Parameterization of the Ångström–Prescott formula based on machine learning benefit estimation of reference crop evapotranspiration with missing solar radiation data

Accurately estimated reference evapotranspiration (ET₀) is essential to regional water management. The FAO recommends coupling the Penman–Monteith (P-M) model with the Ångström–Prescott (A-P) formula as the standard method for ET₀ estimation with missing R_s measurements. However, its application is usually restricted by the two fundamental coefficients (a and b) of the A-P formula. This paper proposes a new method for estimating ET₀ with missing R_s by combining machine learning with physical-based P-M models (PM-ET₀). The benchmark values of the A-P coefficients were first determined at the daily, monthly, and yearly scales, and further evaluated in R_s and ET₀ estimates at 80 national R_s measuring stations. Then, three empirical models and four machine-learning methods were evaluated in estimating the A-P coefficients. Machine learning methods were also used to estimate ET₀ (ML-ET₀) to compare with the PM-ET₀. Finally, the optimal estimation method was used to estimate the A-P coefficients for the 839 regular weather stations for ET₀ estimation without R_s measurement for China. The results demonstrated a descending trend for coefficient a from northwest to southeast China, with larger values in cold seasons. However, coefficient b showed the opposite distribution as the coefficient a. The FAO has recommended a larger a but a smaller b for southeast China, which produced the region's largest R_s and ET₀ estimation errors. Additionally, the A-P coefficients calibrated at the daily scale obtained the best estimation accuracy for both R_s and ET₀, and slightly outperformed the monthly and yearly coefficients without significant difference in most cases. The machine learning methods outperformed the empirical methods for estimating the A-P coefficients, especially for the sites with extreme values. Further, ML-ET₀ outperformed the PM-ET₀ with yearly A-P coefficients but underperformed those with daily and monthly ones. This study indicates an exciting potential for combining machine learning with physical models for estimating ET₀. However, we found that using the A-P coefficients with finer time scales is unnecessary to deal with the missing R_s measurements.     

15万年以来极地冰芯、黄土、深海沉积的对比研究与全球变化

由极地冰芯记录、中纬黄土记录和深海沉积记录的对比可得，１５万年以来全球变化的基本框架为：末次间冰期，从距今１４万年开始到约１２．５万年达鼎盛，可分为五个阶段，即冰芯中三次高温期夹两次低温期；黄土中三层古土壤夹两层黄土；深海沉积中三次高海面夹两次低海面。末次冰期，距今８万年到１万年，可分为三个阶段，早、晚冰段和中间的间冰段。早、晚冰段盛期气候寒冷程度相当，但晚冰段最低温持续期更长，造成海面比早期更低。中间的间冰段，至少可分为７个亚段，即４次暖期（４次弱成土古土壤）夹３次冷期。全新世与末次间冰期对比，总体气候变化较平缓。在上述基本框架中，近几年发现，在末次冰期存在许多短周期、高频快速变化事件。这些事件，温度变化幅度达７℃，变化速率为几年到几十年，对应的环境变化速率可能更快。这些事件的引发机理目前还不清楚，有待进一步研究。     

全球变化背景下多尺度干旱过程及预测研究进展

干旱是一种周期性的气候异常,主要受气候自然变率驱动,具有发展缓慢、持续时间长、影响范围广等特征。然而,气候变化使得干旱不仅有增加趋势,其特征也在发生变化。例如,发展迅速的干旱,也称骤旱,近年来频繁发生。此外,人类活动通过改变陆地水循环,直接或间接地影响干旱过程。在全球变化背景下,干旱机制研究由海-陆-气相互作用影响气象干旱的气候动力学研究扩展到包含干旱传递过程机理认识、考虑区域尺度人类活动影响,面向农业、水利、生态等行业的农业干旱和水文干旱研究,为认识干旱可预报性、发展预测方法带来了新的挑战。本文将针对多尺度干旱过程及预测,讨论相关的研究进展及未来发展方向。     

全球海洋中Argo剖面浮标运行状况分析

A rgo剖面浮标是A rgo全球海洋实时观测网建设中的主要观测设备。自2001年国际A rgo计划组织实施以来,世界上已经有22个国家和团体在全球海洋中陆续布放了3 000多个A rgo剖面浮标,它们的寿命和运行状况一直是人们重视和关心的问题,因为这关系到何时能够实现国际A rgo计划的最终目标,更重要的是,涉及到浮标的性能、资料的可靠性。文中就这些问题进行了探讨,希望能有助于关注者对A rgo全球海洋实时观测网的建设及运行情况有一比较全面的了解。     

网络RTK中几种实用技术的分析比较

针对网络RTK中采用的几种技术作出详尽的论述,在理论分析的基础上对各种技术的优缺点进行了比较。认为合理选取网络RTK中的实用技术能更好地发挥其可靠性高、覆盖地区广和定位精度高等优势。     

基于MapX Mobile开发的个人移动导航系统

随着导航技术的发展,个人移动导航系统成为人们关注的热点。本文介绍了一种基于MapX Mobile开发的个人移动导航系统,实现GPS信号的接收和处理、GPS信号点的地图匹配。通过对地图上导航点、线的设置,可以在PDA上准确、快速地进行自主导航。     

ICESat-2/ATLAS全球高程控制点提取与分析

ICEsat-2/ATLAS是目前高程精度最高的星载激光数据,其数据覆盖全球,能够作为生产高精度全球地面参考高程的基础数据.基于ICESat-2/ATLAS全球激光数据产品ATL08,获取了全球ICESat-2陆地高程点,研究了基于参考高程数据和属性参数提取全球高程控制点的方法,并利用高精度参考高程数据验证了其精度.利...     

Diatom Response to Global Warming in Douhu Lake,Southeast China

A large number of lacustrine sedimentary records indicate that global warming is the main factor leading to significant changes in diatom communities in lakes of the northern hemisphere.However,due to the intensification of human activities since 1850,some scholars have emphasized that the increasing lake trophic level may be the main reason for the changes in diatom communities.The debate is ongoing.In order to avoid falling into the complex relationship between diatom changes and the seasonal cycle that characterizes lakes in mid and high latitudes,we chose a lake located at a low latitude,where the relationship between diatoms and temperature is not indirect but direct.The diatom record spans the past ca.100 years and reveals that the abundance of Aulacoseira granulata increased from 1900 until 1985, replacing the previously dominant Aulacoseira ambigua.These changes are in agreement with the increasing trend in global temperature.Since 1985,the percentages of the small-celled Discostella stelligera and the benthic diatom Navicula heimansioides have increased,while Aulacoseira granulata has decreased.This latest shift is caused by further global warming.We conclude that warming is the main factor leading to changing diatom communities in Douhu Lake.     

全球气候变化对红树林生态系统的影响、挑战与机遇

红树林是以红树植物为主体的常绿灌木或乔木组成的潮滩湿地木本植物群落, 具有“四高”特性(高生产力、高归还率、高分解率和高抗逆性)的典型海洋生态系统; 目前, 全球约有红树林1700万公顷, 主要分布在南北半球25℃等温线内。红树林生态系统的净初级生产力高达2000gC·m^-2·a^-1, 具有高强度的物质循环、能量流动以及丰富的生物多样性, 对热带、亚热带海洋生态系统的维持与发展起到关键作用, 并在全球变化过程中扮演着十分重要的角色。近30年来, 全球气候变化已引起了国内外学者的极大关注。红树林生态系统位于热带、亚热带海岸潮间带, 是一个脆弱的、敏感的生态系统, 也是首先受全球气候变化影响的典型海洋生态系统之一。作为全球海岸带地区应对全球气候变化最为重要的生态屏障之一, 气候变化将严重影响着全球红树林的生存和分布方式。本文将从全球变暖、海平面上升、大气中CO₂浓度的增加和极端天气4个主要方面, 揭示全球气候变化对红树林生态系统的影响与变化特征, 阐述红树林对全球变暖、海平面上升、大气中CO₂浓度增加和极端天气响应与适应的生态学机制, 并简要概述了红树林在减缓全球气候变化危害中的重要作用。全球气候变化也将为红树林的研究、保护和发展带来机遇与挑战。     

面向卫星遥感影像检索定位的深度学习全局表征模型评估与分析

如何快速获取无辅助参数卫星遥感影像地理位置是非合作方式获取的遥感影像信息充分利用的一个关键,利用影像特征的相似性对卫星遥感影像检索来实现定位,是获取无辅助参数卫星遥感影像地理位置的有效手段。为了探寻影像深度学习全局特征用于无辅助参数卫星遥感影像检索定位的可行性,建立了包括Precision@K、平均排序、特征提取时间、特征相似性计算时间、硬件消耗等,涵盖有效性、效率2个方面共计5类指标的评估体系。采用谷歌地球提供的影像数据作为基准影像,在资源三号夏季及冬季数据集上,分别利用AlexNet、VggNet、ResNet、DenseNet、EfficientNet等几种代表性的卷积神经网络预训练模型提取基准影像及查询影像的全局特征,依据评估体系中的指标,对这些网络模型的影像表征效果进行全面的量化评估与分析。试验分析结果表明,DenseNet、ResNet-18、VggNet这3个深度学习神经网络预训练模型提取的全局特征,综合表征效果较好,可有效用于卫星遥感影像检索定位;当K值取200时,DenseNet网络模型的Precision@K值可以达到59.5%,ResNet-18和VggNet网络...