基于广义高斯模型的机载测深LiDAR波形拟合算法

针对传统LiDAR测深波形拟合算法受噪声干扰严重、对微弱回波信号及复杂波形拟合不准确的问题,提出一种基于广义高斯模型的波形拟合算法。首先通过计算滤波前后尾段波形的差异估计波形的噪声;然后利用广义高斯模型提取海面与海底反射信号分量,并对剩余信号进行迭代拟合;最后利用LM算法对参数进行优化,并对优化后的参数进行约束,避免冗余分量。采用南海实测数据进行验证,该算法拟合微弱回波及复杂波形能力强,不论在浅水(回波发生叠加)还是深水,其拟合精度均优于传统算法,具有良好的鲁棒性。     

An elemental abundance analysis of the magnetic chemically peculiar star HR 8216

An elemental abundance analysis of the cool magnetic CP star HR 8216 (= HD204411) was performed using 2.4 Å mm⁻¹ Dominion Astrophysical Observatory Reticon exposures covering  λλ3830–4770  with a typical signal-to-noise ratio of 200 taken with the long camera of the 1.22-m telescope. The spectrograms were measured interactively with the graphics computer program reduce . The fine analysis used an ATLAS9 metal-rich model atmosphere, the predictions of which best-matched the optical region fluxes and the Hγ profile. The anomalies of HR 8216 are primarily an enhancement of many iron peak elements with the rare earths elements much less enhanced than in many similar cool magnetic CP stars. Using the results of the fine analysis the spectrum was synthesized. Comparison with the observed spectrum showed that the agreement is good but did not fully account for all of the observed line spectrum.     

Wind-wave transformations in an elongated bay

In order to determine wave transformations in an elongated bay, a numerical solution was used to interpret yearlong records of bottom pressure and wind velocity obtained at the mouth and head of Concepción Bay, on the Gulf of California side of the Baja California peninsula. Observed wind waves were predominantly produced by southeastward winds in the winter and north–northwestward winds in the summer. Typical mean wave periods at the bay entrance were between 3 and 5 s. In contrast, the waves at the head of the bay had predominant periods <3 s. The energetic long-period swell waves were dissipated somewhere in the bay as they were not observed at the head of the bay. This study centered in identifying the effects that caused swell waves to attenuate in the bay. The ‘Simulating WAves Nearshore (SWAN)’ model was used to determine the cause for such wave attenuation. Model results showed that swell waves were attenuated because of the combined effects of bottom friction, wave breaking, whitecapping, refraction and wave blocking by the coastline. Most of the attenuation (close to 90%), however, was caused by wave blocking owing to the change of coastline orientation of the bay. This wave blocking mechanism should therefore be explored further in embayments of complex coastline morphology.     

The monitoring of red tides based on modular neural networks using airborne hyperspectral remote sensing

This paper proposes a red tide monitoring method based on clustering and modular neural networks. To obtain the features of red tide from a mass of aerial remote sensing hyperspectral data, first the Log Residual Correction （LRC） is used to normalize the data, and then clustering analysis is adopted to select and form the training samples for the neural networks. For rapid monitoring, the discriminator is composed of modular neural networks, whose structure and learning parameters are determined by an Adaptive Genetic Algorithm （AGA）. The experiments showed that this method can monitor red tide rapidly and effectively.