Upload
henry-roman
View
24
Download
0
Tags:
Embed Size (px)
DESCRIPTION
Satellite Land Data Products and Their Assimilation into NCEP Models. X. Zhan, J. Liu, C. Hain , L. Fang, J. Yin NESDIS Center for Satellite Applications & Research, College Park, MD W. Zheng , M. Ek , K. Mo, J. Huang NWS National Centers for Environmental Predictions, College Park, MD - PowerPoint PPT Presentation
Citation preview
X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Satellite Land Data Products and Their Assimilation into NCEP Models
X. Zhan, J. Liu, C. Hain, L. Fang, J. YinNESDIS Center for Satellite Applications & Research, College Park, MD
W. Zheng, M. Ek, K. Mo, J. HuangNWS National Centers for Environmental Predictions, College Park, MD
L. Zhao, H. DingNESDIS Office of Satellite and Product Operations, College Park, MD
2X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NCEP land data needs
NESDIS land satellite data availability
DA efforts at STAR with NCEP models
SMOPS and SM DA for NCEP NWP
GET-D for drought monitoring
Next steps
Outline
3X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NCEP land data needs
NESDIS land satellite data availability
DA efforts at STAR with NCEP models
SMOPS and SM DA for NCEP NWP
GET-D for drought monitoring
Next steps
Outline
NOAA Model Production Suite
Uccellini, 2009
Community NoahLand-Surface Model
Uccellini, 2009
6X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NCEP land data needs
NESDIS land satellite data availability
DA efforts at STAR with NCEP models
SMOPS and SM DA for NCEP NWP
GET-D for drought monitoring
Next steps
Outline
7X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NESDIS Satellite Land Data Availability
Name Satellite/Sensor/System NRT Op
Sfc Type AVHRR, MODIS, VIIRS, GOES/-R Yes
Albedo AVHRR, MODIS, VIIRS, GOES/-R Yes
Fire AVHRR, MODIS, VIIRS, GOES/-R Yes
LST AVHRR, MODIS, VIIRS, GOES/-R Yes
NDVI/GVF AVHRR, MODIS, VIIRS, GOES/-R Yes
Sfc Emissivity MSPPS/MiRS Yes
SM GOES/GOES-R, GCOM-W1, SMOPS Yes
Snow AutoSnow, MSPPS/MiRS Yes
SWE AutoSnow, MSPPS/MiRS Yes
8X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NCEP land data needs
NESDIS land satellite data availability
DA efforts at STAR with NCEP models
SMOPS and SM DA for NCEP NWP
GET-D for drought monitoring
Next steps
Outline
9X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
DA Efforts@STAR with NCEP Models
Name Satellite/Sensor/System Op DA?
Sfc Type AVHRR, MODIS, VIIRS, GOES/-R Yes
Albedo AVHRR, MODIS, VIIRS, GOES/-R ?
Fire AVHRR, MODIS, VIIRS, GOES/-R Testing
LST AVHRR, MODIS, VIIRS, GOES/-R ?
NDVI/GVF AVHRR, MODIS, VIIRS, GOES/-R Tested
Sfc Emissivity MSPPS/MiRS ?
SM GOES/GOES-R, GCOM-W1, SMOPS Tested
Snow AutoSnow, MSPPS/MiRS Yes?
SWE AutoSnow, MSPPS/MiRS Yes?
10X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NCEP land data needs
NESDIS land satellite data availability
DA efforts at STAR with NCEP models
SMOPS and SM DA for NCEP NWP
GET-D for drought monitoring
Next steps
Outline
11X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Two ways to retrieve soil moisture from satellites:
• Microwave (MW): Observed MW brightness temperature depends on soil dielectric constant that is related to soil moisture:
– Strength: higher reliability based on direct physical relationships
– Weakness: antenna technology limits spatial resolution
• Thermal Infrared (TIR): Observed surface temperature changes result from surface energy balance that is dependent on soil moisture:
– Strength: TIR sensor could have higher spatial resolution
– Weakness: relies on land surface energy balance model that is prone to input data errors.
R soil
T cT ac
H s
T s
R aH = H c + H s
R x
H c
T a
5 km
R soil
T cT ac
H s
T s
R aH = H c + H s
R x
H c
T aT a
5 km
Two-S
ourc
e M
od
el (A
LEX
I)
TRAD fc
ABL
TMI
Microwave Sensitivity By Wavelength and Vegetation Density
0.0
1.0
2.0
3.0
4.0
0 5 10 15 20 25Wavelength (cm)
Sen
siti
vity
(D
elta
TB
/ D
elta
Vol
SM
)
BARE
VEGETATION
WATER CONTENT (kg/m2)
1
2
4
0
SSM/I AMSR /WindSat SMOS / SMAP
Satellite Soil Moisture Remote Sensing Science
12X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Soil Moisture Operational Product System (SMOPS)
13X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Increased spatial coverage Multi retrieval variance could
be used as error estimate
Microwave Soil Moisture Products from SMOPS
14X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Microwave Soil Moisture Products from SMOPS
WindSat SMOS ASCAT
Blended
15X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NOAA Global Soil Moisture Data Portal:
15
16X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023 16
17X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
SM Data Assimilation Utility for NCEP GFS
GFSNoah
EnKF
Courtesy of R. Reichle
18X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Assimilation of MW SM into NCEP GFS
GFSNoah
EnKF
Pros: GFS can demonstrate SM impact on forecasts GFS may take advantage of satellite SM obs earlier than
full scale implementation
Cons: Hardwiring limits more flexibility for assimilating other
observational data
19X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Assimilation of MW SM into NCEP GFS
Time: DA at 00z
from April 1 – May 5, 2012
Data: SMOPS Blended Surface SM
Method: EnKF DA within GFS/GSI
Experiments: CTL: Regular GFS run without SM
DAEnKF: Daily EnKF run
20X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Comparison of soil moisture from SMOPS Blended 18Z, 1-30 April 2012
GFS_EnKF EnKF-CTL
GFS_CTL SMOPS
21X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Comparison of soil moisture from SMOPS Blended 18Z, 1-30 April 2012
SMOPS GFS_CTL
GFS_EnKF EnKF-CTL
22X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
GFS Top Lay Soil Moisture ValidationWith USDA-SCAN Measurements
1-30 of April, 2012
East CONUS (26 sites) West CONUS (25 sites) Whole CONUS
RMSE Bias Corr-Coef RMSE Bias Corr-
Coef RMSE Bias Corr-Coef
CTL 0.135 0.046 0.565 0.124 0.033 0.448 0.129 0.040 0.508
EnKF 0.130 -0.031 0.613 0.114 -0.021 0.549 0.123 -0.031 0.587
SMOPS 0.133 -0.055 0.601 0.098 -0.036 0.402 0.117 -0.048 0.524
23X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
+ 0.004
GFS 500hPa Height Anomaly Correlations
24X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
700 hPa
1000 hPa
+ 0.003
+ 0.003
25X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NCEP land data needs
NESDIS land satellite data availability
DA efforts at STAR with NCEP models
SMOPS and SM DA for NCEP NWP
GET-D for drought monitoring
Next steps
Outline
26X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Land surface temperature (LST) and solar insolation (Rs) from NOAA Geostationary Operational Environmental Satellite (GOES) imager and future GOES-R Advance Baseline Imager (ABI) are used in an Atmosphere-Land Exchange Inversion (ALEXI) model to generate ET and an Evaporative Stress Index (ESI) for drought monitoring.
ALEXI model output using GOES data have good agreement with field observations and full-scale land surface model simulations of ET.
ALEXI ET and ESI data products are being used at US operational drought monitoring.
Thermal Infrared Remote Sensing for SM
ALEXI Drought Monitoring Webpage Used by USDA-ARS
• ALEXI Evaporative Stress Index products generated at NESDIS STAR are currently being disseminated to end-users at NDMC and CPC through a website hosted by USDA-ARS (http://hrsl.arsusda.gov/drought/index.php).
27
28
GOES ET and Drought (GET-D) System
STAR
Input Data: LST & solar insolation from GOES imagers, Met forcing from NCEP models via NLDAS
DDS
GET-D
Linux
ESPC Production
GOES-Imager
QC/VAL
Linux
GET-D
Linux
Intranet
Sandia
Internet
gp5
CLASS
Internet
NCEP & Other USERS
ET & Drought Products
LST & Q
NCEP
Met Input
Met Input
ET & Drought Products
LST & Q
28
29X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Palmer Drought Index
Vegetation Health Index LSM SM Output
Current US Drought Monitoring
30X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Palmer Drought Index
Vegetation Health Index
GOES/GOES-R ESI
LSM SM Output
Satellite SM
Satellite ST, Alb, GVF/VI
Enhanced Drought Monitoring
31X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Enhancing US Drought Monitoring with SNPP/JPPP Products
• JPSS products to be used– Surface type (ST) & soil moisture (SM)– Albedo (Al) & Vegetation Index (VI)– Land surface temperature (LST)
• Data assimilation approach– Direct insertion for JPSS ST, LST, Al and VI in NLDAS/GLDAS– EnKF for GCOM-W SM
• Drought-monitoring products– Root-zone SM anomalies– NLDAS/GLDAS runs with and without
assimilations
• Evaluation– Root-zone SM simulations vs. in situ obs– Drought monitoring products vs. standard drought indices and historical records
Proposed drought information system
32X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Impact of Sfc Type on LSM SM simulations
In situ SM: Measurement for every 10 days from 2010-2011LSM SM: Top 10cm layer simulations
Correlation significance: black – no, blue – 0.05 credibility, green – 0.01, red – 0.001
8 0 9 0 1 0 0 1 1 0 1 2 0 1 3 01 7
2 2
2 7
3 2
3 7
4 2
4 7
5 2 N O A H 3 . 2 & A V H R R S T
8 0 9 0 1 0 0 1 1 0 1 2 0 1 3 01 7
2 2
2 7
3 2
3 7
4 2
4 7
5 2 N O A H 3 . 2 & M O D I S S T
8 0 9 0 1 0 0 1 1 0 1 2 0 1 3 01 7
2 2
2 7
3 2
3 7
4 2
4 7
5 2 C L M 2 . 0 & A V H R R S T
8 0 9 0 1 0 0 1 1 0 1 2 0 1 3 01 7
2 2
2 7
3 2
3 7
4 2
4 7
5 2 C L M 2 . 0 & M O D I S S T
33X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Impact of QC on GLDAS Soil Moisture DA
Noah LSM v3.2 was used;
In situ SM data from about 50 SCAN sites were used;
SMOPS SM: SMOS, ASCAT, WindSat
OLP: no SM DA
DA01: SM DA for all GVF
DA02: SM DA for GVF < 0.7
DA03: SM DA for GVF (ST)
Quality controlled SM data assimilation improved Noah LSM simulations
34X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Experiments Parameters Assimilated
OLP Climatological GVF and Climatological albedo
DA01 Monthly albedo
DA02 Monthly GVF
DA03 Weekly GVF
DA04 Monthly GVF and monthly albedo
DA05 Weekly GVF and monthly albedo
Data assimilation experiment matrix
Distributions of RMSE improvement for SM simulations bet’n DA05 and OLP over 2001-2011
Average improved percentage of SM simulations Between DA01- 05 and OLP over
2001 - 2011
Impact of NRT GVF and albedo on Noah LSM SM
35X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
NCEP land data needs
NESDIS land satellite data availability
DA efforts at STAR with NCEP models
SMOPS and SM DA for NCEP NWP
GET-D for drought monitoring
Next steps
Outline
36X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
GFS and LIS “Semi-Coupling”
LIS
GFS Noah
Noah
EnKF
Forcing
States
GFS Noah
LISNoah
EnKF
Forcing
States
37X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Then, what?
38X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Operational mode of land DA?
39X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
More questions?
40X. Zhan, 12th JCSDA Workshop, College Park, MD. April 19, 2023
Thanks!