2019 Vol. 24, No. 1
Display Method:
2019, 24(1): 1-21.
doi: 10.3878/j.issn.1006-9585.2018.18088
Abstract:
As an important background of month-seasonal and interannual climate variations, interdecadal climate variation often affects climate features with interannual and month-seasonal time scales. Along with the development and progress of science and the rise of social requirement, interdecadal climate variability has become an important issue that has attracted more attentions. As one of important contents on climate dynamics and climate foreshadow, research has been launched vigorously in the world. Some research achievements have been published. In this paper, we will focus on systematic and comprehensive discussion on possible mechanisms of interdecadal climate variability. The major contents include:Influences of main patterns of ocean temperature in the global; influences of interdecadal variation of climate system relationship; influences of interdecadal variation of the atmospheric system on the planetary scale; and impacts of solar activities and volcano eruptions. As we know, studies on interdecadal climate variability are important, but the dynamic mechanism of interdecadal climate variability is so complicated. There are more problems that still remain unsolved and need further in-depth study. We believe that further in-depth research achievements will be able to provide reliable scientific basis for the foreshadow of interdecadal climate variation, promote professional work of interdecadal climate variation forecast and improve the forecasting capability.
As an important background of month-seasonal and interannual climate variations, interdecadal climate variation often affects climate features with interannual and month-seasonal time scales. Along with the development and progress of science and the rise of social requirement, interdecadal climate variability has become an important issue that has attracted more attentions. As one of important contents on climate dynamics and climate foreshadow, research has been launched vigorously in the world. Some research achievements have been published. In this paper, we will focus on systematic and comprehensive discussion on possible mechanisms of interdecadal climate variability. The major contents include:Influences of main patterns of ocean temperature in the global; influences of interdecadal variation of climate system relationship; influences of interdecadal variation of the atmospheric system on the planetary scale; and impacts of solar activities and volcano eruptions. As we know, studies on interdecadal climate variability are important, but the dynamic mechanism of interdecadal climate variability is so complicated. There are more problems that still remain unsolved and need further in-depth study. We believe that further in-depth research achievements will be able to provide reliable scientific basis for the foreshadow of interdecadal climate variation, promote professional work of interdecadal climate variation forecast and improve the forecasting capability.
2019, 24(1): 22-36.
doi: 10.3878/j.issn.1006-9585.2018.18027
Abstract:
Urban ventilation corridors can increase urban air liquidity capacity and relieve the urban heat island. In order to quantitatively evaluate the meteorological effects of urban ventilation corridors, the regional boundary layer chemistry model (RBLM-Chem) was used to analyze meteorological effects of the ventilation corridor under different weather types based on the dataset of high resolution surface types and urban buildings in Hangzhou. The horizontal resolution of the model is 250 m. The results indicate that the urban ventilation corridor can increase wind speed, decrease temperature, and increase humidity. Compared with the situation that the ventilation corridor is absent, wind speeds at the surface and 60-m height in the corridor area increase by 1.4 m/s and 1 m/s, respectively. In the summer, temperature in the corridor area can decrease by 2.7℃ due to the cooling effect of the ventilation corridor. This value is much higher than that in winter, which is only about 0.6℃. At 250-m downstream of the ventilation corridor, the maximum values of the increments of wind speed, temperature, and relative humidity are 1.5 m/s, -2.9℃ and 3.1%, respectively. Even at 1500-m downstream of the ventilation corridor, the maximum temperature increments can still reach up to -1.2℃.
Urban ventilation corridors can increase urban air liquidity capacity and relieve the urban heat island. In order to quantitatively evaluate the meteorological effects of urban ventilation corridors, the regional boundary layer chemistry model (RBLM-Chem) was used to analyze meteorological effects of the ventilation corridor under different weather types based on the dataset of high resolution surface types and urban buildings in Hangzhou. The horizontal resolution of the model is 250 m. The results indicate that the urban ventilation corridor can increase wind speed, decrease temperature, and increase humidity. Compared with the situation that the ventilation corridor is absent, wind speeds at the surface and 60-m height in the corridor area increase by 1.4 m/s and 1 m/s, respectively. In the summer, temperature in the corridor area can decrease by 2.7℃ due to the cooling effect of the ventilation corridor. This value is much higher than that in winter, which is only about 0.6℃. At 250-m downstream of the ventilation corridor, the maximum values of the increments of wind speed, temperature, and relative humidity are 1.5 m/s, -2.9℃ and 3.1%, respectively. Even at 1500-m downstream of the ventilation corridor, the maximum temperature increments can still reach up to -1.2℃.
2019, 24(1): 37-49.
doi: 10.3878/j.issn.1006-9585.2018.17166
Abstract:
The influence of sensible and latent heat anomalies over the Tibetan Plateau (TP) on a persistent rainfall in western Sichuan Basin is investigated via a semi-idealized mesoscale numerical model WRF (Weather Research and Forecasting) simulation. Analyses of the simulation show that when the heating on the TP is turned off, rainfall decreases in the TP and western Sichuan Basin but increases in central and eastern Sichuan Basin, while the diurnal variation of precipitation weakens; the trough over the TP on 500 hPa disappears, the strength and scope of the westerly trough reduces slightly, but the meso-cyclone at low levels in eastern Sichuan Basin enhances; quantitative analysis of vorticity budget during the heavy rain period indicates that the intensification of vertical wind shear in the lower troposphere makes positive contribution to strengthening the TILT (tilting term) in eastern Sichuan Basin, and thus facilitates vortex development in the key area of Sichuan Basin, which also leads to increases in precipitation in eastern Sichuan Basin.
The influence of sensible and latent heat anomalies over the Tibetan Plateau (TP) on a persistent rainfall in western Sichuan Basin is investigated via a semi-idealized mesoscale numerical model WRF (Weather Research and Forecasting) simulation. Analyses of the simulation show that when the heating on the TP is turned off, rainfall decreases in the TP and western Sichuan Basin but increases in central and eastern Sichuan Basin, while the diurnal variation of precipitation weakens; the trough over the TP on 500 hPa disappears, the strength and scope of the westerly trough reduces slightly, but the meso-cyclone at low levels in eastern Sichuan Basin enhances; quantitative analysis of vorticity budget during the heavy rain period indicates that the intensification of vertical wind shear in the lower troposphere makes positive contribution to strengthening the TILT (tilting term) in eastern Sichuan Basin, and thus facilitates vortex development in the key area of Sichuan Basin, which also leads to increases in precipitation in eastern Sichuan Basin.
2019, 24(1): 50-60.
doi: 10.3878/j.issn.1006-9585.2017.17072
Abstract:
Based on monthly precipitation observations collected at 93 meteorological stations in Southwest China from 1996 to 2000, this study investigates the spatial interpolation results with the Inverse Distance Weighting (IDW) and O-Kriging interpolation methods. Firstly, we analyze the spatial autocorrelation and spatial variability character of monthly average precipitation data. Secondly, the IDW and O-Kriging based on three semi-variograms (exponential, spherical and, Gaussian model) are used to spatially interpolate monthly precipitation. Finally, the interpolation results are compared and discussed using the cross-validation method. The conclusions are:(1) Monthly precipitation distribution in Southwest China shows a spatial aggregation feature with high spatial autocorrelation and variation, which favors for the spatial interpolation. (2) Compared to the three semi-variograms used in the O-Kriging interpolation method, the best performance is from the exponential model, while the worst is from the Gaussian model. (3) When the O-Kriging and IDW are used in spatial interpolation of monthly average and maximum and minimum precipitation, the former one perform better than the latter one. The errors between interpolated data and observations overall increase with monthly precipitation magnitude, and the errors from both interpolation methods are obviously reduced after removing the maximum monthly precipitation points. (4) For the study area as a whole, the interpolation effect of the O-Kriging is better than that of IDW, however, this is not true at single sites. There is no absolute optimal method in the spatial interpolation of precipitation for every study area and on all time scales. The optimal interpolation method depends on the actual demands and applications.
Based on monthly precipitation observations collected at 93 meteorological stations in Southwest China from 1996 to 2000, this study investigates the spatial interpolation results with the Inverse Distance Weighting (IDW) and O-Kriging interpolation methods. Firstly, we analyze the spatial autocorrelation and spatial variability character of monthly average precipitation data. Secondly, the IDW and O-Kriging based on three semi-variograms (exponential, spherical and, Gaussian model) are used to spatially interpolate monthly precipitation. Finally, the interpolation results are compared and discussed using the cross-validation method. The conclusions are:(1) Monthly precipitation distribution in Southwest China shows a spatial aggregation feature with high spatial autocorrelation and variation, which favors for the spatial interpolation. (2) Compared to the three semi-variograms used in the O-Kriging interpolation method, the best performance is from the exponential model, while the worst is from the Gaussian model. (3) When the O-Kriging and IDW are used in spatial interpolation of monthly average and maximum and minimum precipitation, the former one perform better than the latter one. The errors between interpolated data and observations overall increase with monthly precipitation magnitude, and the errors from both interpolation methods are obviously reduced after removing the maximum monthly precipitation points. (4) For the study area as a whole, the interpolation effect of the O-Kriging is better than that of IDW, however, this is not true at single sites. There is no absolute optimal method in the spatial interpolation of precipitation for every study area and on all time scales. The optimal interpolation method depends on the actual demands and applications.
2019, 24(1): 61-72.
doi: 10.3878/j.issn.1006-9585.2018.18057
Abstract:
Based on hourly PM2.5 monitoring data from Beijing Environmental Protection Monitoring Center and the U.S. Embassy, the 325 m gradient tower data from the Institute of Atmospheric Physics, Chinese Academy of Sciences, the synoptic charts and sounding data, the atmospheric boundary layer characteristics during the heavy PM2.5pollution period from 27 Nov to 1 Dec 2015 are analyzed. The results show that this heavy pollution process was persistent and severe, since the duration of ρ(PM2.5) exceeding 75 μg/m3 was 126 hours in total and the duration of ρ(PM2.5) exceeding 150 μg/m3 was 116 hours in total. The maximum hourly ρ(PM2.5) was up to 522 μg/m3. Under the influence of the weather situation, light winds prevailed in the near-surface layer with multi-layer inverse temperature structure, which inhibited both the horizontal and vertical transport and dispersion of pollutants. In addition, a thick wet layer developed in the boundary layer, in which the aerosols kept absorbing moisture and grew up. As a result, high ρ(PM2.5) concentration maintained during the pollution process. The turbulent kinetic energy was relatively small during the heavy pollution process, which was not conducive to the dispersion of pollutants. Note that horizontal turbulent kinetic energy accounted for the major part of the total turbulent kinetic energy, and the turbulent kinetic energy in the vertical direction was only about 15%-20% of that in the horizontal direction. Friction velocities at different heights exhibited the same characteristics as turbulent kinetic energy. The occurrence of two turbulence intensity spikes was a sign of turbulent flow adjustment and a precursor to the sharp shift in PM2.5 concentration, and the air quality would become worse. During the process of heavy pollution, the sensible heat flux was transported from the ground to the atmosphere; both sensible and latent heat fluxes significantly reduced compared with that in the non-polluting moment and exhibited distinct diurnal changes. Power spectral analysis and calculations show that during the heavy pollution process, mesoscale processes on time scales from 5 min to 6 h made important contributions to the transfer of momentum and heat fluxes from the surface to the atmosphere.
Based on hourly PM2.5 monitoring data from Beijing Environmental Protection Monitoring Center and the U.S. Embassy, the 325 m gradient tower data from the Institute of Atmospheric Physics, Chinese Academy of Sciences, the synoptic charts and sounding data, the atmospheric boundary layer characteristics during the heavy PM2.5pollution period from 27 Nov to 1 Dec 2015 are analyzed. The results show that this heavy pollution process was persistent and severe, since the duration of ρ(PM2.5) exceeding 75 μg/m3 was 126 hours in total and the duration of ρ(PM2.5) exceeding 150 μg/m3 was 116 hours in total. The maximum hourly ρ(PM2.5) was up to 522 μg/m3. Under the influence of the weather situation, light winds prevailed in the near-surface layer with multi-layer inverse temperature structure, which inhibited both the horizontal and vertical transport and dispersion of pollutants. In addition, a thick wet layer developed in the boundary layer, in which the aerosols kept absorbing moisture and grew up. As a result, high ρ(PM2.5) concentration maintained during the pollution process. The turbulent kinetic energy was relatively small during the heavy pollution process, which was not conducive to the dispersion of pollutants. Note that horizontal turbulent kinetic energy accounted for the major part of the total turbulent kinetic energy, and the turbulent kinetic energy in the vertical direction was only about 15%-20% of that in the horizontal direction. Friction velocities at different heights exhibited the same characteristics as turbulent kinetic energy. The occurrence of two turbulence intensity spikes was a sign of turbulent flow adjustment and a precursor to the sharp shift in PM2.5 concentration, and the air quality would become worse. During the process of heavy pollution, the sensible heat flux was transported from the ground to the atmosphere; both sensible and latent heat fluxes significantly reduced compared with that in the non-polluting moment and exhibited distinct diurnal changes. Power spectral analysis and calculations show that during the heavy pollution process, mesoscale processes on time scales from 5 min to 6 h made important contributions to the transfer of momentum and heat fluxes from the surface to the atmosphere.
2019, 24(1): 73-85.
doi: 10.3878/j.issn.1006-9585.2018.18043
Abstract:
Sichuan Basin is located to the east of the Tibet Plateau. Due to its geographical location, weather and climate in this area are complicated and capricious. In particular, the rainstorm forecasting is one of the toughest problems that meteorologists need to solve. In order to understand the rainstorm mechanism in Sichuan Basin, the authors used the ERA-Interim reanalysis data, a gridded precipitation dataset and several conventional upper-air sounding data to analyze the environmental fields of two rainstorms that occurred in the summer of 2015 (the '7.13' process during 13-15 July and the '8.16' process during 16-18 August) from the following perspectives:The circulation background, the moisture condition, and the dynamic and thermal conditions in Sichuan Basin. Results show that:1) The relatively stable large-scale circulation provides a favorable background for the occurrence of the two heavy rainstorms. 2) There existed upper-level jet stream and low-level jet stream in both processes, and the upper-level jet stream in the '8.16' process was much stronger than that in the '7.13' process. This is one of the reasons why there was significant difference in the precipitation intensity between the two processes. For the '7.13' process, the low-level jet stream flowing northward transported water vapor from the Bay of Bengal to southern Sichuan. During the '8.16' process, the water vapor transported from the Bay of Bengal by the low-level jet stream was affected by the southwestern vortex over Sichuan and Chongqing, and then was transported to the rainy area by cyclonic circulation. 3) The atmosphere was unstable during the two rainstorm processes. And θse was a good indicator of the rain intensity and falling area for the two rainstorms. The potential vorticity disturbance propagated downward to lower levels, and its increase indicated the occurrence of heavy rain.
Sichuan Basin is located to the east of the Tibet Plateau. Due to its geographical location, weather and climate in this area are complicated and capricious. In particular, the rainstorm forecasting is one of the toughest problems that meteorologists need to solve. In order to understand the rainstorm mechanism in Sichuan Basin, the authors used the ERA-Interim reanalysis data, a gridded precipitation dataset and several conventional upper-air sounding data to analyze the environmental fields of two rainstorms that occurred in the summer of 2015 (the '7.13' process during 13-15 July and the '8.16' process during 16-18 August) from the following perspectives:The circulation background, the moisture condition, and the dynamic and thermal conditions in Sichuan Basin. Results show that:1) The relatively stable large-scale circulation provides a favorable background for the occurrence of the two heavy rainstorms. 2) There existed upper-level jet stream and low-level jet stream in both processes, and the upper-level jet stream in the '8.16' process was much stronger than that in the '7.13' process. This is one of the reasons why there was significant difference in the precipitation intensity between the two processes. For the '7.13' process, the low-level jet stream flowing northward transported water vapor from the Bay of Bengal to southern Sichuan. During the '8.16' process, the water vapor transported from the Bay of Bengal by the low-level jet stream was affected by the southwestern vortex over Sichuan and Chongqing, and then was transported to the rainy area by cyclonic circulation. 3) The atmosphere was unstable during the two rainstorm processes. And θse was a good indicator of the rain intensity and falling area for the two rainstorms. The potential vorticity disturbance propagated downward to lower levels, and its increase indicated the occurrence of heavy rain.
2019, 24(1): 86-104.
doi: 10.3878/j.issn.1006-9585.2018.17169
Abstract:
Based on the Weather Research and Forecasting (WRF) model driven by the global model IPSL-CM5A-LR results that are included in the model output archive of the Coupled Model Intercomparison Project Phase 5, this study has assessed the model ability for simulating extreme precipitation indices and analyzed possible future changes in the mid-21st century (2041-2060) under the RCP8.5 scenario over East China. Results indicate that WRF performs well in the simulation of extreme precipitation indices. Compared with IPSL-CM5A-LR model, WRF model can better reproduce the spatial distribution and annual cycle of precipitation over East China and the sub-regions. In particular, the simulation of regional features is improved in WRF and the problem in global model to overestimate light precipitation has been overcome. Prediction results show that East China will experience an obvious trend of extremeness on precipitation. WRF simulation results show that indices of annual total wet-day precipitation (PRCPTOT), number of heavy precipitation days (R10mm), and simple daily intensity index (SDⅡ) indices in most regions of East China will increase by more than 20%, the increases of extreme wet days (R95d) and max 5-d precipitation (Rx5day) indices in the northern part of East China will be more than 50% and 35%, and consecutive dry days (CDD) overall will increased in East China. Model grids with significant changes are mainly located in areas with large increases. There will be an extremalization in precipitation with increases in both strong precipitation and drought events, and the degree of extremalization is stronger in the north than in the south of East China.
Based on the Weather Research and Forecasting (WRF) model driven by the global model IPSL-CM5A-LR results that are included in the model output archive of the Coupled Model Intercomparison Project Phase 5, this study has assessed the model ability for simulating extreme precipitation indices and analyzed possible future changes in the mid-21st century (2041-2060) under the RCP8.5 scenario over East China. Results indicate that WRF performs well in the simulation of extreme precipitation indices. Compared with IPSL-CM5A-LR model, WRF model can better reproduce the spatial distribution and annual cycle of precipitation over East China and the sub-regions. In particular, the simulation of regional features is improved in WRF and the problem in global model to overestimate light precipitation has been overcome. Prediction results show that East China will experience an obvious trend of extremeness on precipitation. WRF simulation results show that indices of annual total wet-day precipitation (PRCPTOT), number of heavy precipitation days (R10mm), and simple daily intensity index (SDⅡ) indices in most regions of East China will increase by more than 20%, the increases of extreme wet days (R95d) and max 5-d precipitation (Rx5day) indices in the northern part of East China will be more than 50% and 35%, and consecutive dry days (CDD) overall will increased in East China. Model grids with significant changes are mainly located in areas with large increases. There will be an extremalization in precipitation with increases in both strong precipitation and drought events, and the degree of extremalization is stronger in the north than in the south of East China.
2019, 24(1): 105-115.
doi: 10.3878/j.issn.1006-9585.2018.17116
Abstract:
In order to study long-term precipitation characteristics of the Tibetan Plateau vortex, statistical analysis of precipitation characteristics associated with the Tibetan Plateau vortex over the past 37 years has been conducted based on the dataset of the Tibetan Plateau vortex from 1979 to 2015. Precipitation areas as well as daily precipitation amount observed at weather stations in the Tibetan Plateau are also used. The results show that annual vortex precipitation exhibits an upward trend. The precipitation center is located at Naqu of the Tibetan Plateau, and decreases eastward. The precipitation is largely comprised of summer low vortex precipitation, while both summer precipitation and annual precipitation have a significant positive correlation with total precipitation. The vortex precipitation is declining at Anduo station, whereas its average contribution to annual total is as high as 33%. The vortex precipitation generally shows an upward trend at Naqu and Toto River stations with the increasing rates of 0.21 mm/a and 0.65 mm/a, respectively. The correlation coefficient between low vortex frequency and low vortex precipitation is 0.66 at Naqu area however, the percentage of precipitation accounted for by the vortex precipitation in the Toto River shows a decreasing trend. The low vortex precipitation in the Tibetan Plateau is dominated by light rainfall, while moderate rainfall is the main contributor to total vortex rainfall. The trend analysis shows that the largest increase center of the Tibetan Plateau vortex precipitation is located in northern Qinghai, where the increasing rate reaches 0.88 mm/a, and the second largest center is located along the Yarlung Zangbo River in southwestern Tibet. At the same time, light rainfall induced by the low vortex basically increases in the whole region with the center located in northeastern Tibet and southwestern Qinghai. The upward trend of moderate rainfall is mainly found in southwestern Tibet, Qinghai region and western Sichuan. An obvious rising center is located in southern Qinghai with an increasing rate of up to 0.67mm/a, and the downward trend is mainly distributed in northeastern Tibet.
In order to study long-term precipitation characteristics of the Tibetan Plateau vortex, statistical analysis of precipitation characteristics associated with the Tibetan Plateau vortex over the past 37 years has been conducted based on the dataset of the Tibetan Plateau vortex from 1979 to 2015. Precipitation areas as well as daily precipitation amount observed at weather stations in the Tibetan Plateau are also used. The results show that annual vortex precipitation exhibits an upward trend. The precipitation center is located at Naqu of the Tibetan Plateau, and decreases eastward. The precipitation is largely comprised of summer low vortex precipitation, while both summer precipitation and annual precipitation have a significant positive correlation with total precipitation. The vortex precipitation is declining at Anduo station, whereas its average contribution to annual total is as high as 33%. The vortex precipitation generally shows an upward trend at Naqu and Toto River stations with the increasing rates of 0.21 mm/a and 0.65 mm/a, respectively. The correlation coefficient between low vortex frequency and low vortex precipitation is 0.66 at Naqu area however, the percentage of precipitation accounted for by the vortex precipitation in the Toto River shows a decreasing trend. The low vortex precipitation in the Tibetan Plateau is dominated by light rainfall, while moderate rainfall is the main contributor to total vortex rainfall. The trend analysis shows that the largest increase center of the Tibetan Plateau vortex precipitation is located in northern Qinghai, where the increasing rate reaches 0.88 mm/a, and the second largest center is located along the Yarlung Zangbo River in southwestern Tibet. At the same time, light rainfall induced by the low vortex basically increases in the whole region with the center located in northeastern Tibet and southwestern Qinghai. The upward trend of moderate rainfall is mainly found in southwestern Tibet, Qinghai region and western Sichuan. An obvious rising center is located in southern Qinghai with an increasing rate of up to 0.67mm/a, and the downward trend is mainly distributed in northeastern Tibet.
2019, 24(1): 116-124.
doi: 10.3878/j.issn.1006-9585.2018.18049
Abstract:
Post-forecast data processing is critical for obtaining reliable local weather forecast. In this study, the authors developed three post-processing models based on ridge regression (Ridge), random forest (RF), and deep learning (DL) methods. The post-processing models were trained by observational and forecast data of daily 2-m above surface air temperature in North China (38°N-43°N, 113°E-119°E) from four numerical weather forecast (NWF) models (BABJ model from China Meteorological Administration, ECMF model from ECMWF, RJTD model from Japan Meteorological Agency, and KWBC model from NCEP, respectively), for the training period from February 2014 to September 2016, and then applied to the forecast period from October 2016 to September 2017. The forecast results of the post-processing models together with those of commonly-used multi-model ensemble mean (EMN) and individual NWF models were evaluated according to the root-mean-square error (RMSE). The main results are as follows:1) For the region as a whole, with the increase in the forecast lead time, all the NWF models, EMN and the post-processing models exhibit increasing RMSEs, but the RMSEs of the three post-processing models are all significantly smaller than those of EMN and individual NWF models; especially, DL is slightly better for the short-term (the first few days) forecast and RF is slightly better for the longer-term prediction. 2) The RMSEs are relatively smaller in the southeastern part of North China, approximately in the range of (38°N-41°N, 115.5°E-119°E) than else where; on average, DL is slightly better, and the RMSEs of the three machine learning models are between 1.24℃ and 1.26℃, while the EMN error is 1.69℃. 3) There are seasonal differences:The results of all the models are relatively good for the summer, but poor in general for the winter. All the three post-processing models perform better than EMN and individual NWF models, with a smallest average RMSE of 2.15℃ for Ridge compared with 2.45℃ for EMN.
Post-forecast data processing is critical for obtaining reliable local weather forecast. In this study, the authors developed three post-processing models based on ridge regression (Ridge), random forest (RF), and deep learning (DL) methods. The post-processing models were trained by observational and forecast data of daily 2-m above surface air temperature in North China (38°N-43°N, 113°E-119°E) from four numerical weather forecast (NWF) models (BABJ model from China Meteorological Administration, ECMF model from ECMWF, RJTD model from Japan Meteorological Agency, and KWBC model from NCEP, respectively), for the training period from February 2014 to September 2016, and then applied to the forecast period from October 2016 to September 2017. The forecast results of the post-processing models together with those of commonly-used multi-model ensemble mean (EMN) and individual NWF models were evaluated according to the root-mean-square error (RMSE). The main results are as follows:1) For the region as a whole, with the increase in the forecast lead time, all the NWF models, EMN and the post-processing models exhibit increasing RMSEs, but the RMSEs of the three post-processing models are all significantly smaller than those of EMN and individual NWF models; especially, DL is slightly better for the short-term (the first few days) forecast and RF is slightly better for the longer-term prediction. 2) The RMSEs are relatively smaller in the southeastern part of North China, approximately in the range of (38°N-41°N, 115.5°E-119°E) than else where; on average, DL is slightly better, and the RMSEs of the three machine learning models are between 1.24℃ and 1.26℃, while the EMN error is 1.69℃. 3) There are seasonal differences:The results of all the models are relatively good for the summer, but poor in general for the winter. All the three post-processing models perform better than EMN and individual NWF models, with a smallest average RMSE of 2.15℃ for Ridge compared with 2.45℃ for EMN.
2019, 24(1): 125-134.
doi: 10.3878/j.issn.1006-9585.2018.17060
Abstract:
In order to study the regularity of short-term temperature fluctuation in the context of global warming, daily temperature data collected at the 804 meteorological stations in China are used to count the number of days in which daily average temperature fluctuation is greater than 1℃ and calculate the standard deviation of daily temperature fluctuation. The temporal and spatial distribution and variation of short-term temperature fluctuation days and fluctuation amplitude are analyzed, and the relationships between springtime mean temperature and latitude/longitude/altitude are discussed. The results are as follows. The short-term temperature fluctuation over China in the past six decades can be roughly divided into two stages, i.e., the stage before 1990 when the short-term temperature fluctuation amplitude and fluctuation days did not change in most areas of China, and the stage after 1990 when the short-term temperature fluctuation in most areas of China (about 70%) showed an upward trend. The short-term temperature fluctuation is closely related to latitude and longitude. Combined with regional distribution of the fluctuation trend and fluctuations of daily temperature, it can be seen that the fluctuation amplitude and fluctuation days both increase in the low latitudes, which makes people get a strong feeling about the temperature change. Because the process of biological growth and development is closely related to temperature, results of the present paper suggest that short-term temperature fluctuation may also need to be considered when discussing long-term impacts of climate change on ecosystems.
In order to study the regularity of short-term temperature fluctuation in the context of global warming, daily temperature data collected at the 804 meteorological stations in China are used to count the number of days in which daily average temperature fluctuation is greater than 1℃ and calculate the standard deviation of daily temperature fluctuation. The temporal and spatial distribution and variation of short-term temperature fluctuation days and fluctuation amplitude are analyzed, and the relationships between springtime mean temperature and latitude/longitude/altitude are discussed. The results are as follows. The short-term temperature fluctuation over China in the past six decades can be roughly divided into two stages, i.e., the stage before 1990 when the short-term temperature fluctuation amplitude and fluctuation days did not change in most areas of China, and the stage after 1990 when the short-term temperature fluctuation in most areas of China (about 70%) showed an upward trend. The short-term temperature fluctuation is closely related to latitude and longitude. Combined with regional distribution of the fluctuation trend and fluctuations of daily temperature, it can be seen that the fluctuation amplitude and fluctuation days both increase in the low latitudes, which makes people get a strong feeling about the temperature change. Because the process of biological growth and development is closely related to temperature, results of the present paper suggest that short-term temperature fluctuation may also need to be considered when discussing long-term impacts of climate change on ecosystems.
2019, 24(1): 135-142.
doi: 10.3878/j.issn.1006-9585.2018.17126
Abstract:
Based on daily maximum electric loads and meteorological data in the summer (June-August) from 2006 to 2017 in Beijing, the relationship between electric load and meteorological factors is diagnosed. Using the BP (Back Propagation) neural network algorithm, two maximum electric power load prediction models are established and evaluated. The results indicate that (1) the basic electric load on weekends in Beijing in the summer is much less than that in working days, which should be distinguished when being removed; (2) the influence of meteorological factors on meteorological load has cumulative effect, and the correlation between them is the highest for two days of accumulation; (3) taking the actual situation into account, two different daily maximum electric load forecasting models are established based on different independent variables. Comparing the prediction results with actual data, both of the forecasting models show good prediction performance that can meet the actual demand of the power sector. The forecasting model with meteorological load of the previous day as an independent variable shows better prediction effect.
Based on daily maximum electric loads and meteorological data in the summer (June-August) from 2006 to 2017 in Beijing, the relationship between electric load and meteorological factors is diagnosed. Using the BP (Back Propagation) neural network algorithm, two maximum electric power load prediction models are established and evaluated. The results indicate that (1) the basic electric load on weekends in Beijing in the summer is much less than that in working days, which should be distinguished when being removed; (2) the influence of meteorological factors on meteorological load has cumulative effect, and the correlation between them is the highest for two days of accumulation; (3) taking the actual situation into account, two different daily maximum electric load forecasting models are established based on different independent variables. Comparing the prediction results with actual data, both of the forecasting models show good prediction performance that can meet the actual demand of the power sector. The forecasting model with meteorological load of the previous day as an independent variable shows better prediction effect.
