Search

Abstract Multiple lines of evidence confirm a widespread increase in vegetation growth across China over the past few decades. The relationship between vegetation growth and water availability is thought to be becoming stronger under climate change, that is, water constraints on vegetation growth have been increasing. However, our understanding of how water constraints have influenced these vegetation greening trends, especially those climate change‐driven ones, remains limited. Here, we conduct a comprehensive evaluation of recent water constraints and their implications for vegetation growth in China between 1982 and 2015. By analyzing the spatiotemporal patterns of the relationship between vegetation growth and water availability, we reveal recent water constraints and their changes hidden within an overall greening trend in China. Further analysis demonstrates that two climate change‐related categories, defined broadly as “climate” (e.g., air temperature, precipitation, and so on) and “CO2” (i.e., atmospheric carbon dioxide), have exerted varying levels of importance in regulating vegetation growth across different water constraints. With increasing water constraints, the proportion of the climate‐dominated area has significantly risen, while that of the CO2‐dominated area has sharply declined. Our findings highlight that water constraints can mediate the dominance of climate and atmospheric CO2 on vegetation growth. This has the great potential to exacerbate the uncertainty surrounding current and future sustainable vegetation greening trends.

Abstract The two phases of El‐Niño‐Southern Oscillation (ENSO) influence both regional and global terrestrial vegetation productivity on inter‐annual scales. However, the major drivers for the regional vegetation productivity and their controlling strengths during different phases of ENSO remain unclear. We herein disentangled the impacts of two phases of ENSO on regional carbon cycle using multiple data sets. We found that soil moisture predominantly accounts for ∼40% of the variability in regional vegetation productivity during ENSO events. Our results showed that the satellite‐derived vegetation productivity proxies, gross primary productivity from data‐driven models (FLUXCOM) and observation‐constrained ecosystem model (Carbon Cycle Data Assimilation System) generally agree in depicting the contribution of soil moisture and air temperature in modulating regional vegetation productivity. However, the ensemble of weakly constrained ecosystem models exhibits non‐negligible discrepancies in the roles of vapor pressure deficit and radiation over extra‐tropics. This study highlights the significance of water in regulating regional vegetation productivity during ENSO.

The design and development of machine learning programs require selecting appropriate data and algorithms, and coding and debugging based on specific task requirements and the programming experience of developers. However, the current knowledge structure in the field of machine learning is relatively complex, lacking systematic organization, and developers often face the problem of lack of experience when choosing algorithms and designing programs, resulting in a long development cycle and easy errors in machine learning programs. In response to the above issues, this article proposes and designs a machine learning program generation algorithm based on the AORBCO model. The program generation ability includes two sub abilities: algorithm decision-making ability and code generation ability. AD-EKG has been designed for algorithmic decision-making ability, allowing Ego to select appropriate machine learning algorithms based on datasets in massive data. This algorithm combines the characteristics of the AORBCO model's domain knowledge base, knowledge graph based recommendation algorithm, and collaborative filtering algorithm. By calculating the descriptive and structural information between the dataset and algorithm, the interaction probability between the dataset and algorithm is obtained, allowing Ego to make algorithmic decisions interaction probability based. Results of the experiment have shown that the AD-EKG algorithm can fully utilize structural and descriptive information to improve the accuracy of Ego algorithm decision-making. CodeT5-EKG has been designed for code generation capability, allowing Ego to automatically generate machine learning program code. This algorithm combines the CodeT5 generative model with the domain knowledge base of the AORBCO model, by adding auxiliary information extracted using DPR technology to the code generation task, and performing diversified fusion operations to improve code generation quality. The CodeT5-EKG algorithm combines the creativity and efficiency of generative models and DPR technology, and is an algorithm that can improve the quality of generated code while also having the advantages of generative models. The experiments have proved that the code generated by this algorithm has better quality compared to other generative models with the same number of parameters. [ABSTRACT FROM AUTHOR]