ResoNet:Robust and Explainable ENSO Forecasts with Hybrid Convolution and Transformer Networks

Recent studies have shown that deep learning(DL)models can skillfully forecast El Niño–Southern Oscillation(ENSO)events more than 1.5 years in advance.However,concerns regarding the reliability of predictions made by DL methods persist,including potential overfitting issues and lack of interpretability.Here,we propose ResoNet,a DL model that combines CNN(convolutional neural network)and transformer architectures.This hybrid architecture enables our model to adequately capture local sea surface temperature anomalies as well as long-range inter-basin interactions across oceans.We show that ResoNet can robustly predict ENSO at lead times of 19 months,thus outperforming existing approaches in terms of the forecast horizon.According to an explainability method applied to ResoNet predictions of El Niño and La Niña from 1-to 18-month leads,we find that it predicts the Niño-3.4 index based on multiple physically reasonable mechanisms,such as the recharge oscillator concept,seasonal footprint mechanism,and Indian Ocean capacitor effect.Moreover,we demonstrate for the first time that the asymmetry between El Niño and La Niña development can be captured by ResoNet.Our results could help to alleviate skepticism about applying DL models for ENSO prediction and encourage more attempts to discover and predict climate phenomena using AI methods.

The Forecast Skills and Predictability Sources of Marine Heatwaves in the NUIST-CFS1.0 Hindcasts

Using monthly observations and ensemble hindcasts of the Nanjing University of Information Science and Technology Climate Forecast System(NUIST-CFS1.0) for the period 1983–2020, this study investigates the forecast skill of marine heatwaves(MHWs) over the globe and the predictability sources of the MHWs over the tropical oceans. The MHW forecasts are demonstrated to be skillful on seasonal-annual time scales, particularly in tropical oceans. The forecast skill of the MHWs over the tropical Pacific Ocean(TPO) remains high at lead times of 1–24 months, indicating a forecast better than random chance for up to two years. The forecast skill is subject to the spring predictability barrier of El Nino-Southern Oscillation(ENSO). The forecast skills for the MHWs over the tropical Indian Ocean(TIO), tropical Atlantic Ocean(TAO), and tropical Northwest Pacific(NWP) are lower than that in the TPO. A reliable forecast at lead times of up to two years is shown over the TIO, while a shorter reliable forecast window(less than 17 months) occurs for the TAO and NWP.Additionally, the forecast skills for the TIO, TAO, and NWP are seasonally dependent. Higher skills for the TIO and TAO appear in boreal spring, while a greater skill for the NWP emerges in late summer-early autumn. Further analyses suggest that ENSO serves as a critical source of predictability for MHWs over the TIO and TAO in spring and MHWs over the NWP in summer.

Downscaling Seasonal Precipitation Forecasts over East Africa with Deep Convolutional Neural Networks

This study assesses the suitability of convolutional neural networks(CNNs) for downscaling precipitation over East Africa in the context of seasonal forecasting. To achieve this, we design a set of experiments that compare different CNN configurations and deployed the best-performing architecture to downscale one-month lead seasonal forecasts of June–July–August–September(JJAS) precipitation from the Nanjing University of Information Science and Technology Climate Forecast System version 1.0(NUIST-CFS1.0) for 1982–2020. We also perform hyper-parameter optimization and introduce predictors over a larger area to include information about the main large-scale circulations that drive precipitation over the East Africa region, which improves the downscaling results. Finally, we validate the raw model and downscaled forecasts in terms of both deterministic and probabilistic verification metrics, as well as their ability to reproduce the observed precipitation extreme and spell indicator indices. The results show that the CNN-based downscaling consistently improves the raw model forecasts, with lower bias and more accurate representations of the observed mean and extreme precipitation spatial patterns. Besides, CNN-based downscaling yields a much more accurate forecast of extreme and spell indicators and reduces the significant relative biases exhibited by the raw model predictions. Moreover, our results show that CNN-based downscaling yields better skill scores than the raw model forecasts over most portions of East Africa. The results demonstrate the potential usefulness of CNN in downscaling seasonal precipitation predictions over East Africa,particularly in providing improved forecast products which are essential for end users.

谷物和油料中甲酸乙酯和硫化羰本底水平评估

甲酸乙酯和硫化羰是潜在的储藏物熏蒸剂,广泛存在于大气、植物和土壤中。它们在新收获的谷物和储存的小麦、大麦、燕麦和油菜籽中的本底含量因商品、温度、含水量和储存期而异,其中甲酸乙酯含量为0.5~2.0 mg/kg,硫化羰含量为0.02~1.0 mg/kg。两种成分含量在储存后的前4~5个月呈上升趋势,然后下降,在温度高于20℃,谷物和油菜籽含水量分别大于9.5%和5%时变化最明显。这些本底含量研究对开发甲酸乙酯和硫化羰作为储藏物熏蒸剂时制定最大残留限量(MRL)和市场接受度值有重要参考价值。

Changes in the Boreal Summer Intraseasonal Oscillation under Global Warming in CMIP6 Models

Changes in the activities of the Boreal Summer Intraseasonal Oscillation(BSISO)at the end of 21st century under the SSP5-8.5 scenario are assessed by adopting 17 CMIP6 models and the weak-temperature-gradient assumption.Results show that the intraseasonal variations become more structured.The BSISO-related precipitation anomaly shows a larger zonal scale and propagates further northward.However,there is no broad agreement among models on the changes in the eastward and northward propagation speeds and the frequency of individual phases.In the western North Pacific(WNP),the BSISO precipitation variance is significantly increased,at 4.62%K^(−1),due to the significantly increased efficiency of vertical moisture transport per unit of BSISO apparent heating.The vertical velocity variance is significantly decreased,at−3.51%K^(−1),in the middle troposphere,due to the significantly increased mean-state static stability.Changes in the lower-level zonal wind variance are relatively complex,with a significant increase stretching from the northwestern to southeastern WNP,but the opposite in other regions.This is probably due to the combined impacts of the northeastward shift of the BSISO signals and the reduced BSISO vertical velocity variance under global warming.Changes in strong and normal BSISO events in the WNP are also compared.They show same-signed changes in precipitation and large-scale circulation anomalies but opposite changes in the vertical velocity anomalies.This is probably because the precipitation anomaly of strong(normal)events changes at a rate much larger(smaller)than that of the meanstate static stability,causing enhanced(reduced)vertical motion.

Direct-Seeded Rice: Genetic Improvement of Game-Changing Traits for Better Adaption

The sustainability of rice production continues to be a subject of uncertainty and inquiry attributed to shifts in climatic conditions. In light of the impending climate change crisis and the high labor and water costs accompanying it, direct-seeded rice(DSR) is unquestionably one of the most practical solutions. Despite its resource and climate-friendly advantages, early maturing rice faces weed competitiveness and seedling establishment challenges. Resolving these issues is crucial for promoting its wider adoption among farmers, presenting it as a more effective sustainable rice cultivation method globally. Diverse traditional and contemporary breeding methods are employed to mitigate the limitations of the DSR approach, leveraging advanced techniques such as speed breeding and genome editing. Focusing on key traits like mesocotyl length elongation, early seedling vigor, root system architecture, and weed competitiveness holds promise for transformative improvements in DSR adaptation at a broader scale within farming communities. This review aims to summarize how these features contribute to increased crop production in DSR conditions and explore the research efforts focusing on enhancing DSR adaptation through these traits. Emphasizing the pivotal role of these game-changing traits in DSR adaptation, our analysis sheds light on their potential transformative impact and offers valuable insights for advancing DSR practices.

Clinical Application of Preliminary Breast Cancer Screening for Dense Breasts Using Real-Time AI-Powered Ultrasound with Deep-Learning Computer Vision

Objective:We propose a solution that is backed by cloud computing,combines a series of AI neural networks of computer vision;is capable of detecting,highlighting,and locating breast lesions from a live ultrasound video feed,provides BI-RADS categorizations;and has reliable sensitivity and specificity.Multiple deep-learning models were trained on more than 300,000 breast ultrasound images to achieve object detection and regions of interest classification.The main objective of this study was to determine whether the performance of our Al-powered solution was comparable to that of ultrasound radiologists.Methods:The noninferiority evaluation was conducted by comparing the examination results of the same screening women between our AI-powered solution and ultrasound radiologists with over 10 years of experience.The study lasted for one and a half years and was carried out in the Duanzhou District Women and Children's Hospital,Zhaoqing,China.1,133 females between 20 and 70 years old were selected through convenience sampling.Results:The accuracy,sensitivity,specificity,positive predictive value,and negative predictive value were 93.03%,94.90%,90.71%,92.68%,and 93.48%,respectively.The area under the curve(AUC)for all positives was 0.91569 and the AUC for all negatives was 0.90461.The comparison indicated that the overall performance of the AI system was comparable to that of ultrasound radiologists.Conclusion:This innovative AI-powered ultrasound solution is cost-effective and user-friendly,and could be applied to massive breast cancer screening.

Higher Grain-Filling Rate in Inferior Spikelets of Tolerant Rice Genotype Offset Grain Yield Loss under Post-Anthesis High Night Temperatures

Increased nighttime respiratory losses decrease the amount of photoassimilates available for plant growth and yield. We hypothesized that the increased respiratory carbon loss under high night temperatures(HNT) could be compensated for by increased photosynthesis during the day following HNT exposure. Two rice genotypes, Vandana(HNT-sensitive) and Nagina 22(HNT-tolerant), were exposed to HNT(4 ℃ above the control) from flowering to physiological maturity. They were assessed for alterations in the carbon balance of the source(flag leaf) and its subsequent impact on grain filling dynamics and the quality of spatially differentiated sinks(superior and inferior spikelets). Both genotypes exhibited significantly higher night respiration rates. However, only Nagina 22 compensated for the high respiration rates with an increased photosynthetic rate, resulting in a steady production of total dry matter under HNT. Nagina 22 also recorded a higher grain-filling rate, particularly at 5 and 10 d after flowering, with 1.5- and 4.0-fold increases in the translocation of ^(14)C sugars to the superior and inferior spikelets, respectively. The ratio of photosynthetic rate to respiratory rate on a leaf area basis was negatively correlated with spikelet sterility, resulting in a higher filled spikelet number and grain weight per plant, particularly for inferior grains in Nagina 22. Grain quality parameters such as head rice recovery, high-density grains, and gelatinization temperature were maintained in Nagina 22. An increase in the rheological properties of rice flour starch in Nagina 22 under HNT indicated the stability of starch and its ability to reorganize during the cooling process of product formation. Thus, our study showed that sink adjustments between superior and inferior spikelets favored the growth of inferior spikelets, which helped to offset the reduction in grain weight under HNT in the tolerant genotype Nagina 22.

Solvent engineering towards scalable fabrication of high-quality perovskite films for efficient solar modules

Over the last decade,remarkable progress has been made in metal halide perovskite solar cells(PSCs),which have been a focus of emerging photovoltaic techniques and show great potential for commercialization.However,the upscaling of small-area PSCs to large-area solar modules to meet the demands of practical applications remains a significant challenge.The scalable production of high-quality perovskite films by a simple,reproducible process is crucial for resolving this issue.Furthermore,the crystallization behavior in the solution-processed fabrication of perovskite films can be strongly influenced by the physicochemical properties of the precursor inks,which are significantly affected by the employed solvents and their interactions with the solutes.Thus,a comprehensive understanding of solvent engineering for fabricating perovskite films over large areas is urgently required.In this paper,we first analyze the role of solvents in the solution-processed fabrication of large-area perovskite films based on the classical crystal nucleation and growth mechanism.Recent efforts in solvent engineering to improve the quality of perovskite films for solar modules are discussed.Finally,the basic principles and future challenges of solvent system design for scalable fabrication of high-quality perovskite films for efficient solar modules are proposed.

Measuring the agricultural sustainability of India:An application of Pressure-State-Response(PSR)model

Analyzing agricultural sustainability is essential for designing and assessing rural development initiatives.However,accurately measuring agricultural sustainability is complicated since it involves so many different factors.This study provides a new suite of quantitative indicators for assessing agricultural sustainability at regional and district levels,involving environmental sustainability,social security,and economic security.Combining the PressureState-Response(PSR)model and indicator approach,this study creates a composite agricultural sustainability index for the 14 mainstream agro-climatic regions of India.The results of this study show that the Trans-Gengatic Plain Region(TGPR)ranks first in agricultural sustainability among India's 14 mainstream agro-climatic regions,while the Eastern Himalayan Region(EHR)ranks last.Higher livestock ownership,cropping intensity,per capita income,irrigation intensity,share of institutional credit,food grain productivity,crop diversification,awareness of minimum support price,knowledge sharing with fellow farmers,and young and working population,as well as better transportation facilities and membership of agricultural credit societies are influencing indicators responsible for higher agricultural sustainability in TGPR compared with EHR.Although,the scores of environmental sustainability indicators of EHR are quite good,its scores of social and economic security indicators are fairly low,putting it at the bottom of the rank of agricultural sustainability index among the 14 mainstream agroclimatic regions in India.This demonstrates the need of understanding agricultural sustainability in relation to social and economic dimensions.In a nation as diverse and complicated as India,it is the social structure that determines the health of the economy and environment.Last but not least,the sustainability assessment methodology may be used in a variety of India's agro-climatic regions.

Wave forecast in the Atlantic Ocean using a double-stage ConvLSTM network

海浪预报对海上运输安全至关重要.本研究提出了一种涵盖物理信息的深度学习模型Double-stage ConvLSTM(D-ConvLSTM)以改进大西洋的海浪预报.将D-ConvLSTM模型与海浪持续性预测和原始ConvLSTM模型的预测技巧进行对比.结果表明,预测误差随着预测时长的增加而增加.D-ConvLSTM模型在预测准确度方面优于前二者,且第三天预测的均方根误差低于0.4 m,距平相关系数约在0.8.此外,当使用IFS预测风替代再分析风时,能够产生相似的预测效果.这表明D-ConvLSTM模型的预测能力能够与ECMWF-WAM模式相当,且更节省计算资源和时间.

Prediction of ENSO using multivariable deep learning

本文基于残差神经网络和观测数据构建了一套深度学习多因子预报测模型,以改进厄尔尼诺-南方涛动(ENSO)的预报.该模型基于最大信息系数进行因子时空特征提取,并根据泰勒图的评估标准可自动确定关键预报因子进行预报.该模型在超前8个月以内的预报性能要优于当前传统的业务预报模式.2011–2018年间,该模型的预报性能优于多模式集成预报的结果.在超前6个月预报时效上,模型预报相关性可达0.82,标准化后的均方根误差仅为0.58°C,多模式集成预报的相关性和标准化后的均方根误差分别为0.70和0.73°C.该模型春季预报障碍问题有所缓解,并且自动选取的关键预报因子可用于解释热带和副热带热动力过程对于ENSO变化的影响.

Seasonal Forecasts of Precipitation during the First Rainy Season in South China Based on NUIST-CFS1.0

Current dynamical models experience great difficulties providing reliable seasonal forecasts of regional/local rainfall in South China.This study evaluates seasonal forecast skill for precipitation in the first rainy season(FRS,i.e.,April–June)over South China from 1982 to 2020 based on the global real-time Climate Forecast System of Nanjing University of Information Science and Technology(NUIST-CFS1.0,previously known as SINTEX-F).The potential predictability and the practical forecast skill of NUIST-CFS1.0 for FRS precipitation remain low in general.But NUIST-CFS1.0 still performs better than the average of nine international models in terms of correlation coefficient skill in predicting the interannual precipitation anomaly and its related circulation index.NUIST-CFS1.0 captures the anomalous Philippines anticyclone,which transports moisture and heat northward to South China,favoring more precipitation in South China during the FRS.By examining the correlations between sea surface temperature(SST)and FRS precipitation and the Philippines anticyclone,we find that the model reasonably captures SST-associated precipitation and circulation anomalies,which partly explains the predictability of FRS precipitation.A dynamical downscaling model with 30-km resolution forced by the large-scale circulations of the NUIST-CFS1.0 predictions could improve forecasts of the climatological states and extreme precipitation events.Our results also reveal interesting interdecadal changes in the predictive skill for FRS precipitation in South China based on the NUIST-CFS1.0 hindcasts.These results help improve the understanding and forecasts for FRS precipitation in South China.

Underestimated relationship between westerly wind bursts and ENSO in CMIP6 models

观测表明,热带太平洋的西风爆发(WWBs)在厄尔尼诺南方涛动(ENSO)的发生,发展和多样性中起着关键作用.因此,在耦合模式中真实地再现WWBs对于改进ENSO的模拟和预测有重要意义.在本研究中,作者发现CMIP6的耦合模式的集合平均能很好地再现了热带太平洋WWB发生频率的纬向分布及其年际变动.然而,大多数CMIP6模型极大低估了WWB和ENSO的线性关系.这可能是因为大多数CMIP6模式里海气耦合强度低于观测:海气耦合强度与WWB-ENSO关系的模式间相关系数高达0.91.

Functional Genes in Relation to Residual Feed Intake in Murrah Buffalo Heifers

High Feed efficiency (FE) in growing heifers has economic importance in dairy, but remains less understood in buffaloes. Feed conversion efficiency is defined as dry matter intake (DMI) per unit body weight gain and is determined as residual feed intake (RFI), i.e., the difference between actual and predicted feed intake to gain unit body weight during a feed trial run for 78 days under control feeding. A large variation was identified ranging between -0.42 to 0.35 in growing buffalo heifers (n = 40) of age between 11 to 15 months. An average daily weight gain (ADG) varied between 382.0 and 807.6 g/day when compared with the control-fed heifers at an organized buffalo farm. The whole blood transcriptome data obtained from the selected growing heifers from extremes of estimated high and low RFI efficiency were compared with the reference assembly generated from the transcriptome of multiparous buffaloes (n = 16) of diverse age of maturity, period of regaining post partum cyclicity and level of milk production. Differentially expressed genes (DEGs) were identified using the reference genome of Mediterranean water buffalo. GO: terms (Padj 0.05, FDR 0.05) enriched by annotated DEGs and biological pathways in gene network for RFI efficiency trait were identified. GO: terms specific to pre-transcriptional regulation of nucleus and Chromatin organization under Nucleoplasm, Energy balancing, Immunity, Cell signaling, ROS optimization, ATP generation through the Electron Transport chain and cell proliferation were determined. The study reveals the indicators targeting the actual metabolic changes and molecular functions underlying the feed utilization capacity of buffaloes. Estimated RFI efficiency revealed a large variation over heifers which may lower the DMI even up to 13.6% thus, enabling an increase in ADG up to 16% by involving efficient heifers in breeding plan. The study revealed a scope of high gain by selective breeding for FE in heifers. FE variants catalogued in the study are useful breed-specific RFI markers for future reference. The study contributes to the understanding of feed efficiency in buffaloes and its association with key interactive traits such as reproduction and growth. This knowledge can be utilized to develop more effective breeding programs.

Recombinase Polymerase Amplification Based Rapid Detection of Aroma Gene in Rice

Recombinase polymerase amplification(RPA) has been emerged as an alternative to PCR due to its high specificity and sensitivity in diagnostics of animal and plant pathogens. In this study, an RPA protocol was developed and validated for detection of aroma gene in rice.

Prediction of Seasonal Tropical Cyclone Activity in the NUIST-CFS1.0 Forecast System

Prediction skill for the seasonal tropical cyclone(TC)activity in the Northern Hemisphere is investigated using the coupled climate forecast system(version 1.0)of Nanjing University of Information Science and Technology(NUISTCFS1.0).This assessment is based on the seven-month(May to November)hindcasts consisting of nine ensemble members during 1982–2019.The predictions are compared with the Japanese 55-year Reanalysis and observed tropical storms in the Northern Hemisphere.The results show that the overall distributions of the TC genesis and track densities in model hindcasts agree well with the observations,although the seasonal mean TC frequency and accumulated cyclone energy(ACE)are underestimated in all basins due to the low resolution(T106)of the atmospheric component in the model.NUIST-CFS1.0 closely predicts the interannual variations of TC frequency and ACE in the North Atlantic(NA)and eastern North Pacific(ENP),which have a good relationship with indexes based on the sea surface temperature.In the western North Pacific(WNP),NUIST-CFS1.0 can closely capture ACE,which is significantly correlated with the El Ni?o–Southern Oscillation(ENSO),while it has difficulty forecasting the interannual variation of TC frequency in this area.When the WNP is further divided into eastern and western subregions,the model displays improved TC activity forecasting ability.Additionally,it is found that biases in predicted TC genesis locations lead to inaccurately represented TC–environment relationships,which may affect the capability of the model in reproducing the interannual variations of TC activity.

Controlled drainage in the Nile River delta of Egypt:a promising approach for decreasing drainage off-site effects and enhancing yield and water use efficiency of wheat

North Africa is one of the most regions impacted by water shortage.The implementation of controlled drainage(CD)in the northern Nile River delta of Egypt is one strategy to decrease irrigation,thus alleviating the negative impact of water shortage.This study investigated the impacts of CD at different levels on drainage outflow,water table level,nitrate loss,grain yield,and water use efficiency(WUE)of various wheat cultivars.Two levels of CD,i.e.,0.4 m below the soil surface(CD-0.4)and 0.8 m below the soil surface(CD-0.8),were compared with subsurface free drainage(SFD)at 1.2 m below the soil surface(SFD-1.2).Under each drainage treatment,four wheat cultivars were grown for two growing seasons(November 2018–April 2019 and November 2019–April 2020).Compared with SFD-1.2,CD-0.4 and CD-0.8 decreased irrigation water by 42.0%and 19.9%,drainage outflow by 40.3%and 27.3%,and nitrate loss by 35.3%and 20.8%,respectively.Under CD treatments,plants absorbed a significant portion of their evapotranspiration from shallow groundwater(22.0%and 8.0%for CD-0.4 and CD-0.8,respectively).All wheat cultivars positively responded to CD treatments,and the highest grain yield and straw yield were obtained under CD-0.4 treatment.Using the initial soil salinity as a reference,the soil salinity under CD-0.4 treatment increased two-fold by the end of the second growing season without negative impacts on wheat yield.Modifying the drainage system by raising the outlet elevation and considering shallow groundwater contribution to crop evapotranspiration promoted water-saving and WUE.Different responses could be obtained based on the different plant tolerance to salinity and water stress,crop characteristics,and growth stage.Site-specific soil salinity management practices will be required to avoid soil salinization due to the adoption of long-term shallow groundwater in Egypt and other similar agroecosystems.

Erratum to: An Isentropic Mass Circulation View on the ExtremeCold Events in the 2020/21Winter

Erratum to:Yu,Y.Y.,Y.F.Li,R.C.Ren,M.Cai,Z.Y.Guan,and W.Huang,2021:An isentropic mass circulation view on the extreme cold events in 2020/21 Winter.Adv.Atmos.Sci.,38(6),957−965,https://doi.org/10.1007/s00376-021-1289-2.

不同类型冷却屋顶方案对城市群热环境的缓解效果

利用耦合单层城市冠层模型的中尺度数值模式WRF/UCM,选取8组不同反照率和绿化比例的屋顶冷却方案进行敏感性试验,模拟研究不同冷却屋顶方案对长三角城市群2013年夏季城市热环境的影响,并分析其影响机制。结果表明:不同冷却屋顶方案对城市群热环境的缓解效果与屋顶参数之间呈很强的线性关系。高温热浪天气下,HR4(反照率为1.0)和GR4(屋顶绿化率为100%)方案的制冷度日数分别降低了14.7%和10.9%,节约的能源比普通夏日更多。同时,高温热浪天气会增强热岛强度,高反照率屋顶方案在白天对热岛起到更有效的缓解,热浪天气下日平均热岛强度最大可降低1.36℃。相同方案下,在高温热浪天气下的缓解效果均胜于普通夏日,平均而言,高反照率屋顶和屋顶绿化的降温效果分别增大38.5%和34.9%,增湿效果分别增大29.5%和21.9%,这主要是由于在高温热浪天气下,高反照率屋顶方案能够减少更多的净辐射通量,屋顶绿化方案能够释放更多的潜热通量。此外,城市格点密集区域的降温效果优于分散的城市区域,处于城市群中的常州区域较单独的杭州区域的降温幅度平均高32%。