Artificial intelligence for geoscience: Progress, challenges, and perspectives.

This paper explores the evolution of geoscientific inquiry, tracing the progression from traditional physics-based models to modern data-driven approaches facilitated by significant advancements in artificial intelligence (AI) and data collection techniques. Traditional models, which are grounded in physical and numerical frameworks, provide robust explanations by explicitly reconstructing underlying physical processes. However, their limitations in comprehensively capturing Earth's complexities and uncertainties pose challenges in optimization and real-world applicability. In contrast, contemporary data-driven models, particularly those utilizing machine learning (ML) and deep learning (DL), leverage extensive geoscience data to glean insights without requiring exhaustive theoretical knowledge. ML techniques have shown promise in addressing Earth science-related questions. Nevertheless, challenges such as data scarcity, computational demands, data privacy concerns, and the "black-box" nature of AI models hinder their seamless integration into geoscience. The integration of physics-based and data-driven methodologies into hybrid models presents an alternative paradigm. These models, which incorporate domain knowledge to guide AI methodologies, demonstrate enhanced efficiency and performance with reduced training data requirements. This review provides a comprehensive overview of geoscientific research paradigms, emphasizing untapped opportunities at the intersection of advanced AI techniques and geoscience. It examines major methodologies, showcases advances in large-scale models, and discusses the challenges and prospects that will shape the future landscape of AI in geoscience. The paper outlines a dynamic field ripe with possibilities, poised to unlock new understandings of Earth's complexities and further advance geoscience exploration.

An RFI-suppressed SMOS L-band multi-angular brightness temperature dataset spanning over a decade (since 2010).

The Soil Moisture Ocean Salinity (SMOS) was the first mission providing L-band multi-angular brightness temperature (TB) at the global scale. However, radio frequency interferences (RFI) and aliasing effects degrade, when present SMOS TBs, and thus affect the retrieval of land parameters. To alleviate this, a refined SMOS multi-angular TB dataset was generated based on a two-step regression approach. This approach smooths the TBs and reconstructs data at the incidence angle with large TB uncertainties. Compared with Centre Aval de Traitement des Données SMOS (CATDS) TB product, this dataset shows a better relationship with the Soil Moisture Active Passive (SMAP) TB and enhanced correlation with in-situ measured soil moisture. This RFI-suppressed SMOS TB dataset, spanning more than a decade (since 2010), is expected to provide opportunities for better retrieval of land parameters and scientific applications.

A 21-year dataset (2000-2020) of gap-free global daily surface soil moisture at 1-km grid resolution.

Global soil moisture estimates from current satellite missions are suffering from inherent discontinuous observations and coarse spatial resolution, which limit applications especially at the fine spatial scale. This study developed a dataset of global gap-free surface soil moisture (SSM) at daily 1-km resolution from 2000 to 2020. This is achieved based on the European Space Agency - Climate Change Initiative (ESA-CCI) SSM combined product at 0.25° resolution. Firstly, an operational gap-filling method was developed to fill the missing data in the ESA-CCI SSM product using SSM of the ERA5 reanalysis dataset. Random Forest algorithm was then adopted to disaggregate the coarse-resolution SSM to 1-km, with the help of International Soil Moisture Network in-situ observations and other optical remote sensing datasets. The generated 1-km SSM product had good accuracy, with a high correlation coefficent (0.89) and a low unbiased Root Mean Square Error (0.045 m3/m3) by cross-validation. To the best of our knowledge, this is currently the only long-term global gap-free 1-km soil moisture dataset by far.

A global daily soil moisture dataset derived from Chinese FengYun Microwave Radiation Imager (MWRI)(2010-2019).

Surface soil moisture (SSM) is an important variable in drought monitoring, floods predicting, weather forecasting, etc. and plays a critical role in water and heat exchanges between land and atmosphere. SSM products from L-band observations, such as the Soil Moisture Active Passive (SMAP) Mission, have proven to be optimal global estimations. Although X-band has a lower sensitivity to soil moisture than that of L-band, Chinese FengYun-3 series satellites (FY-3A/B/C/D) have provided sustainable and daily multiple SSM products from X-band since 2008. This research developed a new global SSM product (NNsm-FY) from FY-3B MWRI from 2010 to 2019, transferred high accuracy of SMAP L-band to FY-3B X-band. The NNsm-FY shows good agreement with in-situ observations and SMAP product and has a higher accuracy than that of official FY-3B product. With this new dataset, Chinese FY-3 satellites may play a larger role and provide opportunities of sustainable and longer-term soil moisture data record for hydrological study.

The Influencing Factors of Comprehensive Sexuality Education Capacity and Its Correlation with Subjective Social Status among Chinese Junior High School Students.

(1) Objective: Evidence suggests that comprehensive sexuality education (CSE) can protect and empower younger generations to advocate for their reproductive health and wellbeing. This survey aims to investigate the current status and influencing factors of CSE among Chinese junior high school students, and to evaluate its correlation with the learning experience of sex education and subjective social status (SSS) to provide evidence for the implementation of CSE in the future. (2) Methods: A total of 4109 participants aged 11 to 16 years were recruited using data from a cross-sectional survey among junior high school students in China in 2021. CSE knowledge, attitude, and skills were used to generate the CSE comprehensive capacity by a principal component analysis. One-way ANOVA was used to assess the different effects of school sex education and family sex education. Multiple linear regression was used to assess the association between CSE comprehensive capacity and SSS. (3) Results: The average score of CSE comprehensive capacity was 82.44 ± 8.60 (with a total score of 100 points) among participants. After the adjustment, subjective social status was positively related to CSE comprehensive capacity (B = 0.28, 95% CI: 0.20-0.36), and SSS (School) (beta = 0.62) had a higher impact on CSE comprehensive capacity compared to SSS (Family) (beta = -0.10). School sex education was associated with the CSE knowledge level with a larger magnitude compared to family sex education (mean deviation = -0.53, p = 0.031), whereas family sex education was related to the CSE skill level with a greater magnitude (mean deviation =1.14, p = 0.005). (4) Conclusions: These findings suggest that sex education at school and within the family might have a different impact on CSE capacity, which was positively associated with SSS among junior high school students.

Global spatiotemporally continuous MODIS land surface temperature dataset.

Land surface temperature (LST) plays a critical role in land surface processes. However, as one of the effective means for obtaining global LST observations, remote sensing observations are inherently affected by cloud cover, resulting in varying degrees of missing data in satellite-derived LST products. Here, we propose a solution. First, the data interpolating empirical orthogonal functions (DINEOF) method is used to reconstruct invalid LSTs in cloud-contaminated areas into ideal, clear-sky LSTs. Then, a cumulative distribution function (CDF) matching-based method is developed to correct the ideal, clear-sky LSTs to the real LSTs. Experimental results prove that this method can effectively reconstruct missing LST data and guarantee acceptable accuracy in most regions of the world, with RMSEs of 1-2 K and R values of 0.820-0.996 under ideal, clear-sky conditions and RMSEs of 4-7 K and R values of 0.811-0.933 under all weather conditions. Finally, a spatiotemporally continuous MODIS LST dataset at 0.05° latitude/longitude grids is produced based on the above method.

A long term global daily soil moisture dataset derived from AMSR-E and AMSR2 (2002-2019).

Long term surface soil moisture (SSM) data with stable and consistent quality are critical for global environment and climate change monitoring. L band radiometers onboard the recently launched Soil Moisture Active Passive (SMAP) Mission can provide the state-of-the-art accuracy SSM, while Advanced Microwave Scanning Radiometer for EOS (AMSR-E) and AMSR2 series provide long term observational records of multi-frequency radiometers (C, X, and K bands). This study transfers the merits of SMAP to AMSR-E/2, and develops a global daily SSM dataset (named as NNsm) with stable and consistent quality at a 36 km resolution (2002-2019). The NNsm can reproduce the SMAP SSM accurately, with a global Root Mean Square Error (RMSE) of 0.029 m3/m3. NNsm also compares well with in situ SSM observations, and outperforms AMSR-E/2 standard SSM products from JAXA and LPRM. This global observation-driven dataset spans nearly two decades at present, and is extendable through the ongoing AMSR2 and upcoming AMSR3 missions for long-term studies of climate extremes, trends, and decadal variability.

In situ soil moisture and temperature network in genhe watershed and saihanba area in China.

The dataset presented in this article is related to the work "Evaluation and Analysis of SMAP, AMSR2, and MEaSUREs Freeze/Thaw Products in China [1]". Soil moisture and temperature are important variables of land-atmosphere energy exchange, monitoring vegetation growth, predicting drought disasters and climate and hydrological modelling [2], [3], [4], [5], [6]. This work provides detailed information on in situ soil moisture and temperature data network established in the Genhe watershed and Saihanba area in China, respectively. The Genhe watershed represents the complex surface heterogeneity in Northeast China. Therefore, data from 22 in situ sites were established in the Genhe watershed since March 2016 to improve the dynamic analysis and modeling of remotely sensed information for complex land surfaces. Saihanba is currently China's largest manmade forest and has a unique alpine wetland and a complete aquatic ecosystem. There are 29 in situ sites deployed in Saihanba since August 2018 for studying the cold temperate continental monsoon climate and estimating forest carbon storage capacity and carbon emissions from manmade forests. Soil temperature and permittivity data in the network were measured using ECH2O EC-5TM probes (Decagon Devices, Inc., Washington, USA, https://www.metergroup.com/) and XingShiTu (XST) probes (BEIJING XST Co., Ltd., www.xingshitu.com) every 30 min at depths of 3, 5, and 10 cm for the Genhe watershed continuous automatic observation network, and depths of 5 and 10 cm for the Saihanba continuous automatic observation network. In the Genhe watershed, soil moisture and soil temperature data in the network were automatically collected using the EM50 data collection system. The Saihanba area has the XST data collection system to record soil temperature and permittivity. The permittivity data collected with the XST data collector were transformed to soil moisture data (volumetric water content) based on the formula developed by [7]. The datasets of the Genhe watershed and Saihanba area consist of raw data acquired by the data collector and processed data of soil moisture and temperature. The Saihanba dataset also includes the calibration data based on soil texture. The result of temporal variations analysis in observed data in the Genhe Watershed and the processing in observed data in the saihanba area show that the long-term in situ soil moisture and temperature datasets can be used for the validation/calibration and improvement of the soil moisture and soil freeze/thaw algorithm.

Combining XCO2 measurements derived from SCIAMACHY and GOSAT for potentially generating global CO2 maps with high spatiotemporal resolution.

Global warming induced by atmospheric CO2 has attracted increasing attention of researchers all over the world. Although space-based technology provides the ability to map atmospheric CO2 globally, the number of valid CO2 measurements is generally limited for certain instruments owing to the presence of clouds, which in turn constrain the studies of global CO2 sources and sinks. Thus, it is a potentially promising work to combine the currently available CO2 measurements. In this study, a strategy for fusing SCIAMACHY and GOSAT CO2 measurements is proposed by fully considering the CO2 global bias, averaging kernel, and spatiotemporal variations as well as the CO2 retrieval errors. Based on this method, a global CO2 map with certain UTC time can also be generated by employing the pattern of the CO2 daily cycle reflected by Carbon Tracker (CT) data. The results reveal that relative to GOSAT, the global spatial coverage of the combined CO2 map increased by 41.3% and 47.7% on a daily and monthly scale, respectively, and even higher when compared with that relative to SCIAMACHY. The findings in this paper prove the effectiveness of the combination method in supporting the generation of global full-coverage XCO2 maps with higher temporal and spatial sampling by jointly using these two space-based XCO2 datasets.

Joint effects of asymmetric payoff and reciprocity mechanisms on collective cooperation in water sharing interactions: a game theoretic perspective.

Common-pool resource (CPR) dilemmas distinguish themselves from general public good problems by encompassing both social and physical features. This paper examines how a physical mechanism, namely asymmetric payoff; and a social mechanism, reciprocity; simultaneously affect collective cooperation in theoretical water sharing interactions. We present an iterative N-person game theoretic model to investigate the joint effects of these two mechanisms in a linear fully connected river system under three information assumptions. From a simple evolutionary perspective, this paper quantitatively addresses the conditions for Nash Equilibrium in which collective cooperation might be established. The results suggest that direct reciprocity increases every actor's motivation to contribute to the collective good of the river system. Meanwhile, various upstream and downstream actors manifest individual disparities as a result of the direct reciprocity and asymmetric payoff mechanisms. More specifically, the downstream actors are less willing to cooperate unless there is a high probability that long-term interactions are ensured; however, a greater level of asymmetries is likely to increase upstream actors' incentives to cooperate even though the interactions could quickly end. The upstream actors also display weak sensitivity to an increase in the total number of actors, which generally results in a reduction in the other actors' motivation for cooperation. It is also shown that the indirect reciprocity mechanism relaxes the overall conditions for cooperative Nash Equilibrium.