Abstract. Accurate estimates of the net ecosystem CO2 exchange (NEE) would improve the understanding of the natural carbon sources and sinks and their role in the regulation of the 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 the year-round 6 hourly NEE over 1996–2018 in a pine-dominated boreal forest in southern Finland and analyze the predictability of the 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 the accuracy compared to using only the nearest grid cell and time step. Both ML methods can explain the temporal variability of the NEE in the observational site of this study with the meteorological predictors, but the GB method was more accurate. It was more effective in separating the important predictors from non-important ones, showing no signs of overfitting despite many redundant variables. The accuracy of the GB (RF), here measured mainly using cross-validated Pearson correlation coefficient between the model result and the observed NEE, was high (good), reaching a best estimate value of 0.96 (0.94) and the root mean square value of 1.18 µmol m⁻² s⁻¹ (1.35 µmol m⁻² s⁻¹). We recommend using GB instead of RF for modeling the CO2 fluxes of the ecosystems due to its better performance.

Abstract. To be able to meet global grand challenges (climate change; biodiversity loss; environmental pollution; scarcity of water, food and energy supplies; acidification; deforestation; chemicalization; pandemics), which all are closely interlinked with each other, we need comprehensive open data with proper metadata. The large data sets from ground-base in situ observations, ground and satellite remote sensing and multiscale modelling need to be utilized seamlessly. In this opinion paper, we describe the SMEAR (Station for Measuring Earth surface – Atmosphere Relations) concept. We also demonstrate its power via several examples, such as detection of new particle formation and their subsequent growth, quantifying atmosphere-ecosystem feedback loops, combining comprehensive observations with emergency science and services, as well as studying the effect of COVID restrictions on different air quality and climate variables. The future needs and the potential of comprehensive observations of the environment are summarized.

Abstract. Accurate estimates of net ecosystem CO2 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 R2 score between the model result and the observed NEE, was high, reaching a best estimate value of 0.92 for GB and 0.88 for RF. In addition to the standard RF approach, we recommend using GB for modeling the CO2 fluxes of the ecosystems due to its potential for better performance.

No abstract

Abstract. The Pan-Eurasian Experiment (PEEX) Science Plan, released in 2015, addressed a need for a holistic system understanding and outlined the most urgent research needs for sustainable development in the Artic-boreal region. Air quality in China and long-range transport of the atmospheric pollutants was also indicated as one of the most crucial topics of the research agenda. This paper summarizes results obtained during the last five years in the Northern Eurasian region. It also introduces recent observations on the air quality in the urban environments in China. The main regions of interest are the Russian Arctic, Northern Eurasian boreal forests (Siberia) and peatlands and on the mega cities in China. We frame our analysis against research themes introduced in 2015. We summarize recent progress in the understanding of the land – atmosphere – ocean systems feedbacks. Although the scientific knowledge in these regions has increased, there are still gaps in our understanding of large-scale climate-Earth surface interactions and feedbacks. This arises from limitations in research infrastructures and integrative data analyses, hindering a comprehensive system analysis. The fast-changing environment and ecosystem changes driven by climate change, socio-economic activities like the China Silk Road Initiative, and the global trends like urbanization further complicate such analyses. We recognize new topics with an increasing importance in the near future, such as enhancing biological sequestration capacity of greenhouse gases into forests and soils to mitigate the climate change and the socio-economic development to tackle air quality issues.