Responses of gross primary productivity to diffuse radiation at global FLUXNET sites

Zhou, Hao; Yue, Xu; Lei, Yadong; Zhang, Tianyi; Tian, Chenguang; Ma, Yimian; Cao, Yang

doi:10.1016/j.atmosenv.2020.117905

Cited by 49 publications

(64 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This study provided the global estimates of diffuse fertilization effects with hourly integration that better captures the variability of diffuse radiation induced by cloud and aerosols (Zhang, Boucher, et al, 2020). Furthermore, the derived hourly K d is validated at independent monitoring sites (Zhou, Yue, Lei, Zhang, et al, 2021), the modeled diffuse fertilization efficiency is comparable to observations (Figure 2), and the simulated GPP shows lower biases with updated K d relative to measurements (Figures 3 and 4). With these evaluations, we estimated that diffuse radiation accounts for 64.3% of the global GPP during 1981-2015.…”

Section: Conclusion and Discussionmentioning

confidence: 69%

“…Although ANN improves the underestimation of K d compared to original MERRA (Figure 1), the selection of training datasets may influence the performance of ANN model. As we showed in Zhou, Yue, Lei, and Zhang et al (2021), we randomly select hourly data from 15 FLUXNET sites with more than four-year observations to train the ANN model. Most of these sites are located in Europe, leading to limited spatial representations on the global scale.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…All these variables are interpolated to the resolution of 1.3° longitude by 1° latitude. We use MERRA instead of its updated version 2 (MERRA2) not only because the former was used as input for training the ANN model in Zhou, Yue, Lei, and Zhang et al (2021), but also that the MERRA2 shows larger biases in diffuse radiation than MERRA as we show in the discussion section.…”

Section: Merra Reanalyzesmentioning

confidence: 99%

“…Observations have shown that diffuse radiation promotes carbon assimilation rate more efficiently than direct radiation, because it increases light availability for shaded leaves (Cirino et al, 2014;X. Wang et al, 2018;Zhou, Yue, Lei, Zhang, et al, 2021). For example, Rocha et al (2004) explored the canopy photosynthetic responses to different sky conditions and found that GPP increases by 26% on cloudy days than that on clear days owing the large enhancement of diffuse radiation.…”

mentioning

confidence: 99%

“…For example, Rocha et al (2004) explored the canopy photosynthetic responses to different sky conditions and found that GPP increases by 26% on cloudy days than that on clear days owing the large enhancement of diffuse radiation. Similarly, Zhou, Yue, Lei, and Zhang et al (2021) estimated this diffuse fertilization effect based on observed carbon fluxes and derived diffuse radiation at global FLUXNET sites, and found that diffuse radiation leads to the enhancement of GPP by 0.16%-1.01% per 1 W m −2 radiation, which is about 2.5 times of that caused by direct radiation. Simulations also revealed important impacts of diffuse radiation on terrestrial carbon uptake (Gu et al, 2002;Mercado et al, 2009;Yue & Unger, 2018).…”

mentioning

confidence: 99%

See 4 more Smart Citations

Large Contributions of Diffuse Radiation to Global Gross Primary Productivity During 1981–2015

Zhou

Yue

Lei

et al. 2021

Global Biogeochemical Cycles

Self Cite

View full text Add to dashboard Cite

Ecosystems play an important role in global carbon cycle and climate change (Nemani et al., 2003;S. Q. Wang et al., 2015). Terrestrial gross primary productivity (GPP) is the primary driver of the global carbon cycle (Yuan et al., 2014;Zhang, Song, et al., 2016). Moderate fluctuations in terrestrial production can result in substantial variabilities for global warming (Cai et al., 2014). Accurate quantification of GPP can promote the understanding of the feedback between terrestrial biosphere and climate systems (Xiao et al., 2004).As a traditional method, eddy covariance technique derives site-level GPP as the difference between measured net ecosystem exchange (NEE) and estimated respiration (Reichstein et al., 2005). These GPP measurements are useful for exploring regional carbon budget and validating vegetation models (Mercado et al., 2009;Moore et al., 2018). However, the limited spatiotemporal coverage of site-level observations restricts the assessment of carbon fluxes at large scales and/or over long-term periods (Ran et al., 2016). Satellite remote sensing provides continuous retrieval of ecosystem productivity on a much wider scale than the flux tower data (Running et al., 2000). However, such products are not direct measurements, and as a result may introduce biases in the estimate of global GPP (Li & Xiao, 2019). Therefore, dynamic global vegetation models (DGVMs) have been developed for estimating regional to global carbon budget, in combination with site-level and remote sensing data (He et al., 2013;.The DGVMs help estimate GPP through the modeling of complex biophysical and biogeochemical processes but with uncertainties from multiple sources, including parameterization schemes (Zaehle et al., 2005), model framework (Song & Zeng, 2014), and/or input data (Wu et al., 2017). The first two types of uncertainties are related to model inherent characteristics, and can be improved step by step with the developed understanding of physical processes or better data constraints (Dietze et al., 2014;Raczka et al., 2018). The latter one, however, is related to the external forcing and can be improved independent of model development. Most of DGVMs are forced with climate reanalyzes, which were derived by assimilating numerical

show abstract

Section: Conclusion and Discussionmentioning

confidence: 69%

Section: Conclusion and Discussionmentioning

confidence: 99%

Section: Merra Reanalyzesmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations