Next Article in Journal
Geoid Studies in Two Test Areas in Greece Using Different Geopotential Models towards the Estimation of a Reference Geopotential Value
Previous Article in Journal
Advantages of Using Transfer Learning Technology with a Quantative Measurement
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 15
Issue 17
10.3390/rs15174279
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Impact of the Detection Channels Added by Fengyun Satellite MWHS-II at 183 GHz on Global Numerical Weather Prediction

by
Yali Ju
1,2,
Jieying He
1,*,
Gang Ma
3,4,5,
Jing Huang
3,4,5,
Yang Guo
6,7,8,
Guiqing Liu
3,4,5,
Minjie Zhang
1,2,
Jiandong Gong
3,4,5 and
Peng Zhang
6,7,8
1
Key Laboratory of Microwave Remote Sensing, National Space Science Center, Chinese Academy of Sciences, Beijing 100190, China
2
University of Chinese Academy of Sciences, Beijing 100049, China
3
Center for Earth System Modeling and Prediction of CMA, Beijing 100081, China
4
Key Laboratory of Earth System Modeling and Prediction, China Meteorological Administration, Beijing 100081, China
5
State Key Laboratory of Severe Weather (LaSW), Chinese Academy of Meteorological Sciences, China Meteorological Administration, Beijing 100081, China
6
National Satellite Meteorological Center, Beijing 100081, China
7
Key Laboratory of Radiometric Calibration and Validation for Environmental Satellites, Beijing 100081, China
8
Innovation Center for Fengyun Meteorological Satellite (FYSIC), China Meteorological Administration, Beijing 100081, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2023, 15(17), 4279; https://doi.org/10.3390/rs15174279
Submission received: 14 July 2023 / Revised: 22 August 2023 / Accepted: 28 August 2023 / Published: 31 August 2023
Browse Figures
Versions Notes

Abstract

:
Fine spectral detection can basically solve the problem of low vertical resolution at the 183 GHz water-vapor absorption line, and it is expected to become one of the main methods for next-generation geostationary and polar-orbiting satellites. Here, using data from Microwave Humidity Sounder II (MWHS-II) onboard the Chinese Fengyun 3D (FY-3D) satellite in the Global/Regional Assimilation and Prediction System (GRAPES) Four-Dimensional Variational (4D-Var) system of the China Meteorological Administration (CMA), we explore the assimilation application of the water-vapor absorption line at 183.31 ± 1 GHz, 183.31 ± 3 GHz and 183.31 ± 7 GHz, as well as 183.31 ± 1.8 GHz and 183.31 ± 4.5 GHz, two added channels, to assess the impact of adding the 183.31 ± 1.8 GHz and 183.31 ± 4.5 GHz sampling channels on data assimilation and numerical weather prediction. Our findings reveal a significant increase in the specific-humidity increment, which in the middle–upper troposphere is numerically much larger than in the lower troposphere. Specifically, the assimilation of 183.31 ± 1.8 GHz observations, positioned near the center of the water-vapor absorption line, results in a pronounced adjustment compared with the 183.31 ± 4.5 GHz observations. And under the strong constraint of the numerical model, the Root Mean Square Error (RMSE) of the wind field diminishes more significantly (by an average of 2–4%) after assimilating the water-vapor observations at greater heights.

1. Introduction

Water vapor is the only component of the atmosphere that can undergo phase transitions, and the latent heat is directly related to movement, and the changes in temperature and other variables [1,2]. The water-vapor content varies significantly in the atmosphere and can strongly absorb and emit infrared and microwave radiation [3]. The global three-dimensional water-vapor distribution obtained using detection channels at 183 GHz is one of the most important data sources for numerical weather prediction [4,5]. Fine spectral detection can greatly improve the vertical resolution of water vapor [6], and the assimilation of satellite-borne infrared hyperspectral data at 6.7 μ m has shown effectiveness in typhoon forecasting [7]. Although there are no satellite-borne payloads with microwave fine spectra yet, assimilation of fine spectral detection data in this band may have a better impact than infrared hyperspectral data.
At the beginning of this century, National Oceanic and Atmospheric Administration 15 (NOAA-15) with Advanced Microwave Sounding Unit (AMSU) was launched and put into operation to sound water-vapor profiles with channels centered at 183 GHz. At the same time, many studies focus on the assimilation of the observed data and the assessment of the impact on numerical weather prediction [8,9]. Due to the complex vertical structure and a strong scattering attenuation effect [10,11], the brightness temperature on Advanced Microwave Sounding Unit-B/Microwave Humidity Sounder (AMSU-B/MHS) decreases with the increase in ice particles [12]. When assimilation by Advanced Microwave Sounding Unit-A (AMSU-A) prolongs the effective time period of global numerical forecasts by 1–2 days [13], humidity observations are thought to have little effect on global weather forecasts [14]. Therefore, the assimilation conducted by the early AMSU-B/MHS could not have a significant impact on global numerical prediction [15]. In recent years, the all-sky method developed by the European Centre for Medium-Range Weather Forecasts (ECMWF) has solved the problem of assimilation in the cloud–rain environment and has achieved good results [16,17,18,19]. At the ECMWF, data on satellite microwave radiation sensitive to humidity, cloud and precipitation provided 20% of short-range forecast impact between 2012 and 2016 [20]. This makes them some of the most important sources of data, with an impact equivalent to that of microwave temperature sounding observations.
The main absorbing gases in the microwave band are oxygen and water vapor. Due to the much longer wavelength of microwaves compared with visible and infrared light, they have strong penetration into non-precipitation clouds and work under all weather conditions [21]. Since the vertical distribution of water vapor is located below 300 hPa and the peak height of the AMSU-B/MHS weighting function at 183 GHz is from 800 hPa to 400 hPa, it is mainly used for the detection of tropospheric information [22]. However, AMSU-B/MHS has only three channels, with vertical resolution of 6 km, which is much lower than the 1 km resolution of infrared hyperspectral sounding [23] and the 4 km resolution of AMSU-A [24], so it cannot realize fine detection [25].
To overcome this limitation, microwave sensors with additional channels are deployed [26]. Advanced Technology Microwave Sounding (ATMS) is a 22-channel microwave radiometer that combines AMSU-A and MHS channels with one temperature and two humidity additional sounding channels [27]. Sounder for Probing Vertical Profiles of Humidity (SAPHIR) onboard Megha Tropiques (Indo-French joint satellite) operates at 183 GHz with six channels to measure vertical profiles of atmospheric humidity over land and ocean, and the observations are of good quality, according to Refs. [28,29]. Microwave Humidity Sounder II (MWHS-II) is a microwave humidity sounder onboard the Chinese Fengyun 3 polar-orbiting satellite. Centered at 183 GHz, it has five channels, with additional channels at 183.31 ± 1.8 GHz and 183.31 ± 4.5 GHz, where the weighting function peaks at 500 hPa and 700 hPa [30], respectively. The ECMWF assessment shows that the assimilation of MWHS-II observations contributes to numerical weather prediction comparably to AMSU-B/MHS [31].
The fine spectral detection of the 183 GHz absorption line is expected to become one of the main methods for next-generation geostationary and polar-orbiting satellites [32,33]. However, few studies have evaluated the impact of adding the 183.31 ± 1.8 GHz and 183.31 ± 4.5 GHz sampling channels on data assimilation and numerical weather prediction. This article is based on the Global/Regional Assimilation and Prediction System (GRAPES) Four-Dimensional Variational (4D-Var) system of the China Meteorological Administration (CMA), using the real-time observations of Fengyun 3D (FY-3D) MWHS-II to analyze the assimilation impact of the original three channels at 183.31 ± 1 GHz, 183.31 ± 3 GHz and 183.31 ± 7 GHz, and the additional channels at 183.31 ± 1.8 GHz and 183.31 ± 4.5 GHz, respectively.
The paper is organized as follows: Section 2 introduces the series of instruments and methods used in this study. Section 3 illustrates the assimilation and forecast impacts, and the conclusions are given in Section 4.

2. Data and Method

2.1. FY-3D MWHS-II

The FY-3D satellite is the second member of the Fengyun 3 series from the CMA that is designated for operational use. It was successfully launched on 15 November 2017, with a real-time data stream established at the ECMWF in summer 2019 [34]. MWHS-II onboard FY-3D is a 15-channel cross-track radiometer scanning a 2600 km swath in 98 steps at ±53.35° from the nadir [35]. The in-orbit observations are calibrated to obtain level 1 product data. The present study focuses on MWHS-II 183 GHz frequencies, as they are the channels assimilated in operation at the Met Office. The quality of MWHS-II data was found to be comparable to that of ATMS, with mean global biases of the same magnitude, albeit slightly larger and (for some channels) of opposite sign. The noise and scanning-angle biases were shown to be comparable for both instruments, apart from more variability in ATMS data. Low-magnitude striping similar to or smaller than that of ATMS has been observed in MWHS-II data [36,37].
Figure 1 shows the weighting function distribution of the FY-3D MWHS-II channels. Among them, 183.31 ± 1 GHz (channel 11), 183.31 ± 3 GHz (channel 13) and 183.31 ± 7 GHz (channel 15) are located from the center to the window area of the absorbing line. The weighting height gradually decreases as the frequency approaches the edge, which can be used to retrieve atmospheric-humidity profiles. Channels 183.31 ± 1.8 GHz (channel 12) and 183.31 ± 4.5 GHz (channel 14) are a special set weighted at 500 hPa and 700 hPa, respectively, as effective supplements for the three main detection frequency points. Such channel characteristics can compensate for the limitations of conventional observations with their uneven coverage and poor accuracy in humidity detection in the middle–upper troposphere.

2.2. CMA-GFS 4D-Var Global Numerical Forecasting System

Since 1 July 2018, the GRAPES global 4D-Var data assimilation system has been in operation at the CMA, which marked significant progress in Chinese numerical weather prediction [38]. The system runs four assimilation analyses per day with an assimilation window length of 6 h. In addition, 4D-Var can implicitly propagate the initial background error covariance with a tangent-linear model and an adjoint model [39]. The system operation scheme involved in this article is shown in Table 1.
Before assimilation, we need to eliminate the large errors caused by instruments and other reasons to ensure the quality of observations. A quality control scheme including an extreme-value check (to remove the data with brightness temperatures greater than 340 K or lower than 90 K), a surface-type check (to remove data over mixed surface types), a scanning-angle check (to remove the first eight scan angles in each swath edge) and an absolute-departure check (to remove observations whose OMB is greater than three times the standard error after bias correction) is adopted [40]. Moreover, we need to remove the observations with cloud fraction greater than 0.72 in clear-sky assimilation. To weed out the systematic errors between the simulated brightness temperatures and the observed brightness temperatures, the GRAPES variational bias correction scheme is utilized for the bias correction of FY-3D MWHS-II [41,42].
Radiative Transfer for TOVS (RTTOV), a fast radiative transfer model that can simulate the radiation under clear-sky conditions [43], is selected as an observation operator to achieve functional mapping from model variables such as atmospheric temperature and humidity to channel brightness temperature observed by MWHS-II. Based on variational theory and the RTTOV adjoint model, the initial field is optimized and adjusted.

3. Experimental Analysis

3.1. Assimilation and Forecast Experiment of Single-Point Observation

The experiment is based on the orbital observation data of FY-3D MWHS-II on 15 June 2019 at 19:00 UTC, and the sub-satellite point is selected for clear sky over the ocean with number 50 in a scanning line. In the CMA-GFS 4D-Var system, only precipitation detection is retained in the quality control scheme of MWHS-II assimilation.
Figure 2 shows the spatial distribution of brightness-temperature observation of MWHS-II channel 1 at that moment. The selected single point is located in the Southwest Indian Ocean (28.56°S, 86.47°E), and there is a desire to explore the influence on atmospheric circulation caused by the Northeast Monsoon.
In the experiment, the unique observation is analyzed for assimilation. At this time, the assimilation of other satellite observations and conventional observations are turned off. This study applies the assimilation analysis of a single point to two cases: (1) Exp. 1 only considers the observations of channels 11, 13 and 15; (2) Exp. 2 includes the observations of channels 11–15.
Figure 3 shows the distribution of the humidity increment in single-point observation assimilation at 400 hPa, 600 hPa, 800 hPa and 1000 hPa. In the figure, (a) is the result of Exp. 1; (b) is the result of Exp. 2; and (c) is the difference between (b) and (a). In general, the specific-humidity increment triggered by a single point is mostly negative, and the value is extremely small, with a range of 10 − 4 – 10 − 2 g/kg, meaning that the atmosphere becomes dry after assimilation. The increment center almost coincides with the observation point and is distributed as isotropic concentric circles due to the constraint of the time-varying numerical prediction model. When the isobaric surface is 400 hPa, Exp. 1 has an increment of 1.38 × 10 − 2 g/kg; Exp. 2 has an increment of 1.06 × 10 − 2 g/kg; and the difference between them is approximately 3.20 × 10 − 3 g/kg, which indicates that Exp. 2 produces a reduction of 23.19% compared with Exp. 1. At 600 hPa, the increments of Exp. 1 and Exp. 2 are 1.39 × 10 − 2 g/kg and 1.23 × 10 − 2 g/kg, respectively, and Exp. 2 adjusts the value by 11.31%. Compared with the upper troposphere, the incremental values triggered in the lower troposphere are smaller, of the magnitude of 1.00 × 10 − 3 g/kg. Similarly, the adjusted portions of Exp. 2 at 800 hPa and 1000 hPa are 5.99% and 8.59%.
Overall, compared with assimilating the observations of only channels 11, 13 and 15, increasing the observation data of channels 12 and 14 diminishes, by least 10%, the humidity increment in the upper atmosphere, while the value remains 6% on average in the lower atmosphere. That is, the 183.31 ± 1.8 GHz frequency point near the center of the water-vapor absorption line has a greater impact on data assimilation than the 183.31 ± 4.5 GHz frequency farther away from the center, because it has much greater geographical coverage.
Figure 4 displays the meridional changes in specific humidity, temperature, U wind field and V wind field for Exp. 1 and Exp. 2. Figure 4a shows the distribution of humidity increments. In Exp. 1, the center of the increment extreme value is located at 500 hPa, with an intensity of approximately 1.49 × 10 − 2 g/kg, an impact height of 1000–300 hPa and an impact range of 35°S–23°S. In Exp. 2, the extreme center is adjusted from 500 hPa to 600 hPa, with an intensity of approximately 1.27 × 10 − 2 g/kg. The result indicates that Exp. 2 has an intensity reduction of 14.56% compared with Exp. 1 but that the impact height and range are basically the same. Figure 4b shows that the temperature increment center of Exp. 1 is located at 500 hPa, with an intensity of approximately 4.84 × 10 − 3 K. The height of the extreme center in Exp. 2 is adjusted to 600 hPa, with a strength of approximately 4.38 × 10 − 3 K. The observation assimilation of two newly added 183 GHz absorption lines results in a decrease of approximately 9.50% in the temperature increment. The main impact range of the observation point after assimilation is between 800 hPa and 400 hPa, and in the range of 45°S–15°S. In addition, the humidity and temperature centers coincide with the position of the observation point. Figure 4c shows the distribution of U-wind-field increments. After assimilation, the U wind field is evenly distributed with positive and negative increments in the north–south direction at a single point, with values around 10 − 3 m/s that can be almost ignored. The increase in the number of water-vapor detection channels in Exp. 2 does not change the distribution shape of the increment but decreases the maximum value, with a maximum reduction of about 20%. Figure 4d shows the meridional profile of the increment in the V wind field. The center coordinates of the extreme value are (−28.5°S, 88°E), slightly eastward compared with the point, reaching the maximum height at about 400 hPa. Exp. 2 is reduced by 14.29% compared with Exp. 1.
Figure 5 shows the analysis increments in specific humidity, temperature, U wind field and V wind field in the latitude direction in two experiments. The distribution of latitude profiles is the same as in Figure 4. The influencing height in Exp. 1 and Exp. 2 is 1000–300 hPa, and the range is 78°E–92°E (Figure 5a). The impact range of the temperature increment is also basically the same, about 70°E–95°E (Figure 5b). The U-wind-field increment has a positive increment center at 300 hPa, which is significantly reduced in numerical value compared with the meridional increment distribution (Figure 5c). And the V wind field is evenly distributed with positive and negative increments in the east–west direction at a single point after assimilation (Figure 5d).
As for single-point forecasting, the observation is limited to the range of 75°E–100°E and 40°S–20°S because of the small incremental impact. And the time window is limited to 6 h because of the fast dissipation, with output within 1 h, 3 h and 6 h, respectively. We take logarithmic functions with a base of 10 for the assimilation and prediction of humidity fields in Exp. 1 and Exp. 2 and then compare them. By observing the changes in their differences, we explore the impact at different isobaric surfaces. Index A is defined as
A = log 10 Q i , k m − log 10 Q i , k n
where Q is the humidity forecast field, m denotes the forecast after Exp. 2 assimilation, n denotes the forecast after Exp. 1 assimilation, i denotes the forecast moment and k is the isobaric pressure surface.
Figure 6 shows the evolution of index A at different isobars. The increments at the beginning of the window are local to the observation, but the evolved increments are not at all local. A positive increment is triggered near the single point at 300 hPa, and the intensity value of A is at 10 − 2 , mapped in the range of 81°E–93°E and 33°S–22°S. After 3 h of forecasting, the center of the incremental extreme moves eastward, and the range gradually narrows, with scattered negative increments near the point. After 6 h of forecasting, the positive center continues to move towards north from east, and the shadowing range further narrows. The intensity value of index A decreases to 10 − 4 , indicating that the effect of the humidity increment gradually dissipated as the forecast time increased. The forecast humidity increment range at 500 hPa is relatively stable, approximately within the range of 33°S–22°S and 81°E–91°E. The extreme center approximately coincides with the observation point, resulting in a positive increment of 10 − 2 orders of magnitude in intensity. The negative increment in humidity increases as the forecast duration increases, meaning that the original positive increment is gradually dissipated. Relative to 300 hPa and 500 hPa, the situation is similar to that at 700 hPa and 850 hPa. The result of the 1 h forecast shows that the positive and negative humidity increments are evenly distributed over the region, and the single point is at the boundary between them. The positive incremental shadowing range and intensity tend to increase, while the negative increments tend to decrease with the increase in forecast duration, and the positive increments show a tendency to move southwest with the forecast. After 6 h of forecasting at 850 hPa, all negative increments are dissipated.
It can be seen that the complex evolution of atmospheric water vapor makes assimilation have a much lower impact on forecasting. The enhanced water-vapor remote sensing detection information in the middle and upper troposphere has a stronger impact than that in the lower troposphere, which can be distinguished by index A. However, the impact is basically dissipated after 6 h of forecasting. Meanwhile, the water-vapor increment shows a characteristic of moving with the forecast in the meridional direction, and the newly added water vapor affects the movement direction of the upper troposphere in the opposite direction with respect to the lower troposphere.

3.2. Assimilation and Forecast Experiment of Single-Moment Observations

The experiment is based on orbital observation data of FY-3D MWHS-II on 15 June 2019 from 15:00 to 21:00 UTC, and the observation distribution of channel 1 is shown in Figure 7. The figure shows that a time window contains three orbits, mainly covering the Western Pacific, Southeast Pacific, Southeast Indian Ocean, North Atlantic and Southwest Atlantic regions. The assimilation and forecast experiment of a single moment adopts all quality control schemes in the MWHS-II assimilation operator and turns off the assimilation of all other satellite observations and conventional observations. The experimental setup is basically the same as that of the single-point experiment: (1) Exp. 1 only considers the observations of channels 11, 13 and 15; (2) Exp. 2 includes the observations of channels 11–15.
In this group of experiments, the trend of the humidity increment on each isobaric surface is similar to that in the single point experiment, and the peak value between 700 hPa and 500 hPa is 0.12 g/kg. However, more observations are assimilated, and the area of increment distribution is larger (figures omitted).
By comparing the assimilation analysis fields of Exp. 1 and Exp. 2 with National Centers for Environmental Prediction (NCEP) Final Reanalysis Data (FNL), the tomographic Root Mean Square Error (RMSE) distribution can be obtained as shown in Figure 8. Figure 8a represents the specific humidity, and we can see that the RMSE values of Exp. 1 and Exp. 2 are basically the same in the Southern Hemisphere, and their differences are negligible. In the Northern Hemisphere, due to the fact that near-surface data on vast land have not entered the assimilation system, regions with different RMSE values between the two experiments are distributed in the upper atmosphere (700 hPa), but the RMSE of Exp. 2 is still smaller. Figure 8b shows the RMSE distribution of temperature, and less than 1% difference can be found in the Southern Hemisphere from 925 hPa to 500 hPa, while less than 1% difference can be found in the Northern Hemisphere from 500 hPa to 300 hPa, for the same reasons as in Figure 8a. Figure 8c,d represent the RMSE distributions of U and V wind fields, respectively. It can be seen that under the constraints of the numerical model, the RMSE values of Exp. 1 and Exp. 2 are obviously different throughout the whole atmosphere, regardless of whether it is in the Southern or Northern Hemisphere. Moreover, the RMSE of Exp. 2 has smaller values than that of Exp. 1, indicating a significant decrease in wind field after assimilating more water-vapor detection data at higher altitudes.
Based on the above analysis, the observation assimilation of two new channels in the troposphere can effectively improve the RMSE of wind fields, but the improvement effect is also extremely weak.
Similar to the idea of single-point observation experiments, the assimilation of water-vapor data under clear-sky conditions has a weak impact on global forecasting. Therefore, the difference between the two humidity forecast fields after assimilation is used. We define index B:
B = P i , k m − P i , k n
where m denotes the observed brightness temperature of assimilated MWHS-II channels 11–15; n denotes the observed brightness temperature of assimilated MWHS-II channels 11, 13 and 15; i is the forecast moment; k is the isobaric surface; and P is the difference between the forecast variable after assimilating MWHS-II-observed brightness temperature and the forecast variable without assimilation. We define
P = P A − P C , P { Q , H , U , V }
where Q denotes the specific humidity, H denotes the potential height, U denotes the U-component of the wind filed and V denotes the V-component of wind field. A denotes the forecast field after assimilating the brightness temperature observed in MWHS-II, and C denotes the forecast field without observation assimilation in the control experiment. In the examination of forecast fields, the global forecast model is set to 240 h, with an output interval of 24 h. Forecast fields within 72 h are selected for investigation, and the distributions of incremental index B for each isobaric surface are shown in Figure 9, Figure 10 and Figure 11.
Figure 9 gives the distribution of index B Q at 300 hPa, 500 hPa, 700 hPa and 850 hPa for 24 h, 48 h and 72 h forecasts. At the beginning, the value range of the humidity increment is −0.300–0.300 g/kg at 300 hPa, and the location is between 30°S and 60°N. A positive incremental extreme appears at (10°N, 180°). From 24 h to 72 h of forecasting, the incremental range spreads significantly, and the intensity also increases obviously. From 850 hPa to 500 hPa, the humidity increment is in the range of −5.0–5.0 g/kg, and the overall distribution is between 60°S and 90°N, with a positive increment in southern South America. In addition, it can be seen that the range and intensity decrease gradually as the height increases.
Figure 10 gives the distribution of index B H at each isobaric surface for 24 h, 48 h and 72 h forecasts. In the 24 h forecast at 300 hPa, the positive and negative increments in the geopotential-height field are distributed equally. The B H values exist in the −15–15 m interval, ranging from 30°N to 90°N and from 90°S to 30°S, and two negative centers are (70°S, 130°E) and (60°S, 150°E). The incremental range expands, and the intensity increases gradually as the forecast time and the height increase. In general, the variation in the geopotential-height field is independent of humidity, and there is no significant difference after adding the observations of assimilated 183 GHz absorption-line channels.
Similarly, Figure 11 gives the distribution of index B U . At 300 hPa, for the 24 h forecast, the positive and negative increments in the U wind field are evenly distributed in the global area, and the values are in the range of −16–16 m/s, with no obvious extreme centers. However, there are more and more increments appearing in southern South America while forecasting for 72 h. The incremental range expands, and the intensity increases gradually as the forecast time and the height increase. And the V wind field is similar to the U wind field, so we will not repeat the explanation here.

3.3. Forecast Impact Assessment of Time Series

Based on the MWHS-II assimilation operator in the CMA-GFS 4D-Var system, assimilation analysis is conducted four times a day from 15 June 2019 to 25 June 2019, followed by 240 h forecasts. The assimilation and forecast experiments are designed as follows: (1) Exp. 1 only considers the observations of MWHS-II channels 11, 13 and 15; (2) Exp. 2 includes the observations of channels 11–15. In the two groups of experiments, the global forecast model settings are identical, and the RMSE of each variable is calculated by comparing the results with NCEP FNL data.
Figure 12 displays the RMSE values of the specific-humidity and wind forecast fields in Exp. 1 and Exp. 2. Overall, Exp. 2 shows slightly negative impacts on humidity (1.06% error increase at 300 hPa and 1.39% at 500 hPa) in the Southern Hemisphere compared with Exp. 1. But there is almost no significant difference in the two experiments at middle–lower levels (Figure 12c,d). All detection data were rejected in the Northern Hemisphere with a wide land area, so the values of the RMSE are closer to that of Exp. 1, which means that neutral impacts are achieved. The increase trend of the Northern Hemisphere RMSE is not obvious at all four altitudes, with a maximum increase of about 1.39% at 700 hPa, 192 h.
Unlike the humidity field, the wind-field RMSE of setting five channels at 300 hPa decreases by an average of about 2.37% (before 144 h) compared with setting three channels, and the maximum decrease is about 3.02% (forecast at 96 h) in the Southern Hemisphere (Figure 12a). At the height of 500 hPa, the average decrease is about 2.64% (before 168 h), and the maximum decrease is about 3.86% (forecast at 120 h). The average and maximum reductions at 700 hPa are 2.71% (before 192 h) and 4.24% (forecast at 96 h), respectively. Finally, the values at 850 hPa are 2.47% before 192 h and 3.78% at the forecast time of 120 h. In the Northern Hemisphere, the average decrease at four isobars is between 1.21% and 1.59%, with a maximum decrease of about 2.39% (700 hPa, 120 h). Exp. 2 shows a positive impact on the forecasts of the V-component wind field, and the error reduction in RMSE in the Southern Hemisphere is much higher than that in the Northern Hemisphere.
The trend of forecast errors for other variables is similar, and the figures are omitted. Based on the above analysis, the addition of the 183 GHz absorption line in the troposphere for assimilation shows that there is a neutral impact on the specific-humidity forecast field but a slightly positive impact on the V-component of wind in short-term forecasting.

4. Conclusions and Discussion

Based on the data from MWHS-II onboard the Chinese FY-3D satellite in the CMA-GFS 4D-Var system, to assess the impact of adding two channels at 183.31 ± 1.8 GHz and 183.31 ± 4.5 GHz sampling on global numerical weather prediction, we analyze the assimilation application of the water-vapor absorption line at 183.31 ± 1 GHz, 183.31 ± 3 GHz and 183.31 ± 7 GHz, as well as at 183.31 ± 1.8 GHz and 183.31 ± 4.5 GHz, two added channels. The following conclusions are obtained:
  • In the analysis of single-point assimilation, each isobaric surface shows that the specific-humidity increment is extremely small under clear-sky conditions, but the value of the upper troposphere is much greater than that of the lower troposphere. Compared with assimilating the observations of only channels 11, 13 and 15, increasing the observation data of channels 12 and 14 shows an adjustment of at least 10% for the specific-humidity increment in the upper atmosphere, while the value remains at an average of 6% in the lower atmosphere. That is, the 183.31 ± 1.8 GHz frequency point near the center of the water-vapor absorption line has a greater impact on data assimilation than 183.31 ± 4.5 GHz, which is far from the center.
  • In the analysis of single-point forecasting, the incremental value decreases gradually and is basically dissipated after 6 h of forecasting. The dissipation rate is slower in the middle and upper troposphere and faster in the lower troposphere, which makes the impact of the two additional 183 GHz observation channels on the middle and upper troposphere stronger than that of the lower troposphere.
  • In the analysis of single-moment assimilation, there is almost no difference (less than 1%) in the RMSE of the specific-humidity and temperature fields, but there is a relatively significant decrease in the RMSE value of the wind field after the observation assimilation of newly added channels 12 and 14. The difference between the two sets of RMSE in the Northern Hemisphere is also reflected in the higher atmosphere due to the lack of assimilation of near-surface data over vast land areas.
  • In the analysis of single-moment forecasting, the humidity increment diminishes with the increase in height. However, for the potential-height field, the U-component of the wind field and the V-component of the wind field, their increments increase with the increase in height. And the absolute value of the incremental change is still small and needs to be amplified with an index.
  • The results of the 10-day assimilation and forecast experiments show that there is a neutral impact on the specific-humidity forecast field but a slightly positive impact on the V-component of wind in the short term after adding 183 GHz channels for assimilation in the troposphere. The closer to the new detection frequency the assimilation of data is conducted, the more obvious the decreasing trend is. And the decreases at 500 hPa and 700 hPa are 10% higher than those at other altitudes.
However, the evaluation of the results is dependent on the accuracy of the observation data and the verification method. First, there might still be some problems in the quality of observations and their matching with model grids, although the observation data have undergone a series of quality control processes. If a grid value of the forecast is matched to the nearest observation with great difference, but it is not a truth, there is a large error. In this situation, verification based on observation data is unreliable. Second, the spatial distribution of the observations is also an important factor that directly affects the evaluation of the results. It determines which grid data of the forecast are verified. Finally, many different methods have been developed for forecast verification. In this study, the score skills of RMSE are employed, which can just give an assessment from a certain view point.
In addition, only MWHS-II radiance values under clear-sky conditions are assimilated in the experiments. The scattering index developed by Bennartz et al. [44] is used to perform cloud and precipitation detection. It is simply defined as the difference between two window channels, which cannot accurately identify all clouds. Future research could focus on a new method for cloud detection using cloud products or MWHS-II radiance data assimilation under all-sky conditions. Meanwhile, the extension of the experiments to other Fengyun 3 satellites should be also considered and investigated in the future.

Author Contributions

All authors (Y.J., G.M., J.H. (Jieying He), J.H. (Jing Huang), Y.G., G.L., M.Z., J.G., P.Z.) participated in the design, data collection, data interpretation and critical review of the article. Y.J. performed the statistical analysis and wrote the manuscript. All authors approved the final version. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the National Key Research and Development Program of China (2022YFB3902601) and the Youth Cross Team Scientific Research Project of the Chinese Academy of Sciences (JCTD-2021-10).

Data Availability Statement

The FY-3D satellite data can be downloaded from the website of the China National Satellite Meteorological Center (http://satellite.nsmc.org.cn/portalsite/default.aspx, accessed on 1 July 2022). The NCEP FNL data can be downloaded from the National Center for Atmospheric Research (https://rda.ucar.edu/datasets/ds083.2/, accessed on 27 February 2023).

Acknowledgments

We thank the National Satellite Meteorological Center for providing data from Fengyun 3 Satellite/Microwave Humidity Sounder II, to European Meteorological Satellite team for their technical support, and the editors and anonymous reviewers for their suggestions on modifying the article.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Held, I.M.; Soden, B.J. Water vapor feedback and global warming. Annu. Rev. Energy Environ. 2000, 25, 441–475. [Google Scholar] [CrossRef]
  2. Tu, J.; Lu, E. Relative importance of water vapor and air temperature in the interannual variation of the seasonal precipitation: A comparison of the physical and statistical methods. Clim. Dyn. 2020, 54, 3655–3670. [Google Scholar] [CrossRef]
  3. Shine, K.P.; Ptashnik, I.V.; Rädel, G. The water vapour continuum: Brief history and recent developments. Surv. Geophys. 2012, 33, 535–555. [Google Scholar] [CrossRef]
  4. Andersson, E.; Hólm, E.; Bauer, P.; Beljaars, A.; Kelly, G.; McNally, A.; Simmons, A.; Thépaut, J.N.; Tompkins, A. Analysis and forecast impact of the main humidity observing systems. Q. J. R. Meteorol. Soc. J. Atmos. Sci. Appl. Meteorol. Phys. Oceanogr. 2007, 133, 1473–1485. [Google Scholar] [CrossRef]
  5. Andersson, E.; Hólm, E.; Thépaut, J. Impact studies of main types of conventional and satellite humidity data. In Proceedings of the 3rd WMO Workshop on The Impact of Various Observing Systems on NWP, Alpbach, Austria, 9–12 March 2004. [Google Scholar]
  6. Geer, A.J.; Migliorini, S.; Matricardi, M. All-sky assimilation of infrared radiances sensitive to mid-and upper-tropospheric moisture and cloud. Atmos. Meas. Tech. 2019, 12, 4903–4929. [Google Scholar] [CrossRef]
  7. Yin, R.; Han, W.; Wang, H.; Wang, J. Impacts of FY-4A GIIRS Water Vapor Channels Data Assimilation on the Forecast of “21· 7” Extreme Rainstorm in Henan, China with CMA-MESO. Remote Sens. 2022, 14, 5710. [Google Scholar] [CrossRef]
  8. Eyre, J.R.; English, S.J.; Forsythe, M. Assimilation of satellite data in numerical weather prediction. Part I: The early years. Q. J. R. Meteorol. Soc. 2020, 146, 49–68. [Google Scholar] [CrossRef]
  9. Eyre, J.; Bell, W.; Cotton, J.; English, S.; Forsythe, M.; Healy, S.; Pavelin, E. Assimilation of satellite data in numerical weather prediction. Part II: Recent years. Q. J. R. Meteorol. Soc. 2022, 148, 521–556. [Google Scholar] [CrossRef]
  10. Adler, R.F.; Mack, R.A.; Prasad, N.; Hakkarinen, I.M.; Yeh, H.M. Aircraft microwave observations and simulations of deep convection from 18 to 183 GHz. Part I: Observations. J. Atmos. Ocean. Technol. 1990, 7, 377–391. [Google Scholar] [CrossRef]
  11. Yeh, H.Y.M.; Prasad, N.; Mack, R.A.; Adler, R.F. Aircraft microwave observations and simulations of deep convection from 18 to 183 GHz. Part II: Model results. J. Atmos. Ocean. Technol. 1990, 7, 392–410. [Google Scholar] [CrossRef]
  12. Muller, B.M.; Fuelberg, H.E.; Xiang, X. Simulations of the effects of water vapor, cloud liquid water, and ice on AMSU moisture channel brightness temperatures. J. Appl. Meteorol. Climatol. 1994, 33, 1133–1154. [Google Scholar] [CrossRef]
  13. English, S.; Renshaw, R.; Dibben, P.; Smith, A.; Rayer, P.; Poulsen, C.; Saunders, F.; Eyre, J. A comparison of the impact of TOVS arid ATOVS satellite sounding data on the accuracy of numerical weather forecasts. Q. J. R. Meteorol. Soc. 2000, 126, 2911–2931. [Google Scholar] [CrossRef]
  14. Bengtsson, L.; Hodges, K.I. On the impact of humidity observations in numerical weather prediction. Tellus A Dyn. Meteorol. Oceanogr. 2005, 57, 701–708. [Google Scholar] [CrossRef]
  15. Di Tomaso, E.; Bormann, N. Assimilation of ATOVS Radiances at ECMWF: Second Year EUMETSAT Fellowship Report; ECMWF: Reading, UK, 2012.
  16. Bauer, P.; Geer, A.J.; Lopez, P.; Salmond, D. Direct 4D-Var assimilation of all-sky radiances. Part I: Implementation. Q. J. R. Meteorol. Soc. 2010, 136, 1868–1885. [Google Scholar] [CrossRef]
  17. Geer, A.J.; Bauer, P.; Lopez, P. Direct 4D-Var assimilation of all-sky radiances. Part II: Assessment. Q. J. R. Meteorol. Soc. 2010, 136, 1886–1905. [Google Scholar] [CrossRef]
  18. Candy, B.; Migliorini, S. The assimilation of microwave humidity sounder observations in all-sky conditions. Q. J. R. Meteorol. Soc. 2021, 147, 3049–3066. [Google Scholar] [CrossRef]
  19. Geer, A.J. All-Sky Assimilation: Better Snow-Scattering Radiative Transfer and Addition of SSMIS Humidity Sounding Channels; European Centre for Medium-Range Weather Forecasts: Reading, UK, 2013; Volume 706.
  20. Geer, A.; Baordo, F.; Bormann, N.; Chambon, P.; English, S.; Kazumori, M.; Lawrence, H.; Lean, P.; Lonitz, K.; Lupu, C. The growing impact of satellite observations sensitive to humidity, cloud and precipitation. Q. J. R. Meteorol. Soc. 2017, 143, 3189–3206. [Google Scholar] [CrossRef]
  21. Toong, H.; Staelin, D. Passive microwave spectrum measurements of atmospheric water vapor and clouds. J. Atmos. Sci. 1970, 27, 781–784. [Google Scholar] [CrossRef]
  22. Ferraro, R.R.; Weng, F.; Grody, N.C.; Zhao, L.; Meng, H.; Kongoli, C.; Pellegrino, P.; Qiu, S.; Dean, C. NOAA operational hydrological products derived from the advanced microwave sounding unit. IEEE Trans. Geosci. Remote Sens. 2005, 43, 1036–1049. [Google Scholar] [CrossRef]
  23. Smith, W., Sr.; Revercomb, H.; Bingham, G.; Larar, A.; Huang, H.; Zhou, D.; Li, J.; Liu, X.; Kireev, S. Evolution, current capabilities, and future advances in satellite ultra-spectral IR sounding. Atmos. Chem. Phys. Discuss. 2009, 9, 5563–5574. [Google Scholar] [CrossRef]
  24. Zou, X.; Qin, Z.; Weng, F. Impacts from assimilation of one data stream of AMSU-A and MHS radiances on quantitative precipitation forecasts. Q. J. R. Meteorol. Soc. 2017, 143, 731–743. [Google Scholar] [CrossRef]
  25. Rabier, F. Overview of global data assimilation developments in numerical weather-prediction centres. Q. J. R. Meteorol. Soc. J. Atmos. Sci. Appl. Meteorol. Phys. Oceanogr. 2005, 131, 3215–3233. [Google Scholar] [CrossRef]
  26. Muth, C.; Lee, P.S.; Shiue, J.C.; Webb, W.A. Advanced technology microwave sounder on NPOESS and NPP. In Proceedings of the IGARSS 2004—2004 IEEE International Geoscience and Remote Sensing Symposium, IEEE, Anchorage, AK, USA, 20–24 September 2004; Volume 4, pp. 2454–2458. [Google Scholar]
  27. Niels, B.; Anne, F.; William, B. Evaluation and assimilation of ATMS data in the ECMWF system. Tech. Memo. 2012, 689, 16. [Google Scholar]
  28. Mathur, A.; Gangwar, R.; Gohil, B.; Deb, S.K.; Kumar, P.; Shukla, M.V.; Simon, B.; Pal, P. Humidity profile retrieval from SAPHIR on-board the Megha-Tropiques. Curr. Sci. 2013, 25, 1650–1655. [Google Scholar]
  29. Singh, R.; Ojha, S.P.; Kishtawal, C.; Pal, P. Quality assessment and assimilation of Megha-Tropiques SAPHIR radiances into WRF assimilation system. J. Geophys. Res. Atmos. 2013, 118, 6957–6969. [Google Scholar] [CrossRef]
  30. Kan, W.; Han, Y.; Weng, F.; Guan, L.; Gu, S. Multisource assessments of the FengYun-3D microwave humidity sounder (MWHS) on-orbit performance. IEEE Trans. Geosci. Remote. Sens. 2020, 58, 7258–7268. [Google Scholar] [CrossRef]
  31. Lawrence, H.; Bormann, N.; Geer, A.J.; Lu, Q.; English, S.J. Evaluation and assimilation of the microwave sounder MWHS-2 onboard FY-3C in the ECMWF numerical weather prediction system. IEEE Trans. Geosci. Remote Sens. 2018, 56, 3333–3349. [Google Scholar] [CrossRef]
  32. Blackwell, W.J.; Bickmeier, L.J.; Leslie, R.V.; Pieper, M.L.; Samra, J.E.; Surussavadee, C.; Upham, C.A. Hyperspectral microwave atmospheric sounding. IEEE Trans. Geosci. Remote Sens. 2010, 49, 128–142. [Google Scholar] [CrossRef]
  33. Kummerow, C.D.; Poczatek, J.C.; Almond, S.; Berg, W.; Jarrett, O.; Jones, A.; Kantner, M.; Kuo, C.P. Hyperspectral microwave sensors—Advantages and limitations. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2021, 15, 764–775. [Google Scholar] [CrossRef]
  34. Zhang, S.; Li, J.; Jiang, J.; Zhang, Y.; Wang, Z.; Dong, X. Microwave humidity sounder (MWHS) of Chinese meteorological satellite FY-3. In Proceedings of the Microwave Technology and Techniques Workshop-Enabling Future Space Systems, ESA/ESTEC, Noordwijk, The Netherlands, 15–16 May 2006. [Google Scholar]
  35. Zhang, S.; Li, J.; Wang, Z.; Wang, H.; Sun, M.; Jiang, J.; He, J. Design of the second generation microwave humidity sounder (MWHS-II) for Chinese meteorological satellite FY-3. In Proceedings of the 2012 IEEE International Geoscience and Remote Sensing Symposium, Munich, Germany, 22–27 July 2012; pp. 4672–4675. [Google Scholar]
  36. Guo, Y.; He, J.; Gu, S.; Lu, N. Calibration and validation of Feng Yun-3-D microwave humidity sounder II. IEEE Geosci. Remote Sens. Lett. 2019, 17, 1846–1850. [Google Scholar] [CrossRef]
  37. Carminati, F.; Atkinson, N.; Candy, B.; Lu, Q. Insights into the microwave instruments onboard the Fengyun 3D satellite: Data quality and assimilation in the Met Office NWP system. Adv. Atmos. Sci. 2021, 38, 1379–1396. [Google Scholar] [CrossRef]
  38. Zhang, L.; Liu, Y.; Liu, Y.; Gong, J.; Lu, H.; Jin, Z.; Tian, W.; Liu, G.; Zhou, B.; Zhao, B. The operational global four-dimensional variational data assimilation system at the China Meteorological Administration. Q. J. R. Meteorol. Soc. 2019, 145, 1882–1896. [Google Scholar] [CrossRef]
  39. Zhang, L.; Chen, J. An improvement of the linearized planetary boundary layer parameterization scheme for CMA-GFS 4DVar. J. Appl. Meteor. Sci. 2023, 34, 15–26. [Google Scholar]
  40. Yang, Y.; Wei, H.; Peiming, D. Overview on the quality control in assimilation of AMSU microwave sounding data. Meteorol. Mon. 2011, 37, 1395–1401. [Google Scholar]
  41. Xie, X.; Wu, S.; Xu, H.; Yu, W.; He, J.; Gu, S. Ascending–descending bias correction of microwave radiation imager on board FengYun-3C. IEEE Trans. Geosci. Remote Sens. 2018, 57, 3126–3134. [Google Scholar] [CrossRef]
  42. Harris, B.; Kelly, G. A satellite radiance-bias correction scheme for data assimilation. Q. J. R. Meteorol. Soc. 2001, 127, 1453–1468. [Google Scholar] [CrossRef]
  43. Saunders, R.; Hocking, J.; Turner, E.; Rayer, P.; Rundle, D.; Brunel, P.; Vidot, J.; Roquet, P.; Matricardi, M.; Geer, A.; et al. An update on the RTTOV fast radiative transfer model (currently at version 12). Geosci. Model Dev. 2018, 11, 2717–2737. [Google Scholar] [CrossRef]
  44. Bennartz, R.; Thoss, A.; Dybbroe, A.; Michelson, D.B. Precipitation analysis using the Advanced Microwave Sounding Unit in support of nowcasting applications. Meteorol. Appl. 2002, 9, 177–189. [Google Scholar] [CrossRef]
Remotesensing 15 04279 g001
Figure 1. Weighting function for 5 channels at 183 GHz of FY-3D MWHS-II.
Figure 1. Weighting function for 5 channels at 183 GHz of FY-3D MWHS-II.
Remotesensing 15 04279 g001
Remotesensing 15 04279 g002
Figure 2. Spatial distribution of brightness observation of MWHS-II channel 1 at 19:00 UTC on 15 June 2019. The black circle indicates the selected single-point location (28.56°S, 86.47°E).
Figure 2. Spatial distribution of brightness observation of MWHS-II channel 1 at 19:00 UTC on 15 June 2019. The black circle indicates the selected single-point location (28.56°S, 86.47°E).
Remotesensing 15 04279 g002
Remotesensing 15 04279 g003
Figure 3. Spatial distribution of the observed specific-humidity assimilation increments for (a) Exp. 1, (b) Exp. 2, and (c) the difference between Exp. 2 and Exp. 1 at different isobaric surfaces.
Figure 3. Spatial distribution of the observed specific-humidity assimilation increments for (a) Exp. 1, (b) Exp. 2, and (c) the difference between Exp. 2 and Exp. 1 at different isobaric surfaces.
Remotesensing 15 04279 g003
Remotesensing 15 04279 g004
Figure 4. Meridional profile of assimilation analysis increments in (a) specific humidity, (b) temperature, (c) U wind field and (d) V wind field for Exp. 1 (left panel) and Exp. 2 (right panel).
Figure 4. Meridional profile of assimilation analysis increments in (a) specific humidity, (b) temperature, (c) U wind field and (d) V wind field for Exp. 1 (left panel) and Exp. 2 (right panel).
Remotesensing 15 04279 g004
Remotesensing 15 04279 g005
Figure 5. As in Figure 4, but for latitudinal profiles.
Figure 5. As in Figure 4, but for latitudinal profiles.
Remotesensing 15 04279 g005
Remotesensing 15 04279 g006
Figure 6. Impact on forecasts of Exp. 1 and Exp. 2 after assimilation at (a) 1-h forecast field, (b) 3-h forecast field and (c) 6-h forecast field.
Figure 6. Impact on forecasts of Exp. 1 and Exp. 2 after assimilation at (a) 1-h forecast field, (b) 3-h forecast field and (c) 6-h forecast field.
Remotesensing 15 04279 g006
Remotesensing 15 04279 g007
Figure 7. As in Figure 2, but for 15:00–21:00 UTC.
Figure 7. As in Figure 2, but for 15:00–21:00 UTC.
Remotesensing 15 04279 g007
Remotesensing 15 04279 g008
Figure 8. The RMSE of assimilation analyses of (a) specific humidity, (b) temperature, (c) U wind field and (d) V wind field.
Figure 8. The RMSE of assimilation analyses of (a) specific humidity, (b) temperature, (c) U wind field and (d) V wind field.
Remotesensing 15 04279 g008
Remotesensing 15 04279 g009
Figure 9. Difference B Q between Exp. 1 and Exp. 2 on the impact of forecasting at 300 hPa (first row), 500 hPa (second row), 700 hPa (third row) and 850 hPa (fourth row).
Figure 9. Difference B Q between Exp. 1 and Exp. 2 on the impact of forecasting at 300 hPa (first row), 500 hPa (second row), 700 hPa (third row) and 850 hPa (fourth row).
Remotesensing 15 04279 g009
Remotesensing 15 04279 g010
Figure 10. As in Figure 9, but for potential-height-field B H .
Figure 10. As in Figure 9, but for potential-height-field B H .
Remotesensing 15 04279 g010
Remotesensing 15 04279 g011
Figure 11. As in Figure 9, but for the U-component of wind-field B U .
Figure 11. As in Figure 9, but for the U-component of wind-field B U .
Remotesensing 15 04279 g011
Remotesensing 15 04279 g012
Figure 12. Time series of the forecast field RMSE values from 0 to 240 h at (a) 300 hPa, (b) 500 hPa, (c) 700 hPa and (d) 850 hPa. The red line indicates the specific-humidity field, and the blue line indicates the V wind field.
Figure 12. Time series of the forecast field RMSE values from 0 to 240 h at (a) 300 hPa, (b) 500 hPa, (c) 700 hPa and (d) 850 hPa. The red line indicates the specific-humidity field, and the blue line indicates the V wind field.
Remotesensing 15 04279 g012
Table 1. The operational settings of the CMA-GFS 4D-Var data assimilation system.
Table 1. The operational settings of the CMA-GFS 4D-Var data assimilation system.
Horizontal Resolution0.25°/1.0° (Outer Loop/Inner Loop)
Vertical layers87
Assimilation window6 h
Observation time slot30 min
Maximum number of minimization iterations50
Model top0.1 hPa
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Ju, Y.; He, J.; Ma, G.; Huang, J.; Guo, Y.; Liu, G.; Zhang, M.; Gong, J.; Zhang, P. Impact of the Detection Channels Added by Fengyun Satellite MWHS-II at 183 GHz on Global Numerical Weather Prediction. Remote Sens. 2023, 15, 4279. https://doi.org/10.3390/rs15174279

AMA Style

Ju Y, He J, Ma G, Huang J, Guo Y, Liu G, Zhang M, Gong J, Zhang P. Impact of the Detection Channels Added by Fengyun Satellite MWHS-II at 183 GHz on Global Numerical Weather Prediction. Remote Sensing. 2023; 15(17):4279. https://doi.org/10.3390/rs15174279

Chicago/Turabian Style

Ju, Yali, Jieying He, Gang Ma, Jing Huang, Yang Guo, Guiqing Liu, Minjie Zhang, Jiandong Gong, and Peng Zhang. 2023. "Impact of the Detection Channels Added by Fengyun Satellite MWHS-II at 183 GHz on Global Numerical Weather Prediction" Remote Sensing 15, no. 17: 4279. https://doi.org/10.3390/rs15174279

APA Style

Ju, Y., He, J., Ma, G., Huang, J., Guo, Y., Liu, G., Zhang, M., Gong, J., & Zhang, P. (2023). Impact of the Detection Channels Added by Fengyun Satellite MWHS-II at 183 GHz on Global Numerical Weather Prediction. Remote Sensing, 15(17), 4279. https://doi.org/10.3390/rs15174279

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop