基于多源遥感数据的锡尔河中下游农田土壤水分反演

机器学习结合多源遥感数据反演土壤水分含量（SMC）是目前SMC研究的热点,因较少考虑温度、蒸散等重要SMC影响因子,反演结果存在一定的不确定性。利用Sentinel-1影像、MODIS产品和SRTM数据,提取雷达后向散射系数等32个SMC影响因子,经相关分析选择27个显著的SMC影响因子（P<0.05）作为反演因子,并设计三组因子组合。这三组因子组合分别与随机森林、支持向量回归、BP神经网络三种机器学习方法结合,发现基于随机森林结合所有因子的方案,其SMC反演精度最高,该组合均方根误差RMSE为0.039 m³/m³,将该方案被用于反演2017年生长季锡尔河流域中下游平原区农田SMC。结果表明：从上部至下部SMC总体呈逐渐增加的态势,但存在显著时空差异,春季和秋季SMC较高而夏季较低。SMC差异主要由土壤质地、热量条件和地表植被状况差异引起。春季平原区下部农田SMC要高于上部,SMC的主控因子是土壤质地和地表植被状况;在夏季,土壤水分的主控因子是热量条件,农田灌溉弥补了热量条件差异对土壤水分的影响,导致空间上平原上部和下部土壤SMC空间差异不显著;秋季SMC的主控因子植被状况抵消地表温度和土壤质地差异对SMC的影响,使得秋季SMC空间差异不显著。本文采用的研究方法在一定程度上克服了因考虑SMC影响因子不足而获取更高SMC精度的限制。     

基于可解释性机器学习的城市O3驱动因素挖掘

受前体物排放和气象条件等因素共同驱动,大气臭氧(O₃)已成为影响城市夏季环境空气质量的主要污染物.目前物理化学机制驱动的演绎模型在进行O₃污染解析时需要的模型参数众多,运算时效性较差;数据驱动的归纳模型运算效率高,但存在可解释性差等问题.通过建立可解释性数据驱动的Correlation-ML-SHAP模型,Correlation模块挖掘O₃浓度关联影响因素,机器学习ML模块耦合可解释性SHAP模块计算各驱动因素对O₃浓度的影响贡献,实现对驱动因素的定量解析,并以晋城市2021年夏季O₃污染过程为例开展应用研究.结果表明,Correlation-ML-SHAP模型能够挖掘并利用强驱动因素模拟O₃浓度和量化影响贡献,其中ML模块采用XGBoost模型模拟准确度最佳. 2021年夏季晋城市O₃污染强驱动因素为：气温、日照强度、湿度和前体物排放水平,贡献权重为：32.1%、 21.3%、 16.5%和15.6%,其中气温、日照强度和前体物排放...     

Culturally correct song dialects are correlated with male age and female song preferences in wild populations of brown-headed cowbirds

Variation in vocal behavior among local populations of songbirds may be significant to females in mate choice. In a study of dialect populations of brown-headed cowbirds, estradiol-implanted females from two dialects held the pre-copulatory lordosis posture longer in response to playback of the local flight whistle song than to foreign whistle types from adjacent and more distant dialects. Females were held in captivity for a relatively brief period prior to testing and received no tutoring so discrimination was based solely on experience in the wild. This is the first study to show evidence of discrimination by female cowbirds based on flight whistle type. Evidence is presented from one of the study dialects that the majority of yearling male cowbirds are vocally distinct from resident adults in having either foreign or incomplete local whistles. Although these yearlings are fully mature sexually, they have little or no mating success. The correlation between female whistle preference and male mating success suggests that the ability to give the correct local whistle type may be a characteristic used by females to assess age and quality of a potential mate. A learned female preference for the predominant local song type may be a factor in the stability of these dialects by making it adaptive for males to conform to the local dialect.     

Proposal of Indicators to Assess Urban Sustainability in Brazil

The research related here focuses on municipal situations where problems of sustainability may be defined and tackled on the basis of bottom-up management procedures with the participation of organized society. The aim is to build management models that may be implemented with reasonable administrative effort and cost. Implementation pursues the target of approaching a sustainable situation in the municipality. Environmental parameters for an ideal municipality with undefined geographical location are presented. For each parameter, a set of indicators is developed that can measure the prospect of sustainability. The indicators are defined in terms of numbers or literal concepts according to the possibility of measurement. Examples of management models are presented, which are able to approach the situation defined as sustainable by the indicators. The basic management tool is the learning curve of targeted communities, which is experimentally developed and applied. The indicators are grouped according to their range of applicability.     

Adaptation or conflict? Responses to climate change in water management in Bangladesh

The potential of climate change to impact local conflict and cooperation over natural resources has received relatively little attention. Bangladesh floodplains are highly vulnerable to environmental stresses that are worsening with climate change, and community organisations have to respond to water insecurity − seasonally too little or too much. Two case studies based on action research in contrasting water and climate stressed floodplain environments in Bangladesh investigate local conflicts over water management that worsened when water regimes changed. By overcoming conflicts and improving adaptation for all local actors the cases reveal the importance of local knowledge, innovations in institutions, external facilitation, and incentives provided by disadvantaged groups who contribute towards costs in return for a share in decision making power and better adapted water management. The cases show how community organisations diversified their responsibilities and took up the challenge of water management to address local priorities and overcome conflicts. Without a more flexible and enabling approach, public investments in adaptation are likely to focus on strengthening existing water management infrastructure without understanding local social interactions and complexity. This may strengthen elite dominance and local conflicts if there is no comparable investment in developing robust and fair local institutions.     

INVESTMENT SEQUENCING RECOGNIZING EXTERNALITIES IN WATER DESALTING1

ABSTRACT. The failure to recognize the learning process in new technologies such as desalting may lead to incorrect water resource investment decisions for two reasons. First, to neglect cost reductions stemming from “learning by doing” implies an overestimation of desalting costs. Second, since learning in a particular plant may result in external (learning) benefits to other plants, these may serve as the basis for a subsidy intended to internalize such benefits. Accordingly, the research reported below includes an estimation of learning functions for desalting and the results of a formulation designed to measure external benefits on the basis of these learning functions. These results are then incorporated into a decision framework for water resource investments which recognizes uncertainty in determining optimal timing of desalting construction.     

Associative learning of odor with food- or blood-meal by Culex quinquefasciatus Say (Diptera: Culicidae)

The ability of many insects to learn has been documented. However, a limited number of studies examining associative learning in medically important arthropods has been published. Investigations into the associative learning capabilities of Culex quinquefasciatus Say were conducted by adapting methods commonly used in experiments involving Hymenoptera. Male and female mosquitoes were able to learn a conditioned stimulus that consisted of an odor not normally encountered in nature (synthetic strawberry or vanilla extracts) in association with an unconditioned stimulus consisting of either a sugar (males and females) or blood (females) meal. Such information could lead to a better understanding of the ability of mosquitoes to locate and select host and food resources in nature.     

Song ranging by the dusky antbird,Cercomacra tyrannina: ranging without song learning

In a population of dusky antbirds (Cercomacra tyrannina), less aggressive responses to distance-degraded playbacks than to undegraded playbacks of pair duets show that this tropical suboscine passerine uses sound degradation to range distance from singing conspecifics. This is the first example of song-ranging in a species that does not learn songs, supporting the hypothesis that ranging preceded the song learning that occurs in more recently evolved passerine birds (oscines). Both sexes sing and are able to use song degradation to range distance from singers when their sex-specific song is played back. Received: 27 May 1994/Accepted after revision: 24 March 1996     

A new synergetic paradigm in environmental numerical modeling: Hybrid models combining deterministic and machine learning components

A new type of environmental numerical models, hybrid environmental numerical models (HEMs) based on combining deterministic modeling and machine learning components, is introduced and formulated. Conceptual and practical possibilities of developing HEM, as an optimal synergetic combination of the traditional deterministic/first principles modeling (like that used for solving PDEs on the sphere representing model dynamics of global climate models) and machine learning components (like accurate and fast neural network emulations of model physics or chemistry processes), are discussed. Examples of developed HEMs (hybrid climate models and a hybrid wind–wave ocean model) illustrate the feasibility and efficiency of the new approach for modeling extremely complex multidimensional systems.     

Kinship, familiarity and social status modulate social learning about “micropredators” (biting flies) in deer mice

Animals can learn to recognize and respond to dangerous, threatening factors through either individual or social learning, whereby an individual learns and acquires the defensive behaviors and avoidance responses of another. Here we show that kinship, familiarity, and relative dominance of the interacting individuals affect social learning of defensive responses to micropredators in deer mice, Peromyscus maniculatus. Brief exposure of individual male deer mice to biting flies (stable fly, Stomoxys calcitrans) induced a decrease in pain sensitivity, or analgesia, and active self-burying avoidance responses. The defensive analgesic responses and their fear/anxiety/stress associated correlates facilitate the display of behavioral avoidance responses. Preparatory analgesia and avoidance responses were evident 1–3 days later when the mice were exposed to biting flies that were altered to be incapable of biting but were not displayed to similar-sized non-biting house flies. These anticipatory avoidance and analgesic responses to biting flies were also acquired through social learning—without direct individual aversive experience with biting flies. Fly-naive mice (observers) that witnessed other mice (demonstrators) being attacked by biting flies but themselves were not bitten did not display any avoidance responses. However, when exposed 1–3 days later to altered flies with biting mouth parts removed, the observers displayed socially acquired analgesic and self-burying avoidance responses. Observers whose demonstrators were either kin (siblings) and, or members of a familiar pair (kin or non-kin) displayed enhanced social learning of defensive responses. Also, within the familiar pairs, social status affected learning with subordinate observers displaying better social learning than dominant observers. These findings indicate that kinship, familiarity and social status modulate social learning of defensive responses to, and the recognition of, dangerous and threatening stimuli, likely including that of predators.Communicated by G. Wilkinson