Multi-target retrieval (MTR): the simultaneous retrieval of pressure, temperature and volume mixing ratio profiles from limb-scanning atmospheric measurements

In this paper we describe a retrieval approach for the simultaneous determination of the altitude distributions of p, T and VMR of atmospheric constituents from limb-scanning measurements of the atmosphere. This analysis method, named multi-target retrieval (MTR), has been designed and implemented in a computer code aimed at the analysis of MIPAS-ENVISAT observations; however, the concepts implemented in MTR have a general validity and can be extended to the analysis of all type of limb-scanning observations. In order to assess performance and advantages of the proposed approach, MTR has been compared with the sequential analysis system implemented by ESA as the level-2 processor for MIPAS measurements. The comparison has been performed on a common set of target species and spectral intervals. The performed tests have shown that MTR produces results of better quality than a sequential retrieval. However, the simultaneous retrieval of p, T and water VMR has not lead to satisfactory results below the tropopause, because of the high correlation occurring between p and water VMR in the troposphere. We have shown that this problem can be fixed extending the MTR analysis to at least one further target whose spectral features decouple the retrieval of pressure and water VMR. Ozone was found to be a suitable target for this purpose. The advantages of the MTR analysis system in terms of systematic errors have also been discussed.     

Task-specific deactivation patterns in functional magnetic resonance imaging

In general, image analysis of cognitive experiments using functional magnetic resonance imaging techniques has emphasized those regions of the brain where increases in signal intensity, with regard to the reference state, are associated with activation. Nevertheless, a number of recent papers have shown that there are areas of deactivation as well. In this study, we have used a univariate analysis and echo-planar functional magnetic resonance imaging to address the relationship of the reference state to the deactivations. We employed two dichotomous covert tasks, orthographic lexical retrieval and pure visual retrieval, to contrast with the reference state (baseline) of silent counting. Our analysis yielded extensive, task-specific landscapes of regional incremental and decremental responses. We have specifically demonstrated that the decremental responses are not due to activation in the reference state. We have also demonstrated that they are not an artifact of a specific part of the image analysis, and propose that they represent a physiological, task specific signal that should be considered an integral component of neural networks representing brain function.     

Partition function data and impact on retrieval quality for an mm/sub-mm limb sounder

The partition function plays a major role as it mainly governs the dependence of the spectral line strength (or line intensity) on the temperature. A wrong partition function will lead to an incorrect line strength, which will lead to a systematic error in the retrieval. Therefore, a careful investigation of the available partition function data and the sensitivity of the retrieval to this is required. Two partition function data sets, the HITRAN and the JPL, are considered in this paper. In a first step, we investigate the spread of the partition function ratios from the simple approximation values given by theory. We find that the HITRAN partition functions are usually higher than the values quoted by the simple approximation. The JPL partition functions are much closer to the values given by the approximation. Comparing directly the two datasets, with some exceptions, a good agreement is found. Only for 16 molecules (from 66 considered molecules), the deviations in the two data sets are larger than 2%. The retrieval error analysis shows that, for the case of the molecular species with strong signatures, an uncertainty in the partition function is directly translated into a retrieval error of the species in question. However, the uncertainty in the partition function of this species can have a high impact on the quality of weak species retrieval.     

AIRS红外高光谱卫星数据反演卷云光学厚度和云顶高度

基于大气红外探测器L1B红外高光谱辐射观测资料,结合中分辨率成像光谱仪(moderate resolution imaging spectroradiometer, MODIS)云产品数据,利用通用大气辐射传输模式(combined atmospheric radiative transfer model,CART),根据模式模拟和AIRS实际观测亮温的亮温差,研究从AIRS红外波段1 070～1 135 cm-1高光谱数据反演卷云的光学厚度和云顶高度。将反演的卷云光学厚度与云顶高度作为输入参数模拟计算650～1 150 cm-1波段卷云大气顶的辐射亮温谱,并将模拟值与AIRS观测亮温谱进行了对比分析。将反演的卷云光学厚度和云顶高度和AIRS的760通道(900.56 cm-1,11.1 μm)的亮温以及MODIS卷云反射率进行了对比分析。最后将反演的卷云云顶高度和MODIS云顶高度进行了对比分析。研究结果表明：反演所使用的650～1 150 cm-1波段模式模拟和观测亮温谱吻合得很好,说明CART可以较好的模拟AIRS亮温谱。反演的卷云参数与AIRS在大气窗口区的760通道(900.56 cm-1,11.1 μm)的亮温的分布满足低亮温对应较大的卷云光学厚度和高云顶高度。反演的卷云参数和MODIS卷云的反射率分布满足高卷云光学厚度和云顶高度对应高卷云反射率。反演的卷云云顶高度和MODIS的卷云云顶高度之间线性相关系数相对较高,且都在8.5～11.5 km的概率较高,两者的概率分布趋势一致。说明CART可以用于反演卷云的性质,反演结果具有一定的可靠性。     

基于三光谱带的大气散射效应校正方法

大气散射效应作为CO₂反演的主要误差源,严重影响了全球大气CO₂卫星产品的应用研究。气溶胶作用以及气溶胶与地表综合作用是大气散射的重要来源。基于O₂-A,CO₂ 1.6和2.06 μm三个光谱带中的强、弱吸收峰和连续谱,从大气气溶胶光学厚度和地表反照率的角度,分析三光谱带具有的相关信息,提出改进的全物理反演方法,对相关性很强的气溶胶光学厚度和地表反照率这两个散射相关参数进行同步反演,实现大气CO₂反演中的散射效应校正。模拟计算气溶胶影响、气溶胶和地表反照率两者综合影响导致的CO₂反演误差,并进行校正,极端情况下导致的8% CO₂反演误差可校正到1% 内,气溶胶类型差异导致最高达10%的散射影响可校正到2%内,显示了方法的有效性,同时通过对校正效果的评估,表明该方法应用于卫星数据高精度反演的潜力,也指出了实际应用时需要关注的问题。     

水雾强弥散条件下高温温场多光谱成像反演

高浓度水雾条件下的表面高温温场反演测量在航空航天、冶金铸造等工业领域有着重要的应用。由于水雾的弥散作用,高温表面的辐射透过水雾后,会出现强烈的衰减和散射,导致传统辐射测温方法出现很大误差。现有水雾弥散条件下的温场反演测量主要包括基于试验数据反推及实时测量水雾参数进行修正的测量方法,并基于辐射传输理论对测量结果进行误差分析和评估,测量方式多为单通道或双通道点辐射测温。基于水雾场红外光谱辐射特性的计算,提出了一种水雾强弥散条件下表面高温温场多光谱成像反演方法;根据辐射传输理论,考虑强弥散条件下的邻近效应,建立了相应的反演模型。在水雾场相关参数未知的情况下,通过三个透过水雾场后的高温目标长波红外光谱辐射图像,反演得到表面高温温场的真温分布。反演第一步是辐射温度场反演,即通过长波红外辐射图像,根据定标曲线和高温目标的光谱发射率先验数据,得到高温目标透过弥散水雾场经过发射率校正的辐射温度场;反演的第二步是根据三通道非线性反演模型,得到目标的真温温场分布。设计了一个长波红外三光谱通道反演测量装置,中心波长分别是8.8, 10.7和12.0 μm,对高温目标进行三个长波红外光谱通道的同时探测成像。设计了一套验证测试装置,利用标准高温黑体源和水雾弥散设备,进行了高温目标水雾弥散条件下的辐射图像采集和目标温度的反演试验。试验结果表明8～14 μm长波红外波段比短波波段对水雾弥散具有更强的抗干扰能力,在1 100和1 200 ℃典型温度点反演的平均误差在7%左右,大大减小了未经校正的辐射传输失真,适用于黑体和灰体高温目标,且无需水雾场的浓度和粒径分布等先验信息,基于多光谱成像信息的水雾弥散条件下温场反演方法具有一定的普适性和创新性。     

机器学习的IBBCEAS光谱反演波段优化

非相干宽带腔增强吸收光谱技术（IBBCEAS）利用高精密谐振腔增强吸收光程,实现对痕量气体的高灵敏探测。目前,IBBCEAS技术主要采用发光二极管（LED）作为非相干光源。当谐振腔镜片反射率曲线与带宽有限的LED辐射谱不能很好匹配时,光谱反演波段选择不当可能会对被测气体浓度拟合结果产生较大偏差。以定量探测大气NO₂浓度为例,分析了IBBCEAS光谱反演波段对NO₂拟合结果的影响,发现当反演波段宽度窄到一定程度后,NO₂浓度拟合相对误差会迅速增加。为此,提出了一种基于RBF神经网络结合遗传算法的机器学习IBBCEAS光谱反演波段优化方法,以使浓度拟合误差达到最小。在430～480 nm待选波段内,选择各种宽度和中心波长的子波段作为反演波段,分别进行NO₂浓度拟合,以此获得435个样本数据,并将样本数据按照4∶1比例分成学习样本和测试样本,分别用于RBF神经网络学习训练和测试,得到输入参数“反演波段的起始波长与截止波长”与输出参数“浓度拟合相对误差”之间的非线性映射关系。使用遗传算法搜索最优反演波段,将反演波段的起始波长和截止波长组合进行个体编码,随机产生若干个体形成种群。以RBF神经网络的输出（即浓度拟合相对误差）作为个体适应度,经过多代种群进化过程后,获得适应度最优个体,即获得最优反演波段。在种群规模为100个体,种群进化最大代数为100的情况下,当种群进化第61代时,最优个体出现,对应的最优适应度为3.584%,最优反演波段为445.78～479.44 nm。选择相同带宽的其他4个典型反演波段,与最优反演波段下的NO₂拟合结果进行了对比。结果显示,在最优反演波段下,无论是拟合误差、相对拟合误差还是拟合残差标准偏差,均低于其他4个反演波段,光谱拟合质量达到最优。结果表明,利用机器学习来确定IBBCEAS最优反演波段是可行的。     

Retrieval of cirrus particle sizes using a split-window technique: a sensitivity study

This paper examines the sensitivity of retrieved ice particle sizes using split-window method to the light scattering program for the single scattering calculation. We find that for randomly oriented hexagonal ice particles the retrieval algorithm using the anomalous diffraction theory (ADT) significantly overestimates the mean effective ice particle sizes, D_ge. The retrieved D_ge based on the geometric optics method (GOM) and Mie theory agrees with reference results within 20% when D_ge<30 μm. Based on the speculation that there is no “tunneling” for complex particles, some recent studies suggest that the ADT is an appropriate method to simulate the absorption coefficient for irregularly shaped particles in the infrared. In this study, however, we find that the overestimation of D_ge due to the ADT is largely caused by the neglect of refraction and reflection processes, instead of by the neglect of “tunneling” in the absorption calculations. By considering complex particle shapes such as aggregates with surface roughness, we further find that the retrieved D_ge based on the GOM is not sensitive to the particle shapes. Note that both ADT and GOM do not consider the “tunneling”, but the retrieved D_ge based on the ADT is about two times larger than those based on the GOM. “Tunneling” plays a significant role in the retrieved D_ge only when the D_ge is larger than 30–35 μm. In this study, we also examine the sensitivities of retrieved D_ge to the ice particle size distributions assumed in the retrieval algorithm and to the errors in the emissivities. It is found that when the D_ge is larger than 30–40 μm, the retrieved D_ge becomes very sensitive to the uncertainties related to the ice particle size distributions and to the errors in cirrus emissivities derived from measurements.     

温度对高温超导材料(HTS)性能的影响

超导电力技术将为电力工业带来重大的技术突破,其在电力系统中的应用具有重大意义,而失超保护则是对超导电力系统安全运行的保障。在失超保护设计中正确把握温升对超导材料性能的影响非常重要。该文以高温超导带材为样品,通过实验实际测得加热到不同温度时样品恢复后的临界电流,研究了超导材料经历不同温升后临界电流的变化情况。结果显示:对于同一样品,超过一定温度以后,材料将发生不可恢复的变化,超导材料经历一定温升后,其临界电流会下降,温度越高,下降幅值越大,高温的经历对材料的损伤是不可恢复的。