Estimating nocturnal opaque ice cloud optical depth from MODIS multispectral infrared radiances using a neural network method

Minnis, Patrick; Hong, Gang; Sun-Mack, S.; Smith, William L.; Chen, Yan; Miller, Steven D.

doi:10.1002/2015jd024456

Cited by 43 publications

(33 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This, together with the low computational costs, makes neural networks an interesting alternative to more commonly used physically based methods. Minnis et al (2016) present a similar approach to estimate the optical thickness of opaque ice clouds at night using an ANN trained with coincident CloudSat/CPR (Cloud Profiling Radar) measurements and Aqua/MODIS (Moderate Resolution Imaging Spectroradiometer) infrared radiances. Holl et al (2014) use combined CALIPSO/CALIOP and CloudSat/CPR retrievals for the retrieval of the IWP from AVHRR (Advanced Very High Resolution Radiometer) and MHS (Microwave Humidity Sounder) on the NOAA and Metop satellites using neural networks.…”

Section: Introductionmentioning

confidence: 99%

Cirrus cloud retrieval with MSG/SEVIRI using artificial neural networks

et al. 2017

View full text Add to dashboard Cite

Abstract. Cirrus clouds play an important role in climate as they tend to warm the Earth-atmosphere system. Nevertheless their physical properties remain one of the largest sources of uncertainty in atmospheric research. To better understand the physical processes of cirrus clouds and their climate impact, enhanced satellite observations are necessary. In this paper we present a new algorithm, CiPS (Cirrus Properties from SEVIRI), that detects cirrus clouds and retrieves the corresponding cloud top height, ice optical thickness and ice water path using the SEVIRI imager aboard the geostationary Meteosat Second Generation satellites. CiPS utilises a set of artificial neural networks trained with SE-VIRI thermal observations, CALIOP backscatter products, the ECMWF surface temperature and auxiliary data.CiPS detects 71 and 95 % of all cirrus clouds with an optical thickness of 0.1 and 1.0, respectively, that are retrieved by CALIOP. Among the cirrus-free pixels, CiPS classifies 96 % correctly. With respect to CALIOP, the cloud top height retrieved by CiPS has a mean absolute percentage error of 10 % or less for cirrus clouds with a top height greater than 8 km. For the ice optical thickness, CiPS has a mean absolute percentage error of 50 % or less for cirrus clouds with an optical thickness between 0.35 and 1.8 and of 100 % or less for cirrus clouds with an optical thickness down to 0.07 with respect to the optical thickness retrieved by CALIOP. The ice water path retrieved by CiPS shows a similar performance, with mean absolute percentage errors of 100 % or less for cirrus clouds with an ice water path down to 1.7 g m −2 . Since the training reference data from CALIOP only include ice water path and optical thickness for comparably thin clouds, CiPS also retrieves an opacity flag, which tells us whether a retrieved cirrus is likely to be too thick for CiPS to accurately derive the ice water path and optical thickness.By retrieving CALIOP-like cirrus properties with the large spatial coverage and high temporal resolution of SEVIRI during both day and night, CiPS is a powerful tool for analysing the temporal evolution of cirrus clouds including their optical and physical properties. To demonstrate this, the life cycle of a thin cirrus cloud is analysed.

show abstract

Section: Introductionmentioning

confidence: 99%

Cirrus cloud retrieval with MSG/SEVIRI using artificial neural networks

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Minnis et al (2016) used an NN recently to estimate the COT of ice clouds from MODIS multispectral infrared radiances. An NN is expected to improve cloud retrieval accuracy in presence of 3-D radiative effects because of the complexity of the problem.…”

Section: Introductionmentioning

confidence: 99%

Feasibility study of multi-pixel retrieval of optical thickness and droplet effective radius of inhomogeneous clouds using deep learning

2017

View full text Add to dashboard Cite

Abstract. Three-dimensional (3-D) radiative-transfer effects are a major source of retrieval errors in satellite-based optical remote sensing of clouds. The challenge is that 3-D effects manifest themselves across multiple satellite pixels, which traditional single-pixel approaches cannot capture. In this study, we present two multi-pixel retrieval approaches based on deep learning, a technique that is becoming increasingly successful for complex problems in engineering and other areas. Specifically, we use deep neural networks (DNNs) to obtain multi-pixel estimates of cloud optical thickness and column-mean cloud droplet effective radius from multispectral, multi-pixel radiances. The first DNN method corrects traditional bispectral retrievals based on the plane-parallel homogeneous cloud assumption using the reflectances at the same two wavelengths. The other DNN method uses socalled convolutional layers and retrieves cloud properties directly from the reflectances at four wavelengths. The DNN methods are trained and tested on cloud fields from largeeddy simulations used as input to a 3-D radiative-transfer model to simulate upward radiances. The second DNN-based retrieval, sidestepping the bispectral retrieval step through convolutional layers, is shown to be more accurate. It reduces 3-D radiative-transfer effects that would otherwise affect the radiance values and estimates cloud properties robustly even for optically thick clouds.

show abstract

“…through machine learning. For cloud remote sensing, artificial neural networks (ANNs) have proven to be a powerful tool for this (Kox et al, 2014;Holl et al, 2014;Minnis et al, 2016;Strandgren et al, 2017). Kox et al (2014) developed an ANN-based algorithm trained with coincident SEVIRI thermal observations and CALIOP products for the cloud top height (CTH) and ice optical thickness (IOT) determination of cirrus clouds from SEVIRI.…”

mentioning

confidence: 99%

“…Holl et al (2014) utilise ANNs trained with coincident CALIOP, CPR, AVHRR and MHS (Microwave Humidity Sounder) retrievals for the detection and IWP determination of ice clouds from AVHRR and MHS observations. Minnis et al (2016) estimate the optical thickness of opaque ice clouds at night using an ANN trained with collocated MODIS IR observations and CPR retrievals. The ultimate goal with these approaches is to retrieve, respectively, CALIOP-, CALIOP/CPR-and CPR-like cloud properties from SEVIRI, AVHRR/MHS and MODIS observations alone.…”

mentioning

confidence: 99%

Characterisation of the artificial neural network CiPS for cirrus cloud remote sensing with MSG/SEVIRI

2017

View full text Add to dashboard Cite

Abstract. Cirrus clouds remain one of the key uncertainties in atmospheric research. To better understand the properties and physical processes of cirrus clouds, accurate largescale observations from satellites are required. Artificial neural networks (ANNs) have proved to be a useful tool for cirrus cloud remote sensing. Since physics is not modelled explicitly in ANNs, a thorough characterisation of the networks is necessary.In this paper the CiPS (Cirrus Properties from SE-VIRI) algorithm is characterised using the space-borne lidar CALIOP. CiPS is composed of a set of ANNs for the cirrus cloud detection, opacity identification and the corresponding cloud top height, ice optical thickness and ice water path retrieval from the imager SEVIRI aboard the geostationary Meteosat Second Generation satellites. First, the retrieval accuracy is characterised with respect to different land surface types. The retrieval works best over water and vegetated surfaces, whereas a surface covered by permanent snow and ice or barren reduces the cirrus detection ability and increases the retrieval errors for the ice optical thickness and ice water path if the cirrus cloud is thin (optical thickness less than approx. 0.3). Second, the retrieval accuracy is characterised with respect to the vertical arrangement of liquid, ice clouds and aerosol layers as derived from CALIOP lidar data. The CiPS retrievals show little interference from liquid water clouds and aerosol layers below an observed cirrus cloud. A liquid water cloud vertically close or adjacent to the cirrus clearly increases the average retrieval errors for the optical thickness and ice water path, respectively, only for thin cirrus clouds with an optical thickness below 0.3 or ice water path below 5.0 g m −2 . For the cloud top height retrieval, only aerosol layers affect the retrieval error, with an increased positive bias when the cirrus is at low altitudes. Third, the CiPS retrieval error is characterised with respect to the properties of the investigated cirrus cloud (ice optical thickness and cloud top height). On average CiPS can retrieve the cirrus cloud top height with a relative error around 8 % and no bias and the ice optical thickness with a relative error around 50 % and bias around ±10 % for the most common combinations of cloud top height and ice optical thickness. Similarities with physically based retrieval methods are evident, which implies that even though the retrieval methods differ in the implementation of physics in the model, the retrievals behave similarly due to physical constraints. Finally, we also show that the ANN retrievals have a low sensitivity to radiometric noise in the SEVIRI observations. For optical thickness and ice water path the relative uncertainty due to noise is less than 10 % down to sub-visual cirrus. For the cloud top height retrieval the uncertainty due to noise is around 100 m for all cloud top heights.

show abstract

Estimating nocturnal opaque ice cloud optical depth from MODIS multispectral infrared radiances using a neural network method

Cited by 43 publications

References 48 publications

Cirrus cloud retrieval with MSG/SEVIRI using artificial neural networks

Cirrus cloud retrieval with MSG/SEVIRI using artificial neural networks

Feasibility study of multi-pixel retrieval of optical thickness and droplet effective radius of inhomogeneous clouds using deep learning

Characterisation of the artificial neural network CiPS for cirrus cloud remote sensing with MSG/SEVIRI

Contact Info

Product

Resources

About