration
phenology
time series analysis
parameterization
wetlands
annual grasslands
time series

ASJC Scopus subject areas

  • Earth-Surface Processes
  • Ecology, Evolution, Behavior and Systematics

Cite this

Evaluating the agreement between measurements and models of net ecosystem exchange at different times and timescales using wavelet coherence : An example using data from the North American Carbon Program Site-Level Interim Synthesis. / Stoy, P. C.; Dietze, M. C.; Richardson, A. D.; Vargas, R.; Barr, A. G.; Anderson, R. S.; Arain, M. A.; Baker, I. T.; Black, T. A.; Chen, J. M.; Cook, R. B.; Gough, C. M.; Grant, R. F.; Hollinger, D. Y.; Izaurralde, R. C.; Kucharik, C. J.; Lafleur, P.; Law, B. E.; Liu, S.; Lokupitiya, E.; Luo, Y.; Munger, J. W.; Peng, C.; Poulter, B.; Price, D. T.; Ricciuto, D. M.; Riley, W. J.; Sahoo, A. K.; Schaefer, K.; Schwalm, Christopher R; Tian, H.; Verbeeck, H.; Weng, E.

In: Biogeosciences, Vol. 10, No. 11, 2013, p. 6893-6909.

Research output: Contribution to journalArticle

Stoy, PC, Dietze, MC, Richardson, AD, Vargas, R, Barr, AG, Anderson, RS, Arain, MA, Baker, IT, Black, TA, Chen, JM, Cook, RB, Gough, CM, Grant, RF, Hollinger, DY, Izaurralde, RC, Kucharik, CJ, Lafleur, P, Law, BE, Liu, S, Lokupitiya, E, Luo, Y, Munger, JW, Peng, C, Poulter, B, Price, DT, Ricciuto, DM, Riley, WJ, Sahoo, AK, Schaefer, K, Schwalm, CR, Tian, H, Verbeeck, H & Weng, E 2013, 'Evaluating the agreement between measurements and models of net ecosystem exchange at different times and timescales using wavelet coherence: An example using data from the North American Carbon Program Site-Level Interim Synthesis', Biogeosciences, vol. 10, no. 11, pp. 6893-6909. https://doi.org/10.5194/bg-10-6893-2013
Stoy, P. C. ; Dietze, M. C. ; Richardson, A. D. ; Vargas, R. ; Barr, A. G. ; Anderson, R. S. ; Arain, M. A. ; Baker, I. T. ; Black, T. A. ; Chen, J. M. ; Cook, R. B. ; Gough, C. M. ; Grant, R. F. ; Hollinger, D. Y. ; Izaurralde, R. C. ; Kucharik, C. J. ; Lafleur, P. ; Law, B. E. ; Liu, S. ; Lokupitiya, E. ; Luo, Y. ; Munger, J. W. ; Peng, C. ; Poulter, B. ; Price, D. T. ; Ricciuto, D. M. ; Riley, W. J. ; Sahoo, A. K. ; Schaefer, K. ; Schwalm, Christopher R ; Tian, H. ; Verbeeck, H. ; Weng, E. / Evaluating the agreement between measurements and models of net ecosystem exchange at different times and timescales using wavelet coherence : An example using data from the North American Carbon Program Site-Level Interim Synthesis. In: Biogeosciences. 2013 ; Vol. 10, No. 11. pp. 6893-6909.
@article{d747af3de5874170a6147c5861ac0c4a,
title = "Evaluating the agreement between measurements and models of net ecosystem exchange at different times and timescales using wavelet coherence: An example using data from the North American Carbon Program Site-Level Interim Synthesis",
abstract = "Earth system processes exhibit complex patterns across time, as do the models that seek to replicate these processes. Model output may or may not be significantly related to observations at different times and on different frequencies. Conventional model diagnostics provide an aggregate view of model-data agreement, but usually do not identify the time and frequency patterns of model-data disagreement, leaving unclear the steps required to improve model response to environmental drivers that vary on characteristic frequencies. Wavelet coherence can quantify the times and timescales at which two time series, for example time series of models and measurements, are significantly different. We applied wavelet coherence to interpret the predictions of 20 ecosystem models from the North American Carbon Program (NACP) Site-Level Interim Synthesis when confronted with eddy-covariance-measured net ecosystem exchange (NEE) from 10 ecosystems with multiple years of available data. Models were grouped into classes with similar approaches for incorporating phenology, the calculation of NEE, the inclusion of foliar nitrogen (N), and the use of model-data fusion. Models with prescribed, rather than prognostic, phenology often fit NEE observations better on annual to interannual timescales in grassland, wetland and agricultural ecosystems. Models that calculated NEE as net primary productivity (NPP) minus heterotrophic respiration (HR) rather than gross ecosystem productivity (GPP) minus ecosystem respiration (ER) fit better on annual timescales in grassland and wetland ecosystems, but models that calculated NEE as GPP minus ER were superior on monthly to seasonal timescales in two coniferous forests. Models that incorporated foliar nitrogen (N) data were successful at capturing NEE variability on interannual (multiple year) timescales at Howland Forest, Maine. The model that employed a model-data fusion approach often, but not always, resulted in improved fit to data, suggesting that improving model parameterization is important but not the only step for improving model performance. Combined with previous findings, our results suggest that the mechanisms driving daily and annual NEE variability tend to be correctly simulated, but the magnitude of these fluxes is often erroneous, suggesting that model parameterization must be improved. Few NACP models correctly predicted fluxes on seasonal and interannual timescales where spectral energy in NEE observations tends to be low, but where phenological events, multi-year oscillations in climatological drivers, and ecosystem succession are known to be important for determining ecosystem function. Mechanistic improvements to models must be made to replicate observed NEE variability on seasonal and interannual timescales.",
author = "Stoy, {P. C.} and Dietze, {M. C.} and Richardson, {A. D.} and R. Vargas and Barr, {A. G.} and Anderson, {R. S.} and Arain, {M. A.} and Baker, {I. T.} and Black, {T. A.} and Chen, {J. M.} and Cook, {R. B.} and Gough, {C. M.} and Grant, {R. F.} and Hollinger, {D. Y.} and Izaurralde, {R. C.} and Kucharik, {C. J.} and P. Lafleur and Law, {B. E.} and S. Liu and E. Lokupitiya and Y. Luo and Munger, {J. W.} and C. Peng and B. Poulter and Price, {D. T.} and Ricciuto, {D. M.} and Riley, {W. J.} and Sahoo, {A. K.} and K. Schaefer and Schwalm, {Christopher R} and H. Tian and H. Verbeeck and E. Weng",
year = "2013",
doi = "10.5194/bg-10-6893-2013",
language = "English (US)",
volume = "10",
pages = "6893--6909",
journal = "Biogeosciences",
issn = "1726-4170",
publisher = "European Geosciences Union",
number = "11",

}

TY - JOUR

T1 - Evaluating the agreement between measurements and models of net ecosystem exchange at different times and timescales using wavelet coherence

T2 - An example using data from the North American Carbon Program Site-Level Interim Synthesis

AU - Stoy, P. C.

AU - Dietze, M. C.

AU - Richardson, A. D.

AU - Vargas, R.

AU - Barr, A. G.

AU - Anderson, R. S.

AU - Arain, M. A.

AU - Baker, I. T.

AU - Black, T. A.

AU - Chen, J. M.

AU - Cook, R. B.

AU - Gough, C. M.

AU - Grant, R. F.

AU - Hollinger, D. Y.

AU - Izaurralde, R. C.

AU - Kucharik, C. J.

AU - Lafleur, P.

AU - Law, B. E.

AU - Liu, S.

AU - Lokupitiya, E.

AU - Luo, Y.

AU - Munger, J. W.

AU - Peng, C.

AU - Poulter, B.

AU - Price, D. T.

AU - Ricciuto, D. M.

AU - Riley, W. J.

AU - Sahoo, A. K.

AU - Schaefer, K.

AU - Schwalm, Christopher R

AU - Tian, H.

AU - Verbeeck, H.

AU - Weng, E.

PY - 2013

Y1 - 2013

N2 - Earth system processes exhibit complex patterns across time, as do the models that seek to replicate these processes. Model output may or may not be significantly related to observations at different times and on different frequencies. Conventional model diagnostics provide an aggregate view of model-data agreement, but usually do not identify the time and frequency patterns of model-data disagreement, leaving unclear the steps required to improve model response to environmental drivers that vary on characteristic frequencies. Wavelet coherence can quantify the times and timescales at which two time series, for example time series of models and measurements, are significantly different. We applied wavelet coherence to interpret the predictions of 20 ecosystem models from the North American Carbon Program (NACP) Site-Level Interim Synthesis when confronted with eddy-covariance-measured net ecosystem exchange (NEE) from 10 ecosystems with multiple years of available data. Models were grouped into classes with similar approaches for incorporating phenology, the calculation of NEE, the inclusion of foliar nitrogen (N), and the use of model-data fusion. Models with prescribed, rather than prognostic, phenology often fit NEE observations better on annual to interannual timescales in grassland, wetland and agricultural ecosystems. Models that calculated NEE as net primary productivity (NPP) minus heterotrophic respiration (HR) rather than gross ecosystem productivity (GPP) minus ecosystem respiration (ER) fit better on annual timescales in grassland and wetland ecosystems, but models that calculated NEE as GPP minus ER were superior on monthly to seasonal timescales in two coniferous forests. Models that incorporated foliar nitrogen (N) data were successful at capturing NEE variability on interannual (multiple year) timescales at Howland Forest, Maine. The model that employed a model-data fusion approach often, but not always, resulted in improved fit to data, suggesting that improving model parameterization is important but not the only step for improving model performance. Combined with previous findings, our results suggest that the mechanisms driving daily and annual NEE variability tend to be correctly simulated, but the magnitude of these fluxes is often erroneous, suggesting that model parameterization must be improved. Few NACP models correctly predicted fluxes on seasonal and interannual timescales where spectral energy in NEE observations tends to be low, but where phenological events, multi-year oscillations in climatological drivers, and ecosystem succession are known to be important for determining ecosystem function. Mechanistic improvements to models must be made to replicate observed NEE variability on seasonal and interannual timescales.

AB - Earth system processes exhibit complex patterns across time, as do the models that seek to replicate these processes. Model output may or may not be significantly related to observations at different times and on different frequencies. Conventional model diagnostics provide an aggregate view of model-data agreement, but usually do not identify the time and frequency patterns of model-data disagreement, leaving unclear the steps required to improve model response to environmental drivers that vary on characteristic frequencies. Wavelet coherence can quantify the times and timescales at which two time series, for example time series of models and measurements, are significantly different. We applied wavelet coherence to interpret the predictions of 20 ecosystem models from the North American Carbon Program (NACP) Site-Level Interim Synthesis when confronted with eddy-covariance-measured net ecosystem exchange (NEE) from 10 ecosystems with multiple years of available data. Models were grouped into classes with similar approaches for incorporating phenology, the calculation of NEE, the inclusion of foliar nitrogen (N), and the use of model-data fusion. Models with prescribed, rather than prognostic, phenology often fit NEE observations better on annual to interannual timescales in grassland, wetland and agricultural ecosystems. Models that calculated NEE as net primary productivity (NPP) minus heterotrophic respiration (HR) rather than gross ecosystem productivity (GPP) minus ecosystem respiration (ER) fit better on annual timescales in grassland and wetland ecosystems, but models that calculated NEE as GPP minus ER were superior on monthly to seasonal timescales in two coniferous forests. Models that incorporated foliar nitrogen (N) data were successful at capturing NEE variability on interannual (multiple year) timescales at Howland Forest, Maine. The model that employed a model-data fusion approach often, but not always, resulted in improved fit to data, suggesting that improving model parameterization is important but not the only step for improving model performance. Combined with previous findings, our results suggest that the mechanisms driving daily and annual NEE variability tend to be correctly simulated, but the magnitude of these fluxes is often erroneous, suggesting that model parameterization must be improved. Few NACP models correctly predicted fluxes on seasonal and interannual timescales where spectral energy in NEE observations tends to be low, but where phenological events, multi-year oscillations in climatological drivers, and ecosystem succession are known to be important for determining ecosystem function. Mechanistic improvements to models must be made to replicate observed NEE variability on seasonal and interannual timescales.

UR - http://www.scopus.com/inward/record.url?scp=84887280194&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887280194&partnerID=8YFLogxK

U2 - 10.5194/bg-10-6893-2013

DO - 10.5194/bg-10-6893-2013

M3 - Article

VL - 10

SP - 6893

EP - 6909

JO - Biogeosciences

JF - Biogeosciences

SN - 1726-4170

IS - 11

ER -