Abstract
This study aims to comprehend the propagation of meteorological drought [expressed by the standardized precipitation evapotranspiration index (SPEI)] into hydrological drought [expressed by the standardized runoff index (SRI)] using the combined application of principal component analysis (PCA) and wavelet analysis for a period of 39 years (1980–2018) in the Indus basin, Pakistan. PCA was used to calculate principal components of precipitation, temperature, and streamflow, which were used to systematically propagate drought from one catchment to another, resulting in a catchment-scale drought assessment. The systematic propagation of drought was useful in capturing the effects of local climate variability in the 27 catchments of the Indus basin. Wavelet analyses are used to calculate the variability of SPEI/SRI and propagation (analyzed with the wavelet coherence) from SPEI to SRI. The propagation time from SPEI to SRI was cross correlated. SPEI/SRI time series showed extreme/severe droughts in 16 out of the 39 years, where relatively weak apparent wet and drought events are observed at short periods (1 month) and apparent at longer periods (6 and 12 months). Propagation from SPEI to SRI is catchment specific, with most catchments showing transition in early years (1997–2003). Propagation rate is higher in the upper Indus basin (UIB) and lower Indus basin (LIB) than in the middle Indus basin (MIB), suggesting that climate plays an important role in drought development and propagation. Results also showed a shorter and longer propagation time in the UIB and LIB, respectively. This study has helped us understand the behavior of droughts at catchment scale and will therefore help in the development of drought mitigation plans in Pakistan and similar regions around the world.
Abstract
This study aims to comprehend the propagation of meteorological drought [expressed by the standardized precipitation evapotranspiration index (SPEI)] into hydrological drought [expressed by the standardized runoff index (SRI)] using the combined application of principal component analysis (PCA) and wavelet analysis for a period of 39 years (1980–2018) in the Indus basin, Pakistan. PCA was used to calculate principal components of precipitation, temperature, and streamflow, which were used to systematically propagate drought from one catchment to another, resulting in a catchment-scale drought assessment. The systematic propagation of drought was useful in capturing the effects of local climate variability in the 27 catchments of the Indus basin. Wavelet analyses are used to calculate the variability of SPEI/SRI and propagation (analyzed with the wavelet coherence) from SPEI to SRI. The propagation time from SPEI to SRI was cross correlated. SPEI/SRI time series showed extreme/severe droughts in 16 out of the 39 years, where relatively weak apparent wet and drought events are observed at short periods (1 month) and apparent at longer periods (6 and 12 months). Propagation from SPEI to SRI is catchment specific, with most catchments showing transition in early years (1997–2003). Propagation rate is higher in the upper Indus basin (UIB) and lower Indus basin (LIB) than in the middle Indus basin (MIB), suggesting that climate plays an important role in drought development and propagation. Results also showed a shorter and longer propagation time in the UIB and LIB, respectively. This study has helped us understand the behavior of droughts at catchment scale and will therefore help in the development of drought mitigation plans in Pakistan and similar regions around the world.
A Neural Network–Based Cloud Mask for PREFIRE and Evaluation with Simulated ObservationsAbstract
The Polar Radiant Energy in the Far Infrared Experiment (PREFIRE) will fill a gap in our understanding of polar processes and the polar climate by offering widespread, spectrally resolved measurements through the far-infrared (FIR) with two identical CubeSat spacecraft. While the polar regions are typically difficult for skillful cloud identification due to cold surface temperatures, the reflection by bright surfaces, and frequent temperature inversions, the inclusion of the FIR may offer increased spectral sensitivity, allowing for the detection of even thin ice clouds. This study assesses the potential skill, as well as limitations, of a neural network (NN)-based cloud mask using simulated spectra mimicking what the PREFIRE mission will capture. Analysis focuses on the polar regions. Clouds are found to be detected approximately 90% of time using the derived neural network. The NN’s assigned confidence for whether a scene is “clear” or “cloudy” proves to be a skillful way in which quality flags can be attached to predictions. Clouds with higher cloud-top heights are typically more easily detected. Low-altitude clouds over polar surfaces, which are the most difficult for the NN to detect, are still detected over 80% of the time. The FIR portion of the spectrum is found to increase the detection of clear scenes and increase mid- to high-altitude cloud detection. Cloud detection skill improves through the use of the overlapping fields of view produced by the PREFIRE instrument’s sampling strategy. Overlapping fields of view increase accuracy relative to the baseline NN while simultaneously predicting on a sub-FOV scale.
Significance Statement
Clouds play an important role in defining the Arctic and Antarctic climates. The purpose of this study is to explore the potential of never-before systematically measured radiative properties of the atmosphere to aid in the detection of polar clouds, which are traditionally difficult to detect. Satellite measurements of emitted radiation at wavelengths longer than 15 μm, combined with complex machine learning methods, may allow us to better understand the occurrence of various cloud types at both poles. The occurrence of these clouds can determine whether the surface warms or cools, influencing surface temperatures and the rate at which ice melts or refreezes. Understanding the frequencies of these various clouds is increasingly important within the context of our rapidly changing climate.
Abstract
The Polar Radiant Energy in the Far Infrared Experiment (PREFIRE) will fill a gap in our understanding of polar processes and the polar climate by offering widespread, spectrally resolved measurements through the far-infrared (FIR) with two identical CubeSat spacecraft. While the polar regions are typically difficult for skillful cloud identification due to cold surface temperatures, the reflection by bright surfaces, and frequent temperature inversions, the inclusion of the FIR may offer increased spectral sensitivity, allowing for the detection of even thin ice clouds. This study assesses the potential skill, as well as limitations, of a neural network (NN)-based cloud mask using simulated spectra mimicking what the PREFIRE mission will capture. Analysis focuses on the polar regions. Clouds are found to be detected approximately 90% of time using the derived neural network. The NN’s assigned confidence for whether a scene is “clear” or “cloudy” proves to be a skillful way in which quality flags can be attached to predictions. Clouds with higher cloud-top heights are typically more easily detected. Low-altitude clouds over polar surfaces, which are the most difficult for the NN to detect, are still detected over 80% of the time. The FIR portion of the spectrum is found to increase the detection of clear scenes and increase mid- to high-altitude cloud detection. Cloud detection skill improves through the use of the overlapping fields of view produced by the PREFIRE instrument’s sampling strategy. Overlapping fields of view increase accuracy relative to the baseline NN while simultaneously predicting on a sub-FOV scale.
Significance Statement
Clouds play an important role in defining the Arctic and Antarctic climates. The purpose of this study is to explore the potential of never-before systematically measured radiative properties of the atmosphere to aid in the detection of polar clouds, which are traditionally difficult to detect. Satellite measurements of emitted radiation at wavelengths longer than 15 μm, combined with complex machine learning methods, may allow us to better understand the occurrence of various cloud types at both poles. The occurrence of these clouds can determine whether the surface warms or cools, influencing surface temperatures and the rate at which ice melts or refreezes. Understanding the frequencies of these various clouds is increasingly important within the context of our rapidly changing climate.
A Precise Zenith Hydrostatic Delay Calibration Model in China Based on the Nonlinear Least Square MethodAbstract
Zenith hydrostatic delay (ZHD) is a crucial parameter in Global Navigation Satellite System (GNSS) navigation and positioning and GNSS meteorology. Since the Saastamoinen ZHD model has a larger error in China, it is significant to improve the Saastamoinen ZHD model. This work first estimated the Saastamoinen model using the integrated ZHD as reference values obtained from radiosonde data collected at 73 stations in China from 2012 to 2016. Then, the residuals between the reference values and the Saastamoinen modeled ZHDs were calculated, and the correlations between the residuals and meteorological parameters were explored. The continuous wavelet transform method was used to recognize the annual and semiannual characteristics of the residuals. Because of the nonlinear variation characteristics of residuals, the nonlinear least squares estimation method was introduced to establish an improved ZHD model—China Revised Zenith Hydrostatic Delay (CRZHD)—adapted for China. The accuracy of the CRZHD model was assessed using radiosonde data and International GNSS Service (IGS) data in 2017; the radiosonde data results show that the CRZHD model is superior to the Saastamoinen model with a 69.6% improvement. The three IGS stations with continuous meteorological data present that the BIAS and RMSE are decreased by 2.7 and 1.5 (URUM), 5.9 and 5.3 (BJFS), and 9.6 and 8.8 mm (TCMS), respectively. The performance of the CRZHD model retrieving PWV was discussed using radiosonde data in 2017. It is shown that the CRZHD model retrieving PWV (CRZHD-PWV) outperforms the Saastamoinen model (SAAS-PWV), in which the precision is improved by 44.4%. The BIAS ranged from −1 to 1 mm and RMSE ranged from 0 to 2 mm in CRZHD-PWV account for 89.0% and 95.9%, while SAAS-PWV account for 46.6% and 58.9%.
Significance Statement
Zenith hydrostatic delay (ZHD) is one of the most important parameters in Global Navigation Satellite System (GNSS) navigation and positioning and GNSS meteorology, which can be derived from a precise ZHD model due to its stability. This research established an improved ZHD model for China to obtain accurate ZHD, which is a prerequisite for pinpoint precipitable water vapor (PWV) retrieval. And the PWV value is beneficial to analyze the change in precipitation in some regions, forecast the short-term rainfall, and monitor the climate.
Abstract
Zenith hydrostatic delay (ZHD) is a crucial parameter in Global Navigation Satellite System (GNSS) navigation and positioning and GNSS meteorology. Since the Saastamoinen ZHD model has a larger error in China, it is significant to improve the Saastamoinen ZHD model. This work first estimated the Saastamoinen model using the integrated ZHD as reference values obtained from radiosonde data collected at 73 stations in China from 2012 to 2016. Then, the residuals between the reference values and the Saastamoinen modeled ZHDs were calculated, and the correlations between the residuals and meteorological parameters were explored. The continuous wavelet transform method was used to recognize the annual and semiannual characteristics of the residuals. Because of the nonlinear variation characteristics of residuals, the nonlinear least squares estimation method was introduced to establish an improved ZHD model—China Revised Zenith Hydrostatic Delay (CRZHD)—adapted for China. The accuracy of the CRZHD model was assessed using radiosonde data and International GNSS Service (IGS) data in 2017; the radiosonde data results show that the CRZHD model is superior to the Saastamoinen model with a 69.6% improvement. The three IGS stations with continuous meteorological data present that the BIAS and RMSE are decreased by 2.7 and 1.5 (URUM), 5.9 and 5.3 (BJFS), and 9.6 and 8.8 mm (TCMS), respectively. The performance of the CRZHD model retrieving PWV was discussed using radiosonde data in 2017. It is shown that the CRZHD model retrieving PWV (CRZHD-PWV) outperforms the Saastamoinen model (SAAS-PWV), in which the precision is improved by 44.4%. The BIAS ranged from −1 to 1 mm and RMSE ranged from 0 to 2 mm in CRZHD-PWV account for 89.0% and 95.9%, while SAAS-PWV account for 46.6% and 58.9%.
Significance Statement
Zenith hydrostatic delay (ZHD) is one of the most important parameters in Global Navigation Satellite System (GNSS) navigation and positioning and GNSS meteorology, which can be derived from a precise ZHD model due to its stability. This research established an improved ZHD model for China to obtain accurate ZHD, which is a prerequisite for pinpoint precipitable water vapor (PWV) retrieval. And the PWV value is beneficial to analyze the change in precipitation in some regions, forecast the short-term rainfall, and monitor the climate.
Abstract
This study investigates the mechanisms of low-latitude intraseasonal oscillations affecting regional persistent extreme precipitation events (RPEPEs) over Southwest China (SWC) during rainy seasons. Most of the RPEPEs over SWC are dominated by 7–20-day variability. The RPEPEs over SWC are preconditioned by two different types of 7–20-day Rossby waves with almost opposite phases over the western North Pacific (WNP). The two types of 7–20-day Rossby waves have direct and indirect effects on type-1 and -2 RPEPEs, respectively. For type 1, a coupled 7–20-day low-level anticyclone and suppressed convection originating from the tropical WNP propagate northwestward and cover the region from the South China Sea (SCS) to the Bay of Bengal before the RPEPEs. The anticyclone triggers ascending motion over SWC and transports more moisture to SWC, favoring the SWC RPEPEs. Before the type-2 RPEPEs, a coupled 7–20-day low-level cyclone and enhanced convection propagates from the tropical WNP to the SCS. The enhanced convection over the SCS leads to the westward extension of the western Pacific subtropical high (WPSH) and the eastward shift of the South Asian high (SAH). The variations in the WPSH and the SAH directly cause SWC RPEPEs by inducing ascending motion and transporting moisture. The mechanisms for type-2 RPEPEs tend to work under the background with a strong WPSH. Using a Lagrangian model, we found that both the 7–20-day oscillations and their background atmospheric circulations result in significant differences in moisture sources for the two types of RPEPEs. These findings benefit a better understanding of SWC extreme precipitation events.
Abstract
This study investigates the mechanisms of low-latitude intraseasonal oscillations affecting regional persistent extreme precipitation events (RPEPEs) over Southwest China (SWC) during rainy seasons. Most of the RPEPEs over SWC are dominated by 7–20-day variability. The RPEPEs over SWC are preconditioned by two different types of 7–20-day Rossby waves with almost opposite phases over the western North Pacific (WNP). The two types of 7–20-day Rossby waves have direct and indirect effects on type-1 and -2 RPEPEs, respectively. For type 1, a coupled 7–20-day low-level anticyclone and suppressed convection originating from the tropical WNP propagate northwestward and cover the region from the South China Sea (SCS) to the Bay of Bengal before the RPEPEs. The anticyclone triggers ascending motion over SWC and transports more moisture to SWC, favoring the SWC RPEPEs. Before the type-2 RPEPEs, a coupled 7–20-day low-level cyclone and enhanced convection propagates from the tropical WNP to the SCS. The enhanced convection over the SCS leads to the westward extension of the western Pacific subtropical high (WPSH) and the eastward shift of the South Asian high (SAH). The variations in the WPSH and the SAH directly cause SWC RPEPEs by inducing ascending motion and transporting moisture. The mechanisms for type-2 RPEPEs tend to work under the background with a strong WPSH. Using a Lagrangian model, we found that both the 7–20-day oscillations and their background atmospheric circulations result in significant differences in moisture sources for the two types of RPEPEs. These findings benefit a better understanding of SWC extreme precipitation events.
Abstract
The Turkana Jet in northern Kenya is shown to modulate the climate of southwest Ethiopia’s Omo River Valley using in situ hydrometeorological data, satellite measurements, and atmospheric reanalyses from decadal to diurnal time scales. Temporal statistics from lowland (2.5°–5°N, 35°–38°E) and highland (6°–9°N, 35°–38°E) areas show that 850-hPa westward airflow over Lake Turkana is stronger in March and October but is weakened when western Indian Ocean sea temperatures become warmer than usual at intervals of 2–7 years. A case study on 24 March 2019 reveals how a stronger Turkana Jet induces warming and drying of the Omo Valley. A second case study on 27 September 2018 reveals Hadley cell subsidence over the southern flank of the Turkana Jet. We demonstrate how nocturnal airflow draining off the mountains joins the channelized jet. Satellite and atmospheric reanalyses exhibit realistic diurnal cycles in the east Omo mountains, but some products have incorrect phase and warm bias. Omo Valley soil moisture and runoff exhibit little trend in historical records and model projections; however, unpredictable multiyear wet and dry spells and a growing demand for water are ongoing concerns.
Abstract
The Turkana Jet in northern Kenya is shown to modulate the climate of southwest Ethiopia’s Omo River Valley using in situ hydrometeorological data, satellite measurements, and atmospheric reanalyses from decadal to diurnal time scales. Temporal statistics from lowland (2.5°–5°N, 35°–38°E) and highland (6°–9°N, 35°–38°E) areas show that 850-hPa westward airflow over Lake Turkana is stronger in March and October but is weakened when western Indian Ocean sea temperatures become warmer than usual at intervals of 2–7 years. A case study on 24 March 2019 reveals how a stronger Turkana Jet induces warming and drying of the Omo Valley. A second case study on 27 September 2018 reveals Hadley cell subsidence over the southern flank of the Turkana Jet. We demonstrate how nocturnal airflow draining off the mountains joins the channelized jet. Satellite and atmospheric reanalyses exhibit realistic diurnal cycles in the east Omo mountains, but some products have incorrect phase and warm bias. Omo Valley soil moisture and runoff exhibit little trend in historical records and model projections; however, unpredictable multiyear wet and dry spells and a growing demand for water are ongoing concerns.
Abstract
Wind and turbulence effects on raindrop fall speeds were elucidated using field observations over a 2-yr time period. Motivations for this study include the recent observations of raindrop fall speed deviations from the terminal fall speed predictions (Vt ) based upon laboratory studies and the utilizations of these predictions in various important meteorological and hydrological applications. Fall speed (Vf ) and other characteristics of raindrops were observed using a high-speed optical disdrometer (HOD), and various rainfall and wind characteristics were observed using a 3D ultrasonic anemometer, a laser-type disdrometer, and rain gauges. A total of 26 951 raindrops were observed during 17 different rainfall events, and of these observed raindrops, 18.5% had a subterminal fall speed (i.e., 0.85Vt ≥ Vf ) and 9.5% had a superterminal fall speed (i.e., 1.15Vt ≤ Vf ). Our observations showed that distributions of sub- and superterminal raindrops in the raindrop size spectrum are distinct, and different physical processes are responsible for the occurrence of each. Vertical wind speed, wind shear, and turbulence were identified as the important factors, the latter two being the dominant ones, for the observed fall speed deviations. Turbulence and wind shear had competing effects on raindrop fall. Raindrops of different sizes showed different responses to turbulence, indicating multiscale interactions between raindrop fall and turbulence. With increasing turbulence levels, while the raindrops in the smaller end of the size spectrum showed fall speed enhancements, those in the larger end of the size spectrum showed fall speed reductions. The effect of wind shear was to enhance the raindrop fall speed toward a superterminal fall.
Abstract
Wind and turbulence effects on raindrop fall speeds were elucidated using field observations over a 2-yr time period. Motivations for this study include the recent observations of raindrop fall speed deviations from the terminal fall speed predictions (Vt ) based upon laboratory studies and the utilizations of these predictions in various important meteorological and hydrological applications. Fall speed (Vf ) and other characteristics of raindrops were observed using a high-speed optical disdrometer (HOD), and various rainfall and wind characteristics were observed using a 3D ultrasonic anemometer, a laser-type disdrometer, and rain gauges. A total of 26 951 raindrops were observed during 17 different rainfall events, and of these observed raindrops, 18.5% had a subterminal fall speed (i.e., 0.85Vt ≥ Vf ) and 9.5% had a superterminal fall speed (i.e., 1.15Vt ≤ Vf ). Our observations showed that distributions of sub- and superterminal raindrops in the raindrop size spectrum are distinct, and different physical processes are responsible for the occurrence of each. Vertical wind speed, wind shear, and turbulence were identified as the important factors, the latter two being the dominant ones, for the observed fall speed deviations. Turbulence and wind shear had competing effects on raindrop fall. Raindrops of different sizes showed different responses to turbulence, indicating multiscale interactions between raindrop fall and turbulence. With increasing turbulence levels, while the raindrops in the smaller end of the size spectrum showed fall speed enhancements, those in the larger end of the size spectrum showed fall speed reductions. The effect of wind shear was to enhance the raindrop fall speed toward a superterminal fall.
Abstract
Current climate models project that Antarctic sea ice will decrease by the end of 21st century. Previous studies have suggested that Antarctic sea ice change have impacts on atmospheric circulation and the mean state of the Southern Hemisphere. However, little is known whether Antarctic sea ice loss may have a tangible impact on climate extremes over southern continents and whether ocean-atmosphere coupling plays an important role in changes of climate extremes over southern continents. In this study, we conduct a set of fully coupled and atmosphere-only model experiments forced by present and future Antarctic sea ice cover. It is found that the projected Antarctic sea ice loss by the end of 21st century leads to increase in the frequency and duration of warm extremes (especially warm nights) over southern continents, and decrease in cold extremes over most regions. The frequency and duration of wet extremes are projected to increase over South America and Antarctica, whereas changes in dry days and longest dry spell vary with regions. Further Antarctic sea ice loss under a quadrupling of CO2 leads to similar but larger changes. Comparison between the coupled and atmosphere-only model experiments suggests that ocean dynamics and their interactions with the atmosphere induced by Antarctic sea ice loss plays a key role in driving the identified changes in temperature and precipitation extremes over southern continents. By comparing with global warming experiments, we find that Antarctic sea ice loss may affect temperature and precipitation extremes for some regions under greenhouse warming, especially for Antarctica.
Abstract
Current climate models project that Antarctic sea ice will decrease by the end of 21st century. Previous studies have suggested that Antarctic sea ice change have impacts on atmospheric circulation and the mean state of the Southern Hemisphere. However, little is known whether Antarctic sea ice loss may have a tangible impact on climate extremes over southern continents and whether ocean-atmosphere coupling plays an important role in changes of climate extremes over southern continents. In this study, we conduct a set of fully coupled and atmosphere-only model experiments forced by present and future Antarctic sea ice cover. It is found that the projected Antarctic sea ice loss by the end of 21st century leads to increase in the frequency and duration of warm extremes (especially warm nights) over southern continents, and decrease in cold extremes over most regions. The frequency and duration of wet extremes are projected to increase over South America and Antarctica, whereas changes in dry days and longest dry spell vary with regions. Further Antarctic sea ice loss under a quadrupling of CO2 leads to similar but larger changes. Comparison between the coupled and atmosphere-only model experiments suggests that ocean dynamics and their interactions with the atmosphere induced by Antarctic sea ice loss plays a key role in driving the identified changes in temperature and precipitation extremes over southern continents. By comparing with global warming experiments, we find that Antarctic sea ice loss may affect temperature and precipitation extremes for some regions under greenhouse warming, especially for Antarctica.
Abstract
Which processes control the mean amounts of precipitation received by tropical land and ocean? Do large-scale constraints exist on the ratio between the two? We address these questions using a conceptual box model based on water balance equations. With empirical but physically motivated parametrizations of the water balance components, we construct a set of coupled differential equations which describe the dynamical behavior of the water vapor content over land and ocean as well as the land’s soil moisture content. For a closed model configuration with one ocean and one land box, we compute equilibrium solutions across the parameter space and analyze their sensitivity to parameter choices. The precipitation ratio χ, defined as the ratio between mean land and ocean precipitation rates, quantifies the land-sea precipitation contrast. We find that χ is bounded between zero and one as long as the presence of land does not affect the relationship between water vapor path and precipitation. However, for the tested parameter values, 95% of the obtained χ values are even larger than 0.75. The sensitivity analysis reveals that χ is primarily controlled by the efficiency of atmospheric moisture transport rather than by land surface parameters. We further investigate under which conditions precipitation enhancement over land (χ > 1) would be possible. An open model configuration with an island between two ocean boxes and nonzero external advection into the domain can yield χ values larger than one, but only for a small subset of parameter choices, characterized by small land fractions and a sufficiently large moisture influx through the windward boundary.
Abstract
Which processes control the mean amounts of precipitation received by tropical land and ocean? Do large-scale constraints exist on the ratio between the two? We address these questions using a conceptual box model based on water balance equations. With empirical but physically motivated parametrizations of the water balance components, we construct a set of coupled differential equations which describe the dynamical behavior of the water vapor content over land and ocean as well as the land’s soil moisture content. For a closed model configuration with one ocean and one land box, we compute equilibrium solutions across the parameter space and analyze their sensitivity to parameter choices. The precipitation ratio χ, defined as the ratio between mean land and ocean precipitation rates, quantifies the land-sea precipitation contrast. We find that χ is bounded between zero and one as long as the presence of land does not affect the relationship between water vapor path and precipitation. However, for the tested parameter values, 95% of the obtained χ values are even larger than 0.75. The sensitivity analysis reveals that χ is primarily controlled by the efficiency of atmospheric moisture transport rather than by land surface parameters. We further investigate under which conditions precipitation enhancement over land (χ > 1) would be possible. An open model configuration with an island between two ocean boxes and nonzero external advection into the domain can yield χ values larger than one, but only for a small subset of parameter choices, characterized by small land fractions and a sufficiently large moisture influx through the windward boundary.
