Estimating forest carbon fluxes using four different data-driven techniques based on long-term eddy covariance measurements: Model comparison and evaluation

Dou, Xiankang; Yang, Yongguo

doi:10.1016/j.scitotenv.2018.01.202

Cited by 46 publications

(22 citation statements)

References 108 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NN have been used to identify and model factors influencing NEE and to 385 partition NEE into ER and GPP (Moffat et al, 2010). NNs have even been used to upscale fluxes from the ecosystem level to the continental scale (Dou and Yang, 2018;Papale et al, 2003).…”

Section: Appendix A: Neural Network Analysis and Uncertainty Calculationsmentioning

confidence: 99%

Vegetation Influence and Environmental Controls on Greenhouse Gas Fluxes from a Drained Thermokarst Lake in the Western Canadian Arctic

Skeeter¹,

Christen²,

Laforce³

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract. Thermokarst features are widespread in ice-rich regions of the circumpolar Arctic. The rate of thermokarst lake formation and drainage is anticipated to accelerate as the climate warms. However, it is uncertain how these dynamic features impact the terrestrial Arctic carbon cycle. Methane (CH4) and carbon dioxide (CO2) fluxes were measured during peak growing season using eddy covariance and chambers at Illisarvik, a 0.16 km2 thermokarst lake basin that was experimentally drained in 1978 on Richards Island, Northwest Territories, Canada. Vegetation in the basin differs markedly from the surrounding dwarf-shrub tundra and included patches of tall shrubs, grasses and sedges with some bare ground and a small pond in the centre. During the study period, temperature and wind conditions were highly variable and soil water content decreased steadily. Basin scaled net ecosystem exchange (NEE) measured by eddy covariance was −1.5 [CI95 % ± 0.2] g C-CO2 m−2 d−1; NEE followed a marked diurnal pattern with no trend during the study period. NEE was primary controlled by photosynthetic photon flux density and influenced by vapor pressure deficit, volumetric water content and the presence of shrubs. By contrast, net methane exchange (NME) was low (8.7 [CI95 % ± 0.4] mg CH4 m−2 d−1 and had little impact on the carbon balance of the basin during the study period. NME displayed high spatial variability, sedge areas in the basin were the strongest source of CH4 while upland areas outside the basin were a net sink. Soil moisture and temperature were the main environmental factors influencing NME, having a positive and negative effect respectively.

show abstract

Section: Appendix A: Neural Network Analysis and Uncertainty Calculationsmentioning

confidence: 99%

Vegetation Influence and Environmental Controls on Greenhouse Gas Fluxes from a Drained Thermokarst Lake in the Western Canadian Arctic

Skeeter¹,

Christen²,

Laforce³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Although these ML models have been successful in estimating RE at different temporal and spatial scales, some uncertainties still exist. For instance, these ML models are usually constructed based on different learning principles; however, few attempts have been made to compare the predictive performance of these models in estimating RE [12]. Some advanced ML models, such as deep learning (DL) models, have not yet been tested for estimating RE.…”

Section: Introductionmentioning

confidence: 99%

Estimating Ecosystem Respiration in the Grasslands of Northern China Using Machine Learning: Model Evaluation and Comparison

Zhu

et al. 2020

Sustainability

View full text Add to dashboard Cite

While a number of machine learning (ML) models have been used to estimate RE, systematic evaluation and comparison of these models are still limited. In this study, we developed three traditional ML models and a deep learning (DL) model, stacked autoencoders (SAE), to estimate RE in northern China’s grasslands. The four models were trained with two strategies: training for all of northern China’s grasslands and separate training for the alpine and temperate grasslands. Our results showed that all four ML models estimated RE in northern China’s grasslands fairly well, while the SAE model performed best (R2 = 0.858, RMSE = 0.472 gC m−2 d−1, MAE = 0.304 gC m−2 d−1). Models trained with the two strategies had almost identical performances. The enhanced vegetation index and soil organic carbon density (SOCD) were the two most important environmental variables for estimating RE in the grasslands of northern China. Air temperature (Ta) was more important than the growing season land surface water index (LSWI) in the alpine grasslands, while the LSWI was more important than Ta in the temperate grasslands. These findings may promote the application of DL models and the inclusion of SOCD for RE estimates with increased accuracy.

show abstract

“…Due to the above-mentioned drawbacks of estimating LAI from space-borne data and crop growth models, this study obtains LAI in forests from meteorological data via employing the GEP technique. In the recent years, GEP has been applied to many hydrological problems such as estimation of solar radiation (Landeras et al, 2012), air temperature (Kisi et al, 2013a;Kisi and Shiri, 2014;Shiri et al, 2014a), pan evaporation (Kisi et al, 2012;Pour Alibaba et al, 2013;Kisi, 2015;Kim et al, 2015), forest carbon fluxes (Dou and Yang, 2018), crop evapotranspiration (Torres et al, 2011;Shiri et al, 2014b, c;Shrestha and Shukla, 2015;Karimi et al, 2016), soil parameters (Shiri et al, 2017a, b), river flows (Liu et al, 2014;Karimi et al, 2017) and river qualitative parameters (Kisi et al, 2013b). Recently, Karimi et al (2018) utilized GEP and random forest to estimate LAI in croplands and grasslands from in-situ meteorological data.…”

Section: Introductionmentioning

confidence: 99%

Estimation of Forest Leaf Area Index Using Meteorological Data: Assessment of Heuristic Models

Karimi¹,

Nazemi²,

Sadraddini³

et al. 2020

J ENVIRON INFORM

View full text Add to dashboard Cite

Leaf Area Index (LAI) is an important structural feature of our ecosystem as it affects energy, carbon, and water exchanges between the land surface and overlying atmosphere. Global scale LAI datasets have been obtained by regression, heuristic data driven, and radiative transfer models using remotely sensed land surface reflectance data. However, the estimation of LAI from remotely sensed data is limited only to clear sky conditions. Also, it is problematic to estimate LAI in forests by using conventional remote sensing image analysis of multi-spectral data. Due to the above-mentioned shortcomings of estimating LAI from remotely sensed data, this study obtained LAI from meteorological data using the Gene Expression Programming (GEP) technique. The new approach was tested in different forest sites with broad-leaf and needle-leaf trees in USA. The results showed that the GEP technique can accurately estimate LAI from meteorological data in different forest sites.

show abstract

Estimating forest carbon fluxes using four different data-driven techniques based on long-term eddy covariance measurements: Model comparison and evaluation

Cited by 46 publications

References 108 publications

Vegetation Influence and Environmental Controls on Greenhouse Gas Fluxes from a Drained Thermokarst Lake in the Western Canadian Arctic

Vegetation Influence and Environmental Controls on Greenhouse Gas Fluxes from a Drained Thermokarst Lake in the Western Canadian Arctic

Estimating Ecosystem Respiration in the Grasslands of Northern China Using Machine Learning: Model Evaluation and Comparison

Estimation of Forest Leaf Area Index Using Meteorological Data: Assessment of Heuristic Models

Contact Info

Product

Resources

About