Spatiotemporal lagging of predictors improves machine learning estimates of atmosphere–forest CO<sub>2</sub> exchange

to related objects

Abstract 2 exchange (NEE) would improve the understanding of natural carbon sources and sinks and their role in the regulation of global atmospheric carbon. In this work, we use and compare the random forest (RF) and the gradient boosting (GB) machine learning (ML) methods for predicting year-round 6 h NEE over 1996–2018 in a pine-dominated boreal forest in southern Finland and analyze the predictability of NEE. Additionally, aggregation to weekly NEE values was applied to get information about longer term behavior of the method. The meteorological ERA5 reanalysis variables were used as predictors. Spatial and temporal neighborhood (predictor lagging) was used to provide the models more data to learn from, which was found to improve considerably the accuracy of both ML approaches compared to using only the nearest grid cell and time step. Both ML methods can explain temporal variability of NEE in the observational site of this study with meteorological predictors, but the GB method was more accurate. Only minor signs of overfitting could be detected for the GB algorithm when redundant variables were included. The accuracy of the approaches, measured mainly using cross-validated R 2 2 fluxes of the ecosystems due to its potential for better performance.

Location
Deutsche Nationalbibliothek Frankfurt am Main
Extent
Online-Ressource
Language
Englisch

Bibliographic citation
Spatiotemporal lagging of predictors improves machine learning estimates of atmosphere–forest CO2 exchange ; volume:20 ; number:4 ; year:2023 ; pages:897-909 ; extent:13
Biogeosciences ; 20, Heft 4 (2023), 897-909 (gesamt 13)

Creator
Kämäräinen, Matti
Tuovinen, Juha-Pekka
Kulmala, Markku
Mammarella, Ivan
Aalto, Juha
Vekuri, Henriikka
Lohila, Annalea
Lintunen, Anna

DOI
10.5194/bg-20-897-2023
URN
urn:nbn:de:101:1-2023033006131115283478
Rights
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
Last update
14.08.2025, 10:51 AM CEST

Data provider

This object is provided by:
Deutsche Nationalbibliothek. If you have any questions about the object, please contact the data provider.
Show original at data provider

Associated

Other Objects (12)

A simple model for the time evolution of the condensation sink in the atmosphere for intermediate Knudsen numbers

A simple model for the time evolution of the condensation sink in the atmosphere for intermediate Knudsen numbers

Anthropogenic CO2 emission estimates in the Tokyo metropolitan area from ground-based CO2 column observations

Anthropogenic CO<sub>2</sub> emission estimates in the Tokyo metropolitan area from ground-based CO<sub>2</sub> column observations

Measurement report: Method for evaluating CO2 emission from a cement plant by atmosphere O2 /N2 and CO2 measurements and its applicability to the detection of CO2 capture signals

Measurement report: Method for evaluating CO<sub>2</sub> emission from a cement plant by atmosphere O<sub>2 </sub>/N<sub>2</sub> and CO<sub>2</sub> measurements and its applicability to the detection of CO<sub>2</sub> capture signals

Simulating soil atmosphere exchanges and CO2 fluxes for a restored peatland

Simulating soil atmosphere exchanges and CO<sub>2</sub> fluxes for a restored peatland

Central Arctic Ocean surface-atmosphere exchange of CO2 and CH4 constrained by direct measurements

Central Arctic Ocean surface-atmosphere exchange of CO<sub>2</sub> and CH<sub>4</sub> constrained by direct measurements

CO2 and CH4 exchanges between moist moss tundra and atmosphere on Kapp Linné, Svalbard

CO<sub>2</sub> and CH<sub>4</sub> exchanges between moist moss tundra and atmosphere on Kapp Linné, Svalbard

Central Arctic Ocean surface–atmosphere exchange of CO2 and CH4 constrained by direct measurements

Central Arctic Ocean surface–atmosphere exchange of CO<sub>2</sub> and CH<sub>4</sub> constrained by direct measurements

CEDAR-GPP: spatiotemporally upscaled estimates of gross primary productivity incorporating CO2 fertilization

CEDAR-GPP: spatiotemporally upscaled estimates of gross primary productivity incorporating CO<sub>2</sub> fertilization

The Northeast Greenland shelf as a late-summer CO2 source to the atmosphere

The Northeast Greenland shelf as a late-summer CO<sub>2</sub> source to the atmosphere

Revision of the World Meteorological Organization Global Atmosphere Watch (WMO/GAW) CO2 calibration scale

Revision of the World Meteorological Organization Global Atmosphere Watch (WMO/GAW) CO<sub>2</sub> calibration scale

Revised alkenone &delta;13C based CO2 estimates during the Plio-Pleistocene

Revised alkenone &delta;<sup>13</sup>C based CO<sub>2</sub> estimates during the Plio-Pleistocene

To what extent does CO2 diurnal cycle impact carbon flux estimates in CarboScope?

To what extent does CO<sub>2</sub> diurnal cycle impact carbon flux estimates in CarboScope?

Related objects