scholarly journals Applying machine learning methods to detect convection using Geostationary Operational Environmental Satellite-16 (GOES-16) advanced baseline imager (ABI) data

2021 ◽  
Vol 14 (4) ◽  
pp. 2699-2716
Author(s):  
Yoonjin Lee ◽  
Christian D. Kummerow ◽  
Imme Ebert-Uphoff
Keyword(s):  
Temporal Resolution ◽  
Radar Data ◽  
Probability Of Detection ◽  
High Temporal Resolution ◽  
Convective System ◽  
Mountainous Regions ◽  
Convective Clouds ◽  
Geostationary Operational Environmental Satellite ◽  
Resolution Data ◽  
Infrared Data

Abstract. An ability to accurately detect convective regions is essential for initializing models for short-term precipitation forecasts. Radar data are commonly used to detect convection, but radars that provide high-temporal-resolution data are mostly available over land, and the quality of the data tends to degrade over mountainous regions. On the other hand, geostationary satellite data are available nearly anywhere and in near-real time. Current operational geostationary satellites, the Geostationary Operational Environmental Satellite-16 (GOES-16) and Satellite-17, provide high-spatial- and high-temporal-resolution data but only of cloud top properties; 1 min data, however, allow us to observe convection from visible and infrared data even without vertical information of the convective system. Existing detection algorithms using visible and infrared data look for static features of convective clouds such as overshooting top or lumpy cloud top surface or cloud growth that occurs over periods of 30 min to an hour. This study represents a proof of concept that artificial intelligence (AI) is able, when given high-spatial- and high-temporal-resolution data from GOES-16, to learn physical properties of convective clouds and automate the detection process. A neural network model with convolutional layers is proposed to identify convection from the high-temporal resolution GOES-16 data. The model takes five temporal images from channel 2 (0.65 µm) and 14 (11.2 µm) as inputs and produces a map of convective regions. In order to provide products comparable to the radar products, it is trained against Multi-Radar Multi-Sensor (MRMS), which is a radar-based product that uses a rather sophisticated method to classify precipitation types. Two channels from GOES-16, each related to cloud optical depth (channel 2) and cloud top height (channel 14), are expected to best represent features of convective clouds: high reflectance, lumpy cloud top surface, and low cloud top temperature. The model has correctly learned those features of convective clouds and resulted in a reasonably low false alarm ratio (FAR) and high probability of detection (POD). However, FAR and POD can vary depending on the threshold, and a proper threshold needs to be chosen based on the purpose.

scholarly journals Applying machine learning methods to detect convection using GOES-16 ABI data

2020 ◽  
Author(s):  
Yoonjin Lee ◽  
Christian D. Kummerow ◽  
Imme Ebert-Uphoff
Keyword(s):  
Temporal Resolution ◽  
Radar Data ◽  
Probability Of Detection ◽  
High Temporal Resolution ◽  
Convective System ◽  
Cloud Optical Depth ◽  
Mountainous Regions ◽  
Convective Clouds ◽  
Resolution Data ◽  
Infrared Data

Abstract. An ability to accurately detect convective regions is essential for initializing models for short term precipitation forecasts. Radar data are commonly used to detect convection, but radars that provide high temporal resolution data are mostly available over land and the quality of the data tends to degrade over mountainous regions. On the other hand, geostationary satellite data are available nearly anywhere and in near-real time. Current operational geostationary satellites, the Geostationary Operational Environmental Satellite-16 (GOES-16) and -17 provide high spatial and temporal resolution data, but only of cloud top properties. One-minute data, however, allow us to observe convection from visible and infrared data even without vertical information of the convective system. Existing detection algorithms using visible and infrared data look for static features of convective clouds such as overshooting top or lumpy cloud top surface, or cloud growth that occurs over periods of 30 minutes to an hour. This study represents a proof-of-concept that Artificial Intelligence (AI) is able, when given high spatial and temporal resolution data from GOES-16, to learn physical properties of convective clouds and automate the detection process. A neural network model with convolutional layers is proposed to identify convection from the high-temporal resolution GOES-16 data. The model takes five temporal images from channel 2 (0.65 μm) and 14 (11.2 μm) as inputs and produces a map of convective regions. In order to provide products comparable to the radar products, it is trained against Multi-Radar Multi-Sensor (MRMS), which is a radar-based product that uses rather sophisticated method to classify precipitation types. Two channels from GOES-16, each related to cloud optical depth (channel 2) and cloud top height (channel 14), are expected to best represent features of convective clouds: high reflectance, lumpy cloud top surface, and low cloud top temperature. The model has correctly learned those features of convective clouds, and resulted reasonably low false alarm ratio (FAR) and high probability of detection (POD). However, FAR and POD can vary depending on the threshold, and a proper threshold needs to be chosen based on the purpose.

scholarly journals Eruptive Styles Recognition Using High Temporal Resolution Geostationary Infrared Satellite Data

2019 ◽  
Vol 11 (6) ◽  
pp. 669 ◽  
Author(s):  
Valerio Lombardo ◽  
Stefano Corradini ◽  
Massimo Musacchio ◽  
Malvina Silvestri ◽  
Jacopo Taddeucci
Keyword(s):  
Temporal Resolution ◽  
Hot Spot ◽  
High Temporal Resolution ◽  
Spot Detection ◽  
Infrared Imager ◽  
Mt Etna ◽  
Flow Through ◽  
Radiance Distribution ◽  
Strombolian Explosions ◽  
Infrared Data

The high temporal resolution of the Spinning Enhanced Visible and InfraRed Imager (SEVIRI) instrument aboard Meteosat Second Generation (MSG) provides the opportunity to investigate eruptive processes and discriminate different styles of volcanic activity. To this goal, a new detection method based on the wavelet transform of SEVIRI infrared data is proposed. A statistical analysis is performed on wavelet smoothed data derived from SEVIRI Mid-Infrared( MIR) radiances collected from 2011 to 2017 on Mt Etna (Italy) volcano. Time-series analysis of the kurtosis of the radiance distribution allows for reliable hot-spot detection and precise timing of the start and end of eruptive events. Combined kurtosis and gradient trends allow for discrimination of the different activity styles of the volcano, from effusive lava flow, through Strombolian explosions, to paroxysmal fountaining. The same data also allow for the prediction, at the onset of an eruption, of what will be its dominant eruptive style at later stages. The results obtained have been validated against ground-based and literature data.

A Strategy for Scaling ERA5 Sea Winds Using Climate Model Data

2020 ◽  
Author(s):  
Silvio Davison ◽  
Francesco Barbariol ◽  
Alvise Benetazzo ◽  
Luigi Cavaleri ◽  
Paola Mercogliano
Keyword(s):  
Temporal Resolution ◽  
Climate Model ◽  
Grid Point ◽  
High Temporal Resolution ◽  
Widespread Application ◽  
Northern Adriatic ◽  
Wind Speeds ◽  
The Past ◽  
Statistical Strategy ◽  
Resolution Data

<p>Over the past decade, reanalysis data products have found widespread application in many areas of research and have often been used for the assessment of the past and present climate. They produce reliable atmospheric fields at high temporal resolution, albeit at low-to-mid spatial resolution. On the other hand, climatological analyses, quite often down-scaled to represent conditions also in enclosed basins, lack the historical sequence of stormy events and are often provided at poor temporal resolution.</p><p>In this context, we investigated the possibility of using the ERA5 reanalysis 10-m wind (25-km and 1-hour resolution data) to assess the Mediterranean Sea wind climate (past and scenario). We propose a statistical strategy to relate ERA5 wind speeds over the sea to the past and future wind speeds produced by the COSMO-CLM (8-km and 6-hour resolution data) climatological model. In particular, the probability density function of the ERA5 wind speed at each grid point is adjusted to match that of COSMO-CLM. In this way, past ERA5 winds are corrected to account for the COSMO-CLM energy, while ERA5 scaled wind sequence can be projected in the future with COSMO-CLM scenario energy. Comparison with past observations confirms the validity of the adopted method.</p><p>In the Venezia2021 project, we have applied this strategy for the assessment of the changing wind and, after WAVEWATCH III model runs, also the wave climate in the Northern Adriatic Sea, especially in front of Venice and the MOSE barriers, under two IPCC (RCP 4.5 and 8.5) scenarios.</p><p>In general, this strategy may be applied to produce a scaled wind dataset in enclosed basins and improve past wave modeling applications based on any reanalysis wind data.</p>

scholarly journals Prediction of storm transfers and annual loads with data-based mechanistic models using high-frequency data

2017 ◽  
Vol 21 (12) ◽  
pp. 6425-6444 ◽  
Author(s):  
Mary C. Ockenden ◽  
Wlodek Tych ◽  
Keith J. Beven ◽  
Adrian L. Collins ◽  
Robert Evans ◽  
...  
Keyword(s):  
Temporal Resolution ◽  
Dynamic Models ◽  
Second Order ◽  
Dynamic Responses ◽  
High Temporal Resolution ◽  
Agricultural Catchments ◽  
Antecedent Conditions ◽  
Poor Water Quality ◽  
The Uk ◽  
Resolution Data

Abstract. Excess nutrients in surface waters, such as phosphorus (P) from agriculture, result in poor water quality, with adverse effects on ecological health and costs for remediation. However, understanding and prediction of P transfers in catchments have been limited by inadequate data and over-parameterised models with high uncertainty. We show that, with high temporal resolution data, we are able to identify simple dynamic models that capture the P load dynamics in three contrasting agricultural catchments in the UK. For a flashy catchment, a linear, second-order (two pathways) model for discharge gave high simulation efficiencies for short-term storm sequences and was useful in highlighting uncertainties in out-of-bank flows. A model with non-linear rainfall input was appropriate for predicting seasonal or annual cumulative P loads where antecedent conditions affected the catchment response. For second-order models, the time constant for the fast pathway varied between 2 and 15 h for all three catchments and for both discharge and P, confirming that high temporal resolution data are necessary to capture the dynamic responses in small catchments (10–50 km2). The models led to a better understanding of the dominant nutrient transfer modes, which will be helpful in determining phosphorus transfers following changes in precipitation patterns in the future.

scholarly journals Technical Note: Temporal disaggregation of spatial rainfall fields with generative adversarial networks

2021 ◽  
Vol 25 (6) ◽  
pp. 3207-3225
Author(s):  
Sebastian Scher ◽  
Stefanie Peßenteiner
Keyword(s):  
Probability Distribution ◽  
Temporal Resolution ◽  
Radar Data ◽  
Technical Note ◽  
Geographical Information ◽  
Generative Adversarial Networks ◽  
High Temporal Resolution ◽  
Adversarial Networks ◽  
Rainfall Patterns ◽  
Lower Temporal Resolution

Abstract. Creating spatially coherent rainfall patterns with high temporal resolution from data with lower temporal resolution is necessary in many geoscientific applications. From a statistical perspective, this presents a high- dimensional, highly underdetermined problem. Recent advances in machine learning provide methods for learning such probability distributions. We test the usage of generative adversarial networks (GANs) for estimating the full probability distribution of spatial rainfall patterns with high temporal resolution, conditioned on a field of lower temporal resolution. The GAN is trained on rainfall radar data with hourly resolution. Given a new field of daily precipitation sums, it can sample scenarios of spatiotemporal patterns with sub-daily resolution. While the generated patterns do not perfectly reproduce the statistics of observations, they are visually hardly distinguishable from real patterns. Limitations that we found are that providing additional input (such as geographical information) to the GAN surprisingly leads to worse results, showing that it is not trivial to increase the amount of used input information. Additionally, while in principle the GAN should learn the probability distribution in itself, we still needed expert judgment to determine at which point the training should stop, because longer training leads to worse results.

A high temporal resolution data set of ERS scatterometer radar backscatter for research in Arctic and sub-Arctic regions

Polar Record ◽  
2002 ◽  
Vol 38 (205) ◽  
pp. 115-120 ◽  
Author(s):  
Yongwei Sheng ◽  
Laurence C. Smith ◽  
Karen E. Frey ◽  
Douglas E. Alsdorf
Keyword(s):  
Temporal Resolution ◽  
Incident Angle ◽  
High Temporal Resolution ◽  
Grid Spacing ◽  
Radar Backscatter ◽  
Data Set ◽  
Arctic Regions ◽  
Full Resolution ◽  
Data Products ◽  
Resolution Data

AbstractRadar backscatter in Arctic and sub-Arctic regions is temporally dynamic and reflects changes in sea ice, glacier facies, soil thaw state, vegetation cover, and moisture content. Wind scatterometers on the ERS-1 and ERS-2 satellites have amassed a global archive of C-band radar backscatter data since 1991. This paper derives three high temporal resolution data products from this archive that are designed to facilitate scatterometer research in high-latitude environments. Radar backscatter data have a grid spacing of 25 km and are mapped northwards from 60°N latitude over intervals of one, three, and seven days for the period 1991–2000. Data are corrected to a normalized incident angle of 40°. Animations and full-resolution data products are freely available for scientific use at http://merced.gis.ucla.edu/scatterometer/index.htm.

Sign in / Sign up

Export Citation Format

Share Document