共查询到20条相似文献，搜索用时 538 毫秒
1.
利用地基微波辐射计资料反演0-10km大气温湿廓线试验研究 总被引：3，自引：0，他引：3
实测与模拟的微波辐射计亮温存在偏差,导致基于BP神经网络模型的大气温湿廓线反演精度的降低。研究了一种基于资料订正后的BP神经网络反演大气温湿廓线的方法。首先,基于2014年6月南京江宁探空资料,利用MonoRTM模式,模拟中心频率在22.2GHz~58.8GHz范围内22通道亮温;对比模拟和实测南京站微波辐射计资料,建立实测微波辐射计资料订正模型。然后,以南京地区2011-2013年探空资料为输入,模拟22通道亮温数据,并基于模拟的22通道亮温数据和当地探空资料,利用BP神经网络算法,建立大气温度、水汽密度、相对湿度廓线反演模型。最后,利用构建的订正模型,对2014年7月试验获取的微波辐射计资料进行订正,并将订正后的微波辐射计资料输入BP神经网络反演模型,反演0-10km高度58层的大气温度、水汽密度和相对湿度,对比实际探空资料以及微波辐射计二级产品,评估分析反演效果。实验结果表明：所建的反演模型提高了大气温湿廓线反演精度,大气温度、水汽密度和相对湿度均方根误差范围分别为1.0~2.0K、0.20 ~1.93g/m3和2.5%~18.6%。 相似文献
2.
以2007~2018年西宁二十里铺气象站探空资料为模拟样本,利用MonoRTM模式模拟中心频率21.985~58.759GHz的35通道亮温,应用BP神经网络对模拟数据进行反复训练,构建最优反演模型,并以2019年探空资料为测试样本,对比分析了不同季节和不同天气条件下BP神经网络与微波辐射计的反演效果。结果表明:晴空条件下,BP神经网络与微波辐射计在温度反演上效果最佳,水汽密度次之,相对湿度最差,其中冬春季BP神经网络反演效果优于微波辐射计,夏秋季反之;有云条件下,BP神经网络温度反演效果在冬、春和夏季均优于微波辐射计,其水汽密度反演效果在四季均较微波辐射计有明显提升,其相对湿度反演效果在冬、春和夏季均较微波辐射计更佳。晴空和有云条件下,BP神经网络在不同季节反演温度、水汽密度和相对湿度的平均绝对误差和标准偏差均小于微波辐射计,尤其是相对湿度的反演精度提升最为明显。晴空条件下,BP神经网络反演温度廓线在春、夏和秋季效果最佳,反演水汽密度廓线在中低层精度较高,反演相对湿度廓线的精度较差,但基本和探空资料趋势一致;有云条件下,BP神经网络反演温度廓线与晴空时基本一致,较微波辐射计精度更高,反演水汽密度和相对湿度廓线在8km以上效果较好。 相似文献
3.
基于一维变分算法的地基微波辐射计遥感大气温湿廓线研究 总被引：3，自引：1，他引：2
为研究地基微波辐射计遥感温、湿度廓线的一维变分算法的反演能力,用北京地区2010—2011年00和12时(世界时)的多通道地基微波辐射计亮温资料进行试验。首先,利用同时次的地面观测资料、红外亮温(由地基微波辐射计自带红外传感器测得)及探空观测数据,给出提取无云样本的方案,得到432个无云样本;再以辐射传输模式计算得到的模拟亮温为参考,对无云条件下的观测亮温进行质量控制;然后利用探空数据进行模拟试验,结果发现,一维变分算法对3 km以下的温度廓线有较大调整。使反演结果更加接近探空,而对湿度廓线在0—10 km都有不同程度的优化;最后利用一维变分算法对地基微波辐射计观测亮温进行大气温湿廓线反演,将结果与探空对比可以看出,温度廓线的均方根误差小于2.9 K,绝对湿度的均方根误差小于0.47 g/m~3;进一步与地基微波辐射计自带神经网络的反演结果比较表明,一维变分的反演结果更接近实际大气。 相似文献
4.
为了研究ERA5再分析资料在微波辐射计数据反演中的应用效果,利用探空和ERA5作为训练样本分别训练得到两个神经网络,对微波辐射计观测数据进行反演。并以探空资料作为参考分别统计基于探空和ERA5再分析资料反演得到的数据的相关系数、绝对偏差和均方根误差,用于评估ERA5再分析资料在微波辐射计数据反演中的应用效果。结果表明：基于ERA5和探空反演的温度廓线与探空的相关系数分别为0.988、0.99,绝对偏差分别为2.402 ℃、2.607 ℃。反演水汽与探空的相关系数分别为0.975、0.979,绝对偏差分别为0.412 g/m3、0.369 g/m3。反演的相对湿度相关系数分别为0.663、0.696,绝对偏差分别为15.587%、13.976%。反演的积分水汽总量相关系数分别为0.959、0.968,绝对偏差分别为0.258 cm、0.227 cm。从结果来看,基于ERA5资料反演的温度、水汽资料与利用探空反演的结果相差较小,温度廓线反演效果优于相对湿度廓线。 相似文献
5.
总结了自主研制的MWP967KV型地基35通道微波辐射计系统设计原理和主要特点。系统将K、V双频段宽带天线及接收机、宽带调谐本振、温湿压计以及电源模块紧凑集成为一台整机设备。对各传感器的测量输出实时进行一体化综合处理,采用BP神经网络实时反演大气温度、湿度廓线和汽、水总量。廓线的垂直覆盖范围为地表至顶空10km,共划分为58层。辐射计样机于2012年秋冬季节在北京地区开展了为期2个月的观测试验,利用这段时期内的69个探空资料样本对辐射计2个层次的实时观测输出都进行了对比检验,计算了水汽、氧气通道亮温和反演所得大气温湿廓线的平均偏差、均方差以及相关性。结果表明该系统能够满足实时气象监测的需求,达到国际先进水平。 相似文献
6.
基于暴雨中尺度数值预报模式AREMv2.3和三维变分同化系统GRAPES-3DVAR,以2008年7月22日08时至23日08时发生在我国东部地区的一次特大暴雨过程为例,研究了荆州、咸宁、常德3部地基微波辐射计反演的相对湿度廓线资料同化对降水预报的影响。结果表明:1)同化荆州、咸宁、常德3部微波辐射计反演的相对湿度廓线资料后,模式对降水落区预报改进不明显,但对降水强度预报改进明显,24 h降水最大增幅为45 mm;2)分别同化荆州、咸宁、常德微波辐射计湿度廓线资料,降水强度模拟均有所改进,24 h降水最大增幅分别达到42、10、24 mm;3)单站及两站微波辐射计资料同化均对降水强度预报有所改善,但不及对3部微波辐射计资料同时同化的结果,这表明同化的中尺度水汽场信息越多,初始场的质量越高,降水模拟效果越好;4)对上述暴雨过程而言,荆州微波辐射计湿度廓线资料同化对降水强度预报改进贡献最大,常德站次之,咸宁站作用最不明显。 相似文献
7.
8.
为提高地基微波辐射计大气探测精度,融合BP神经网络与遗传算法,研究0~10 km大气温湿度廓线。首先,结合数据特征,基于数值模拟技术,建立一套TP/WVP-3000型号地基微波辐射计的一级数据质量控制和订正模型。然后,为减小训练样本代表性误差对模型反演精度的影响,利用遗传算法优化训练样本数据,建立一套精度更高的神经网络大气温湿度反演模型。最后,利用构建的反演模型,开展大气温湿度反演试验,结合探空资料和微波辐射计二级产品,评价反演模型精度。研究结果表明:(1)经过质量控制后的实测数据与模拟数据之间的相关性有显著提升;(2)经过质量控制与订正后建立的神经网络模型对比原微波辐射计二级产品的反演精度有一定提升,温度提升6.77%,湿度提升20.11%;(3)经过遗传算法优化后的训练样本所建立的神经网络反演模型对比原微波辐射计二级产品反演精度有进一步的提升,温度提升10.21%,湿度提升23.75%,反演结果与该地区同类型研究结果相比有着较大提升。 相似文献
9.
地基微波辐射计反演水汽密度廓线精度分析 总被引：2，自引：0，他引：2
以探空水汽廓线为标准,对00:00UTC、12:00UTC北京南郊(39.93°E,116.28°N)观测站的微波辐射计水汽廓线与WRF(Weather ResearchForecasting Model)模拟的水汽廓线进行精度分析。结果表明,整体上WRF水汽密度接近探空,但在近地面处微波辐射计的更接近探空;水汽密度偏差在夏季较大冬季较小;微波辐射计的水汽密度在3km左右偏大,根据北京地区历史天气记录中有中云的天气较多的信息,推测可能是由于云对微波辐射计反演有影响。06:00UTC和18:00UTC以WRF水汽廓线为参考对微波辐射计的水汽廓线进行精度分析,得到的结果与00:00UTC和12:00UTC的类似。 相似文献
10.
为提升地基微波辐射计在不同天气条件下, 特别是云天条件下温湿廓线的反演精度, 利用2011年1月—2016年12月中国气象局北京国家综合气象观测试验基地探空数据, 在微波辐射计反演温湿度廓线的过程中通过区分晴天和云天条件并引入全固态Ka波段测云仪云高及云厚信息, 对反演输入亮温进行质量控制和偏差订正, 建立BP神经网络模型, 采用2017年1月—2018年3月微波辐射计探测数据评估检验, 结果表明:在亮温订正前提下, 晴天温度模型、云天温度模型、晴天相对湿度模型和云天相对湿度模型反演结果与探空的相关系数分别为0.99, 0.99, 0.80和0.78, 均方根误差为2.3℃, 2.3℃, 9%和16%, 较微波辐射计自带产品（LV2产品）减小约0.4℃, 0.3℃, 11%和9%, 准确性提升约30%, 28%, 64%和45%;温度模型偏差在±2℃以内、湿度模型偏差在±20%以内的占比分别为68%, 70%和95%, 78%, 较LV2产品分别提高了7%, 5%和27%, 23%, 其中相对湿度改善明显。可见亮温订正、区分天气类型训练反演模型有利于改善地基微波辐射温湿廓线反演精度。 相似文献
11.
Stephen F. Wathen 《Climatic change》2011,108(1-2):333-356
This paper provides both a detailed history of environmental change in the Sierra Nevada over the past 1,800 years and evidence for climate teleconnections between the Sierra Nevada and Greenland during the late Holocene. A review of Greenland ice core data suggests that the magnitudes of abrupt changes in temperature and precipitation increased beginning c. 3,700 and 3,000 years ago, respectively. Precipitation increased abruptly 1,300 years ago. Comparing paleotemperature data from Cirque Peak, CA with paleoprecipitation data from Pyramid Lake, NV suggests that hot temperatures occurred at the beginnings of most severe droughts in the Sierra Nevada over the past 1,800 years. Severe fires and erosion also occurred at Coburn Lake, CA at the beginning of all severe droughts in the Sierra Nevada over the past 1,800 years. This suggests that abrupt climate change during the late Holocene caused vegetation and mountain slopes in some areas to be out of equilibrium with abruptly changed climates. Finally, the ending of drought conditions in Greenland coincided with the beginning of drought conditions in the Sierra Nevada over the past 1,800 years, perhaps as a result of the rapidly changed locations of the Earth??s major precipitation belts during abrupt climate change events. 相似文献
12.
B. Kochtubajda 《Atmospheric Research》1995,35(2-4)
Physical experiments designed to explore the potential of rain augmentation through airborne glaciogenic seeding on small, isolated non-precipitating cumuliform clouds near Red Deer, Alberta were carried out during the period 1982–1985. The microstructure of 90 cumulus congestus clouds have been documented through repeated in-situ sampling using a cloud physics instrumented aircraft platform. Observations from the inspection passes of 57 clouds seeded with either dry ice pellets or silver iodide pyrotechnics, and all the passes of 33 natural clouds are presented.Measurements of the cloud droplet concentration indicate that Alberta cumulus clouds are typically continental in nature, with an average droplet concentration of 535 cm−3 and an average droplet diameter of 10.6 μm. Alberta clouds have average liquid water contents of 0.57 g m−3, with a peak 1-sec value of 3.17 g m−3. The 1-km average liquid water contents are 0.83 g m−3, with a peak value of 2.81 g m−3. Cloud lifetimes vary between 11 and 20 minutes. Concentrations of naturally occurring ice crystals are found to be low. The average maximum 1-km ice concentration was 31−1, and the peak 1-km concentration was 73.11−1 in the natural cloud dataset. Evidence of precipitation-sized particles was detected in 21% (7 of 33) of the clouds, and precipitation below cloud base was detected in 6% (2 of 33) of the clouds.A comparison of the Alberta cloud characteristics to the cumulus clouds from different locations showed that there are some distinct differences between Alberta clouds and the clouds from the other regions. 相似文献
13.
14.
The present study aims at the assessment of six satellite rainfall estimates (SREs) in Pakistan. For each assessed products, both real-time (RT) and post adjusted (Adj) versions are considered to highlight their potential benefits in the rainfall estimation at annual, monthly, and daily temporal scales. Three geomorphological climatic zones, i.e., plain, mountainous, and glacial are taken under considerations for the determination of relative potentials of these SREs over Pakistan at global and regional scales. All SREs, in general, have well captured the annual north-south rainfall decreasing patterns and rainfall amounts over the typical arid regions of the country. Regarding the zonal approach, the performance of all SREs has remained good over mountainous region comparative to arid regions. This poor performance in accurate rainfall estimation of all the six SREs over arid regions has made their use questionable in these regions. Over glacier region, all SREs have highly overestimated the rainfall. One possible cause of this overestimation may be due to the low surface temperature and radiation absorption over snow and ice cover, resulting in their misidentification with rainy clouds as daily false alarm ratio has increased from mountainous to glacial regions. Among RT products, CMORPH-RT is the most biased product. The Bias was almost removed on CMORPH-Adj thanks to the gauge adjustment. On a general way, all Adj versions outperformed their respective RT versions at all considered temporal scales and have confirmed the positive effects of gauge adjustment. CMORPH-Adj and TMPA-Adj have shown the best agreement with in situ data in terms of Bias, RMSE, and CC over the entire study area. 相似文献
15.
广州地Ⅸ的高温天气主要是受副热带高压和台风外围下沉气流的影响所致.文中采用BDA(Bogus Data Assimila-tion)方法,探讨BDA方案对广州地区台风背景条件下高温预报的改进能力.选取2005年7月中旬广州地区出现的高温天气进行研究.这是比较典型的受副热带高压和台风(海棠)共同影响造成高温的天气过程.分析有无采用BDA方案的模式初始场.结果表明:采用BDA方案同化Bogus模型可以调整台风中心位置和强度,使所得到的初始场中心位置与观测更为接近,台风强度(气压梯度力、风速)比末用Bogus的情况强,与观测值更为接近.数值模拟的结果表明,采用了BDA方案的敏感试验可以更好地预撤台风路径和台风中心强度变化,从而更好地预报高温天气,对高温区分布、日平均温度大小等的预报都有改进.文中对引起这种预报差异的原因进行了讨论,并探讨高温预报改进的可能机制.大气下沉运动的增强是高温预报改进的主要原因.敏感试验由于广州中低层大气的水汽减少,大气的下沉增强,致使天空的云量减少,对太阳短波辐射的阻挡减小,从而地面吸收热量增多,温度升高,输送给大气的感热增加,大气气温升高.采用BDA方案可以改进模式在台风"海棠"过程对广州高温的预报. 相似文献
16.
17.
A. C. Chamberlain 《Boundary-Layer Meteorology》1983,25(4):405-409
Profiles of wind speed in driving sand and snow show roughness length increasing as the square of the friction velocity, in accordance with Charnock's relation for the roughness length of the sea. 相似文献
18.
Distribution,speciation, and budget of atmospheric mercury 总被引：9，自引：0，他引：9
Total gaseous mercury (TGM) concentrations over the Atlantic Ocean and over Central Europe were measured repeatedly in the years 1978–1981. The latitudinal TGM distribution showed a pronounced and reproducible interhemispherical difference with higher TGM concentrations in the Northern Hemisphere. TGM was found to be vertically well mixed within the troposphere. The TGM concentration seems to increase with time at a rate of 10±8%/yr in the Northern and 8±3%/yr in the Southern Hemisphere. Measurements of mercury speciation showed that elemental mercury is the main TGM component contributing more than 92% and 83% of TGM in marine and continental air, respectively. The tropospheric mercury burden was calculated to be 6×109g. The interhemispheric distribution and temporal and spatial variability of TGM imply a tropospheric residence time of TGM of about 1 yr. Sink strengths calculated independently from the measured mercury concentration on particles and in rainwater are consistent with the above figures. 相似文献
19.
Data collected in 2007 from a dense commercial network (operated by AWS Convergence Technologies, Inc.) of roof-mounted temperature
sensors are used to explore the heat island characteristics of Washington, DC, and New York City, NY. Considerable spatial
detail is revealed, but aggregating data in annuli centered on assumed central locations in the business districts of the
two cities reveals that the heat islands extend out to more than 30 km, with the New York City island being somewhat larger.
The results from both arrays reveal the influence of the surroundings, with large scatter of daytime results being characteristic
of sites with the greatest local surface inhomogeneity. Nighttime data are more ordered, and suggest that surface air temperatures
decrease by about 0.02°C km−1 for the Washington case, and 0.04°C km−1 for New York, with the winter behaviour being more pronounced than for other seasons. Scatter of the data in the daytime
is a common feature for all seasons, but mainly for those with the strongest insolation. Comparison between working day and
weekend temperatures provides convincing verification that the air responds quite slowly to changes in surface (radiometric)
temperatures, with distance constants of the order of many tens of km. There appears to be a small wind speed effect, which
is evident in the nighttime data but is largely obscured by scatter for the daytime. 相似文献
20.
Preliminary Comparison of the CMA, ECMWF, NCEP, and JMA Ensemble Prediction Systems 总被引：1，自引：0，他引：1 
下载免费PDF全文
下载免费PDF全文 Based on The Observing System Research and Predictability Experiment (THORPEX) Interactive Grand Global Ensemble (TIGGE) dataset,using various verification methods,the performances of four typical ense... 相似文献