The drag coefficient （Ca） and heat transfer coefficient （Ch） with the bulk transfer scheme are usually used to calculate the momentum and heat fluxes in meteorological models. The aerodynamic roughness length （z0...The drag coefficient （Ca） and heat transfer coefficient （Ch） with the bulk transfer scheme are usually used to calculate the momentum and heat fluxes in meteorological models. The aerodynamic roughness length （z0m） and thermal roughness length （z0h） are two crucial parameters for bulk transfer equations. To improve the meteorological models, the seasonal and interannual variations of Z0m, Z0h, coefficient kB-1, Cd, and Ch were investigated based on eddy covariance data over different grazed semiarid grasslands of Inner Mongolia during the growing seasons （May to September） from 2005 to 2008. For an ungrazed Leymus chinensis grassland （ungrazed since 1979）, Z0m and z0h had significant seasonal and in- terannual variations. Zorn was affected by the amount and distribution of rainfall, kB 1 exhibited a relatively negative variation compared with z0h, which indicates that the seasonal variation of z0h cannot be described by kB 1. To parameterize Zorn and z0h, the linear regressions between ln（z0m）, ln（z0h）, and the leaf area index （LAI） were performed with R2=0.71 and 0.83. The monthly average kB-1 was found to decrease linearly with LAI. The four-year averaged values of Ca and Ch were 4.5 × 10^-3 and 3.9× 10^-3, respectively. The monthly average Cd only varied by 8% while the variation of Ch was 18%, which reflects the dif- ferent impacts of dead vegetation on momentum and heat transfer at this natural grassland. Moreover, with the removal of vegetation cover, grazing intensities reduced Z0m, Z0h, Cd, and Ch.展开更多
In this study, recurrent networks to downscale meteorological fields of the ERA-40 re-analysis dataset with focus on the meso-scale water balance were investigated. Therefore two types of recurrent neural networks wer...In this study, recurrent networks to downscale meteorological fields of the ERA-40 re-analysis dataset with focus on the meso-scale water balance were investigated. Therefore two types of recurrent neural networks were used. The first approach is a coupling between a recurrent neural network and a distributed watershed model and the second a nonlinear autoregressive with exogenous inputs (NARX) network, which directly predicted the component of the water balance. The approaches were deployed for a meso-scale catchment area in the Free State of Saxony, Germany. The results show that the coupled approach did not perform as well as the NARX network. But the meteorological output of the coupled approach already reaches an adequate quality. However the coupled model generates as input for the watershed model insufficient daily precipitation sums and not enough wet days were predicted. Hence the long-term annual cycle of the water balance could not be preserved with acceptable quality in contrary to the NARX approach. The residual storage change term indicates physical restrictions of the plausibility of the neural networks, whereas the physically based correlations among?the components of the water balance were preserved more accurately by the coupled approach.展开更多
Based on the carbon fluxes measured over the grassland ecosystems in Inner Mongolia （UG79 site）, Loess Plateau （SACOL site）, and Tongyu, Jilin Province （TY site） in the semiarid areas from 2007 to 2008 with the ...Based on the carbon fluxes measured over the grassland ecosystems in Inner Mongolia （UG79 site）, Loess Plateau （SACOL site）, and Tongyu, Jilin Province （TY site） in the semiarid areas from 2007 to 2008 with the eddy covariance method, we have investigated the carbon exchange processes over semiarid grassland ecosystem and its main affecting environmental variables. The precipitations at UG79 and TY sites in 2007 were below the historical average, especially for TY site, which was 50% be- low the historical average annual precipitation. The precipitation in SACOL site was close to average in 2007 but below average in 2008. The variation of monthly diurnal average NEE showed that the diurnal mean NEE decreased in the order of TY site, UG79 site, and SACOL site. However, a longer net carbon uptake period was observed at SACOL site. The diurnal course of NEE at UG79 site was similar between 2007 and 2008. The diurnal average NEE remained large during July and August in growing sea- son （May to September） at UG79 site, with maximum values approaching 0.08 mg C m^-2 s^-1 in August of 2008. The diurnal av- erage NEE of 2007 was larger than 2008 at SACOL site, with maximum values of 0.07 mg C m^-2 sq in September of 2007. A shorter carbon uptake period was recorded in 2007 at TY site, lasting from July to August. A larger diurnal average NEE oc- curred in 2008 at TY site, with maximum values of 0.12 mg C m^-2 s^-1. The ecosystem respirations of three sites were controlled by both soil temperature and soil volumetric water content （at a depth of 5 cm below the land surface）. Both UG79 site and SACOL site acted as a carbon sink during the growing periods of 2007 and 2008. Annual NEE in the growing seasons of 2007 and 2008 ranged from -68 to -50 g C m^-2 at UG79 site and from -109 to -55 g C m^-2at SACOL site. Alternation between car- bon source and carbon sink was found at TY site, with respective values of annual NEE in the growing seasons of 0.32 g C m^-2 and -73 g C m^-2 in 2007 and 2008. The magnitude and duration of carbon uptake depended mainly on the amount and timing of precipitation and the timing of the first effective rainfall during the growing season in semiarid grassland ecosystems.展开更多
基金supported by the German Science Foundation (DFG) within the Research Group 536"MAGIM" (Matter fluxes in grasslands of Inner Mongolia as influenced by stocking rate) under Grant BE 172/7-1 in cooperation with Inner Mongolia Grassland Ecosystem Research Stationthe National Basic Research Program of China (973 Program) under Grant 2010CB951801the Strategic Priority Research Program of Chinese Academy of Sciences under Grant XDA05110102
文摘The drag coefficient （Ca） and heat transfer coefficient （Ch） with the bulk transfer scheme are usually used to calculate the momentum and heat fluxes in meteorological models. The aerodynamic roughness length （z0m） and thermal roughness length （z0h） are two crucial parameters for bulk transfer equations. To improve the meteorological models, the seasonal and interannual variations of Z0m, Z0h, coefficient kB-1, Cd, and Ch were investigated based on eddy covariance data over different grazed semiarid grasslands of Inner Mongolia during the growing seasons （May to September） from 2005 to 2008. For an ungrazed Leymus chinensis grassland （ungrazed since 1979）, Z0m and z0h had significant seasonal and in- terannual variations. Zorn was affected by the amount and distribution of rainfall, kB 1 exhibited a relatively negative variation compared with z0h, which indicates that the seasonal variation of z0h cannot be described by kB 1. To parameterize Zorn and z0h, the linear regressions between ln（z0m）, ln（z0h）, and the leaf area index （LAI） were performed with R2=0.71 and 0.83. The monthly average kB-1 was found to decrease linearly with LAI. The four-year averaged values of Ca and Ch were 4.5 × 10^-3 and 3.9× 10^-3, respectively. The monthly average Cd only varied by 8% while the variation of Ch was 18%, which reflects the dif- ferent impacts of dead vegetation on momentum and heat transfer at this natural grassland. Moreover, with the removal of vegetation cover, grazing intensities reduced Z0m, Z0h, Cd, and Ch.
基金supported by the Erasmus Mundus Action 2 Programme of the European Union and the German Weather Service(DWD)and the Czech Hydrological-Meteorological Service(CHMI).
文摘In this study, recurrent networks to downscale meteorological fields of the ERA-40 re-analysis dataset with focus on the meso-scale water balance were investigated. Therefore two types of recurrent neural networks were used. The first approach is a coupling between a recurrent neural network and a distributed watershed model and the second a nonlinear autoregressive with exogenous inputs (NARX) network, which directly predicted the component of the water balance. The approaches were deployed for a meso-scale catchment area in the Free State of Saxony, Germany. The results show that the coupled approach did not perform as well as the NARX network. But the meteorological output of the coupled approach already reaches an adequate quality. However the coupled model generates as input for the watershed model insufficient daily precipitation sums and not enough wet days were predicted. Hence the long-term annual cycle of the water balance could not be preserved with acceptable quality in contrary to the NARX approach. The residual storage change term indicates physical restrictions of the plausibility of the neural networks, whereas the physically based correlations among?the components of the water balance were preserved more accurately by the coupled approach.
基金supported by the National Basic Research Program of China (Grant Nos.2010CB951801and 2006CB400501)the Foundation for Innovative Research Groups of the National Natural Science Foundation of China (Grant No.41021004)
文摘Based on the carbon fluxes measured over the grassland ecosystems in Inner Mongolia （UG79 site）, Loess Plateau （SACOL site）, and Tongyu, Jilin Province （TY site） in the semiarid areas from 2007 to 2008 with the eddy covariance method, we have investigated the carbon exchange processes over semiarid grassland ecosystem and its main affecting environmental variables. The precipitations at UG79 and TY sites in 2007 were below the historical average, especially for TY site, which was 50% be- low the historical average annual precipitation. The precipitation in SACOL site was close to average in 2007 but below average in 2008. The variation of monthly diurnal average NEE showed that the diurnal mean NEE decreased in the order of TY site, UG79 site, and SACOL site. However, a longer net carbon uptake period was observed at SACOL site. The diurnal course of NEE at UG79 site was similar between 2007 and 2008. The diurnal average NEE remained large during July and August in growing sea- son （May to September） at UG79 site, with maximum values approaching 0.08 mg C m^-2 s^-1 in August of 2008. The diurnal av- erage NEE of 2007 was larger than 2008 at SACOL site, with maximum values of 0.07 mg C m^-2 sq in September of 2007. A shorter carbon uptake period was recorded in 2007 at TY site, lasting from July to August. A larger diurnal average NEE oc- curred in 2008 at TY site, with maximum values of 0.12 mg C m^-2 s^-1. The ecosystem respirations of three sites were controlled by both soil temperature and soil volumetric water content （at a depth of 5 cm below the land surface）. Both UG79 site and SACOL site acted as a carbon sink during the growing periods of 2007 and 2008. Annual NEE in the growing seasons of 2007 and 2008 ranged from -68 to -50 g C m^-2 at UG79 site and from -109 to -55 g C m^-2at SACOL site. Alternation between car- bon source and carbon sink was found at TY site, with respective values of annual NEE in the growing seasons of 0.32 g C m^-2 and -73 g C m^-2 in 2007 and 2008. The magnitude and duration of carbon uptake depended mainly on the amount and timing of precipitation and the timing of the first effective rainfall during the growing season in semiarid grassland ecosystems.
