Jungho Im, PhD (ersgis@unist.ac.kr) School of Urban and Environmental Engineering Ulsan National Institute of Science and Technology Ulsan, S. Korea November 7, 2019
Intelligent Remote sensing and geospatial Information Science
Reference data Uniform spatial and High spatio-temporal temporal coverage High temporal resolution resolution Point-based data Difficult to parameterize Cloud contamination High uncertainty Spatial discontinuity (optical sensor data)
Disaster Monitoring and Prediction Multi-source Data fusion
Image source: NVIDIA’s blog
Image by Théophile Gonos Image source: Difference Between
Research Examples 01 Estimation of Fugacity of Carbon Dioxide ( f CO 2 ) 02 Prediction of Monthly Arctic Sea Ice Concentrations 03 Detection of Convective Initiation 04 Overshooting Tops Detection 05 Estimation of Tropical Cyclone Intensity 06 Estimation of Ground Particulate Matter Concentrations 07 Heatwave Monitoring and Prediction
Estimation of Fugacity of Carbon Dioxide ( f CO 2 ) Jang, E., Im, J.*, Park, G., Park, Y. (2017). Estimation of fugacity of carbon dioxide in the East Sea using in situ measurements and Geostationary Ocean Color Imager satellite data. Remote Sensing , 9, 821
Source : NOAA PMEL Carbon Program • Ocean control the climate of Earth by regulating the concentration of Carbon Dioxide • Ocean acidification by increasing carbon dioxide • Destroy the ocean ecosystem • Monitoring carbon dioxide is important to determine ocean acidification and climate change
• Dynamic phenomena occur • Active deep convection occurs because • The East Sea of Korea of the deep water formation and weak • 20,903 in situ samples vertical stability
Machine Learning Estimate Random Ocean 𝐠𝑫𝑷 𝟑 Forest GOCI GOCI, MODIS Based on Chlorophyll Chlorophyll CDOM CDOM Satellite Data Band Reflectance Band Reflectance Support HYCOM HYCOM Vector Mixed layer depth Mixed layer depth Regression Sea surface temperature Sea surface temperature Sea surface Salinity Sea surface salinity Calculate Sea-Air 𝑫𝑷 𝟑 Flux In situ Fugacity of CO 2
2015 (RF model) • Seasonal variability of surface seawater f CO 2 averaged by month in 2015 (RF model) • Similar spatial patterns with SST, SSS, and MLD • f CO 2 in summer might be affected by an inflow of warm current from South • Low f CO 2 in coastal areas appears to be related to biological activity f CO 2 (µatm) 300 350 400
• Sea-air CO 2 flux using estimated surface seawater f CO 2 based on the RF model • The East Sea absorbs CO 2 from the atmosphere throughout the whole region, acts as a sink for atmospheric CO 2 • The annual mean CO 2 flux value was -1.53 mol · m -2 · year -1 • The largest CO 2 flux to the ocean was estimated in winter and the Flux ( mol m -2 yr -1 ) lowest flux in summer -5 -2 0 1
Prediction of monthly Arctic sea ice concentrations Kim, Y., Kim, H., Han, D., Im, J.* , Lee, S. (2019). Prediction of monthly Arctic sea ice concentrations using satellite and reanalysis data based on convolutional neural network deep learning. The Cryosphere , in revision
• Research objectives ✓ Development of a novel monthly SIC prediction model using a deep learning approach (CNN) ✓ Examination of the prediction performance by comparing Random Forests model ✓ Analysis of the sensitivity of variables affecting SICs
• Variable description ✓ Eight predictors were used to predict the monthly SIC ✓ i.e., the SICs in September is affected by the SICs in August Temporal Spatial Normaliza Variable Source Unit resolution resolution tion SIC one-year before (sic_1y) 0 - 1 SIC one-month before (sic_1m) 0 - 1 NSIDC, The Nimbus 7 SIC anomaly one-year before -1 - 1 SMMR and the DMSP % Daily 25km (ano_1y) SSM/I and SSMIS SIC anomaly one-month before -1 - 1 (ano_1m) Sea surface temperature one- NOAA OISST ver.2 K Daily 0.25° 0 - 1 month before (sst) 2-meter air temperature one- K 0 - 1 month before (t2m) forecast albedo one-month ECMWF ERA Interim % Monthly 0.125° 0 - 1 before (fal) the amount of v-wind one-month m/s 0 - 1 before (v-wind)
MAE RMSE nRMSE NSE Simple prediction 9.36% 21.93% 61.94% 0.83 All range of SICs RF 2.45% 6.61% 18.64% 0.96 (0-100%) CNN 2.28% 5.76% 16.15% 0.97 Simple prediction 3.88% 11.96% 33.22% 0.60 Low SICs RF 2.38% 7.23% 19.87% 0.90 (0-40%) CNN 2.13% 6.18% 16.87% 0.93 (a) mean absolute (b) MAE of simple (c) MAE of RF (RF- (d) MAE of CNN SIC anomaly (%) prediction model NSIDC, %) (CNN-NSIDC, %) (predicted-NSIDC%)
(a) SIC (NSIDC, %) (b) SIC (RF, %) (c) SIC (CNN, %) (e) Error (RF-NSIDC, (f) Error (CNN-NSIDC, (d) SIC anomaly (NSIDC, %) RMSE = 7.47%; RMSE = 5.00%; nRMSE = 32.71%) nRMSE = 21.93%)
Detection of Convective Initiation Han, H., Lee, S., Im, J.* , Kim, M., Lee, M., Ahn, M., Chung, S. (2015). Detection of convective initiation using Meteorological Imager on board Communication, Ocean and Meteorological Satellite based on machine learning approaches. Remote Sensing , 7, 9184-9204 . Lee, S., Han, H., Im, J. *, Jang, E. (2017). Detection of deterministic and probabilistic convective initiation using Himawari-8 Advanced Himawari Imager data. Atmospheric Measurement Techniques , 10, 1859 – 1874 Han, D., Lee, J., Im, J.*, Sim, S., Lee, S., Han, H. (2019). A novel framework of detecting convective initiation combining automated sampling, machine learning, and repeated model tuning using geostationary satellite data. Remote Sensing , 11, 1454.
Heavy Rainfall & Severe Storm Numerical weather prediction (NWP) Weather radar ▪ High uncertainty for nowcasting ▪ ▪ Low spatial resolution Limited spatial coverage
• Data and methods GOES-R Interest fields Himawari-8 AHI Machine learning approaches Machine learning approach & logistic regression Interest fields optimized for identifying CIs over Korean Peninsula Developing optimized interest field of CI over Korean peninsula
• An example - CI prediction using RF : 20150801 0610-0800 (for 120 mins) - Radar CAPPI reference: 20150801 0810 UTC CI
Overshooting Tops Detection Kim, M., Im, J.* , Park, H., Park, S., Lee, M., Ahn, M. (2017). Detection of tropical overshooting cloud tops using Himawari-8 imagery. Remote Sensing , 9, 685. Kim, M., Lee, J., Im, J.* (2018). Deep learning-based monitoring of overshooting cloud tops from geostationary satellite data. GIScience and Remote Sensing , 55, 763-792.
• Overshooting cloud tops (OT) • “A domelike protrusion above a cumulonimbus anvil, representing the intrusion of an updraft through its equilibrium level” (AMS) • Cumulonimbus clouds with OTs can cause severe weather conditions, influencing in-flight and ground aviation operations. • We try to develop a system to mimic the “human being’s system” which is more intuitive and generalizable. → “Convolutional Neural Network (CNN)” . • Rather than identifying individual pixels as OTs or non-OTs, human beings typically find OTs through visual inspection of contextual information of pixels (i.e. dome-like shape).
• Overshooting cloud tops viewed by Himawari-8 Time series infrared images of 13th May, 2015 Infrared Visible [Source: Kristopher M. Bedka at NASA Langley Research Center]
• Final CNN structure in this study
• POD and FAR for whole images • POD (Probability Of Date POD FAR Detection) 0600 UTC on Sep. 25th 2015 90.59% 13.27% = # of OTs detected correctly /Total # of OT reference 0600 UTC on May 25th 2016 85.19% 8.89% • FAR (False Alarm Rate) 0600 UTC July 8th 2016 86.93% 31.77% = # of misclassified OTs Average 87.57% 17.98% /Total # of detected OTs • Comparisons of model performances Authors Technique POD FAR Bedka et al. (2010)+ IRW-texture candidates+ Bedka and 35.1% 24.9% detection criteria Khlopenkov (2016) OT probability 69.2% 18.4% Bedka and OT probability with visible Khlopenkov (2016) 51.4% 1.6% rating detection Kim et al. (2017) Random Forest 77.76% 31.73% This study CNN 79.68% 9.78%
CNN RF CNN RF
• Product: Overshooting tops (OT) • Spatial coverage: Northeast Asia • Product period : 10min < 가시영상과 산출 결과 > < 지상국 산출 영상 > 32
Estimation of Tropical Cyclone Intensity Lee, J., Im, J.*, Cha, D.*, Park, H., Sim, S. (2019). Tropical cyclone intensity estimation through multidimensional convolutional neural networks using geostationary satellite data, in review.
Estimation of Ground Particulate Matter Concentrations Park, S., Shin, M., Im, J.* , Song, C., Choi, M., Kim, J., Lee, S., Park, R., Kim, J., Lee, D., Kim, S. (2019). Estimation of ground level particulate matter concentrations through the synergistic use of satellite observations and process-based models over South Korea, Atmospheric Chemistry and Physics , 19, 1097-1113. Park, S., Lee, J., Im, J.* , Song, C., Kim, J., Lee, S., Park, R., Kim, S., Yoon, J., Lee, D., Quackenbush, L. (2019). Estimation of spatially continuous particulate matter concentrations under all sky conditions through the synergistic use of satellite-based AOD and numerical models, in review.
Recommend
More recommend
