Effect of including land-use driven radiative forcing of the surface albedo of land on climate response in the 16th–21st centuries

A change in ecosystem types, such as through natural-vegetation-agriculture conversion, alters the surface albedo and triggers attendant shortwave radiative forcing (RF). This paper describes numerical experiments performed using the climate model (CM) of the Institute of Atmospheric Physics (IAP), Russian Academy of Sciences, for the 16th–21st centuries; this model simulated the response to a change in the contents of greenhouse gases (tropospheric and stratospheric), sulfate aerosols, solar constant, as well as the response to change in surface albedo of land due to natural-vegetation-agriculture conversion. These forcing estimates relied on actual data until the late 20th century. In the 21st century, the agricultural area was specified according to scenarios of the Land Use Harmonization project and other anthropogenic impacts were specified using SRES scenarios. The change in the surface vegetation during conversion from natural vegetation to agriculture triggers a cooling RF in most regions except for those of natural semiarid vegetation. The global and annual average RF derived from the IAP RAS CM in late 20th century is ?0.11 W m^?2. Including the land-use driven RF in IAP RAS CM appreciably reconciled the model calculations to observations in this historical period. For instance, in addition to the net climate warming, IAP RAS CM predicted an annually average cooling and reduction in precipitation in the subtropics of Eurasia and North America and in Amazonia and central Africa, as well as a local maximum in annually average and summertime warming in East China. The land-use driven RF alters the sign in the dependence that the amplitude of the annual cycle of the near-surface atmospheric temperature has on the annually averaged temperature. One reason for the decrease in precipitation as a result of a change in albedo due to land use may be the suppression of the convective activity in the atmosphere in the warm period (throughout the year in the tropics) and the corresponding decrease in convective precipitation. In the 21st century, the effect that the land-use driven RF has on the climate response for scenarios of anthropogenic impact is generally small.     

Changes in climatic characteristics of Northern Hemisphere extratropical land in the 21st century: Assessments with the IAP RAS climate model

Assessments of future changes in the climate of Northern Hemisphere extratropical land regions have been made with the IAP RAS climate model (CM) of intermediate complexity (which includes a detailed scheme of thermo- and hydrophysical soil processes) under prescribed greenhouse and sulfate anthropogenic forcing from observational data for the 19th and 20th centuries and from the SRES B1, A1B, and A2 scenarios for the 21st century. The annual mean warming of the extratropical land surface has been found to reach 2–5 K (3–10 K) by the middle (end) of the 21st century relative to 1961–1990, depending on the anthropogenic forcing scenario, with larger values in North America than in Europe. Winter warming is greater than summer warming. This is expressed in a decrease of 1–4 K (or more) in the amplitude of the annual harmonic of soil-surface temperature in the middle and high latitudes of Eurasia and North America. The total area extent of perennially frozen ground S _p in the IAP RAS CM changes only slightly until the late 20th century, reaching about 21 million km², and then decreases to 11–12 million km² in 2036–2065 and 4–8 million km² in 2071–2100. In the late 21st century, near-surface permafrost is expected to remain only in Tibet and in central and eastern Siberia. In these regions, depths of seasonal thaw exceed 1 m (2 m) under the SRES B1 (A1B or A2) scenario. The total land area with seasonal thaw or cooling is expected to decrease from the current value of 54–55 million km² to 38–42 in the late 21st century. The area of Northern Hemisphere snow cover in February is also reduced from the current value of 45–49 million km² to 31–37 million km². For the basins of major rivers in the extratropical latitudes of the Northern Hemisphere, runoff is expected to increase in central and eastern Siberia. In European Russia and in southern Europe, runoff is projected to decrease. In western Siberia (the Ob watershed), runoff would increase under the SRES A1B and A2 scenarios until the 2050s–2070s, then it would decrease to values close to present-day ones; under the anthropogenic forcing scenario SRES B1, the increase in runoff will continue up to the late 21st century. Total runoff from Eurasian rivers into the Arctic Ocean in the IAP RAS CM in the 21st century will increase by 8–9% depending on the scenario. Runoff from the North American rivers into the Arctic Ocean has not changed much throughout numerical experiments with the IAP RAS CM.     

Estimation of changes in characteristics of the climate and carbon cycle in the 21st century accounting for the uncertainty of terrestrial biota parameter values

ensemble simulations with the A.M. Obukhov Institute of Atmospheric Physics, Russian Academy of Sciences (IAP RAS) climate model (CM) for the 21st century are analyzed taking into account anthropogenic forcings in accordance with the Special Report on Emission Scenarios (SRES) A2, A1B, and B1, whereas agricultural land areas were assumed to change in accordance with the Land Use Harmonization project scenarios. Different realizations within these ensemble experiments were constructed by varying two governing parameters of the terrestrial carbon cycle. The ensemble simulations were analyzed with the use of Bayesian statistics, which makes it possible to suppress the influence of unrealistic members of these experiments on their results. It is established that, for global values of the main characteristics of the terrestrial carbon cycle, the SRES scenarios used do not differ statistically from each other, so within the framework of the model, the primary productivity of terrestrial vegetation will increase in the 21st century from 74 ± 1 to 102 ± 13 PgC yr⁻¹ and the carbon storage in terrestrial vegetation will increase from 511 ± 8 to 611 ± 8 PgC (here and below, we indicate the mean ± standard deviations). The mutual compensation of changes in the soil carbon stock in different regions will make global changes in the soil carbon storage in the 21st century statistically insignificant. The global CO₂ uptake by terrestrial ecosystems will increase in the first half of the 21st century, whereupon it will decrease. The uncertainty interval of this variable in the middle (end) of the 21st century will be from 1.3 to 3.4 PgC yr⁻¹ (from 0.3 to 3.1 PgC yr⁻¹). In most regions, an increase in the net productivity of terrestrial vegetation (especially outside the tropics), the accumulation of carbon in this vegetation, and changes in the amount of soil carbon stock (with the total carbon accumulation in soils of the tropics and subtropics and the regions of both accumulation and loss of soil carbon at higher latitudes) will be robust within the ensemble in the 21st century, as will the CO₂ uptake from the atmosphere only by terrestrial ecosystems located at extratropical latitudes of Eurasia, first and foremost by the Siberian taiga. However, substantial differences in anthropogenic emissions between the SRES scenarios in the 21st century lead to statistically significant differences between these scenarios in the carbon dioxide uptake by the ocean, the carbon dioxide content in the atmosphere, and changes in the surface air temperature. In particular, according to the SRES A2 (A1B, B1) scenario, in 2071–2100 the carbon flux from the atmosphere to the ocean will be 10.6 ± 0.6 PgC yr⁻¹ (8.3 ± 0.5, 5.6 ± 0.3 PgC yr⁻¹), and the carbon dioxide concentration in the atmosphere will reach 773 ± 28 ppmv (662 ± 24, 534 ± 16 ppmv) by 2100. The annual mean warming in 2071–2100 relatively to 1961–1990 will be 3.19 ± 0.09 K (2.52 ± 0.08, 1.84 ± 0.06 K).     

Estimation of the uncertainty of future changes in atmospheric carbon dioxide concentration and its radiative forcing

An ensemble experiment with the IAP RAS CM was performed to estimate future changes in the atmospheric concentration of carbon dioxide, its radiative forcing, and characteristics of the climate-carbon cycle feedback. Different ensemble members were obtained by varying the governing parameters of the terrestrial carbon cycle of the model. For 1860–2100, anthropogenic CO₂ emissions due to fossil-fuel burning and land use were prescribed from observational estimates for the 19th and 20th centuries. For the 21st century, emissions were taken from the SRES A2 scenario. The ensemble of numerical experiments was analyzed via Bayesian statistics, which made the uncertainty range of estimates much narrower. To distinguish between realistic and unrealistic ensemble members, the observational characteristics of the carbon cycle for the 20th century were used as a criterion. For the given emission scenario, the carbon dioxide concentration expected by the end of the 21st century falls into the range 818 ± 46 ppm (an average plus or minus standard deviation). The corresponding global instantaneous radiative forcing at the top of the atmosphere (relative to the preindustrial state) lies in the uncertainty range 6.8 ± 0.4 W m^?2. The uncertainty range of the strength of the climate-carbon cycle feedback by the end of the 21st century reaches 59 ± 98 ppm in terms of the atmospheric carbon dioxide concentration and 0.4 ± 0.7 W m^?2 in terms of the radiative forcing.     

The influence of lightning activity and anthropogenic factors on large-scale characteristics of natural fires

A module for simulating of natural fires (NFs) in the climate model of the A.M. Obukhov Institute of Atmospheric Physics, Russian Academy of Sciences (IAP RAS CM), is extended with respect to the influence of lightning activity and population density on the ignition frequency and fire suppression. The IAP RAS CM is used to perform numerical experiments in accordance with the conditions of the project that intercompares climate models, CMIP5 (Coupled Models Intercomparison Project, phase 5). The frequency of lightning flashes was assigned in accordance with the LIS/OTD satellite data. In the calculations performed, anthropogenic ignitions play an important role in NF occurrences, except for regions at subpolar latitudes and, to a lesser degree, tropical and subtropical regions. Taking into account the dependence of fire frequency on lightning activity and population density intensifies the influence of characteristics of natural fires on the climate changes in tropics and subtropics as compared to the version of the IAP RAS CM that does not take the influence of ignition sources on the large-scale characteristics of NFs into consideration.     

Influence of volcanic activity on climate change in the past several centuries: Assessments with a climate model of intermediate complexity

The climate model of intermediate complexity developed at the A.M. Obukhov Institute of Atmospheric Physics of the Russian Academy of Sciences (IAP RAS CM) is supplemented by a scheme which takes into account the volcanic forcing of climate. With this model, ensemble experiments have been conducted for the 1600s–1900s, in which, along with the volcanic forcing, the anthropogenic forcing due to greenhouse gases and sulfate aerosols and the natural forcing due to variations in solar irradiance were taken into account. The model realistically reproduces the annual mean response of surface air temperature and precipitation to major eruptions both globally and regionally. In particular, the decreases in the annual mean global temperature T _g in the IAP RAS CM after the largest eruptions in the latter half of the 20th century, the Mt. Agung (1963), El Chichon (1982), and Mt. Pinatubo (1991) volcanic eruptions, are 0.28, 0.27, and 0.46 K, respectively, in agreement with estimates from observational data. Moreover, in the IAP RAS CM, the volcanic eruptions result in a general precipitation decrease, especially over land in the middle and high latitudes of the Northern Hemisphere. The seasonal distribution of the response shows good agreement with observations for high-latitude eruptions and worse agreement for tropical and subtropical volcanoes. On interdecadal scales, volcanism leads to variations in T _g on the order of 0.1 K. In numerical experiments with anthropogenic and natural forcings, the model reproduces a general change in surface air temperature over the past several centuries. Taking into account the volcanic forcing, along with that due to variations in solar irradiance, the model has partly reproduced the nonmonotonic global warming for the 20th century.     

Interaction of the methane cycle and processes in wetland ecosystems in a climate model of intermediate complexity

The climate model of the Institute of Atmospheric Physics of the Russian Academy of Sciences (IAP RAS CM) has been supplemented with a module of soil thermal physics and the methane cycle, which takes into account the response of methane emissions from wetland ecosystems to climate changes. Methane emissions are allowed only from unfrozen top layers of the soil, with an additional constraint in the depth of the simulated layer. All wetland ecosystems are assumed to be water-saturated. The molar amount of the methane oxidized in the atmosphere is added to the simulated atmospheric concentration of CO₂. A control preindustrial experiment and a series of numerical experiments for the 17th–21st centuries were conducted with the model forced by greenhouse gases and tropospheric sulfate aerosols. It is shown that the IAP RAS CM generally reproduces preindustrial and current characteristics of both seasonal thawing/freezing of the soil and the methane cycle. During global warming in the 21st century, the permafrost area is reduced by four million square kilometers. By the end of the 21st century, methane emissions from wetland ecosystems amount to 130–140 Mt CH₄/year for the preindustrial and current period increase to 170–200 MtCH₄/year. In the aggressive anthropogenic forcing scenario A2, the atmospheric methane concentration grows steadily to ≈3900 ppb. In more moderate scenarios A1B and B1, the methane concentration increases until the mid-21st century, reaching ≈2100–2400 ppb, and then decreases. Methane oxidation in air results in a slight additional growth of the atmospheric concentration of carbon dioxide. Allowance for the interaction between processes in wetland ecosystems and the methane cycle in the IAP RAS CM leads to an additional atmospheric methane increase of 10–20% depending on the anthropogenic forcing scenario and the time. The causes of this additional increase are the temperature dependence of integral methane production and the longer duration of a warm period in the soil. However, the resulting enhancement of the instantaneous greenhouse radiative forcing of atmospheric methane and an increase in the mean surface air temperature are small (globally < 0.1 W/m² and 0.05 K, respectively).     

Influence of direct sulfate-aerosol radiative forcing on the results of numerical experiments with a climate model of intermediate complexity

The climate model of intermediate complexity developed at the Institute of Atmospheric Physics of the Russian Academy of Sciences (IAP RAS CM) is extended by a block for the direct anthropogenic sulfate-aerosol (SA) radiative forcing. Numerical experiments have been performed with prescribed scenarios of the greenhouse and anthropogenic sulfate radiative forcings from observational estimates for the 19th and 20th centuries and from SRES scenarios A1B, A2, and B1 for the 21st century. The globally averaged direct anthropogenic SA radiative forcing F _ASA by the end of the 20th century relative to the preindustrial state is ?0.34 W/m², lying within the uncertainty range of the corresponding present-day estimates. The absolute value of F _ASA is the largest in Europe, North America, and southeastern Asia. A general increase in direct radiative forcing in the numerical experiments that have been performed continues until the mid-21st century. With both the greenhouse and the sulfate loadings included, the global climate warming in the model is 1.5–2.8 K by the end of the 21st century relative to the late 20th century, depending on the scenario, and 2.1–3.4 K relative to the preindustrial period. The sulfate aerosol reduces global warming by 0.1–0.4 K in different periods depending on the scenario. The largest slowdown (>1.5 K) occurs over land at middle and high latitudes in the Northern Hemisphere in the mid-21st century for scenario A2. The IAP RAS CM response to the greenhouse and the aerosol forcing is not additive.     

Simulation of characteristics of thermal and hydrologic soil regimes in equilibrium numerical experiments with a climate model of intermediate complexity

The IAP RAS CM (Institute of Atmospheric Physics, Russian Academy of Sciences, climate model) has been extended to include a comprehensive scheme of thermal and hydrologic soil processes. In equilibrium numerical experiments with specified preindustrial and current concentrations of atmospheric carbon dioxide, the coupled model successfully reproduces thermal characteristics of soil, including the temperature of its surface, and seasonal thawing and freezing characteristics. On the whole, the model also reproduces soil hydrology, including the winter snow water equivalent and river runoff from large watersheds. Evapotranspiration from the soil surface and soil moisture are simulated somewhat worse. The equilibrium response of the model to a doubling of atmospheric carbon dioxide shows a considerable warming of the soil surface, a reduction in the extent of permanently frozen soils, and the general growth of evaporation from continents. River runoff increases at high latitudes and decreases in the subtropics. The results are in qualitative agreement with observational data for the 20th century and with climate model simulations for the 21st century.     

Climate and carbon cycle variations in the 20th and 21st centuries in a model of intermediate complexity

The climate model of intermediate complexity developed at the Oboukhov Institute of Atmospheric Physics, Russian Academy of Sciences (IAP RAS CM), has been supplemented by a zero-dimensional carbon cycle model. With the carbon dioxide emissions prescribed for the second half of the 19th century and for the 20th century, the model satisfactorily reproduces characteristics of the carbon cycle over this period. However, with continued anthropogenic CO₂ emissions (SRES scenarios A1B, A2, B1, and B2), the climate-carbon cycle feedback in the model leads to an additional atmospheric CO₂ increase (in comparison with the case where the influence of climate changes on the carbon exchange between the atmosphere and the underlying surface is disregarded). This additional increase is varied in the range 67–90 ppmv depending on the scenario and is mainly due to the dynamics of soil carbon storage. The climate-carbon cycle feedback parameter varies nonmonotonically with time. Positions of its extremes separate characteristic periods of the change in the intensity of anthropogenic emissions and of climate variations. By the end of the 21st century, depending on the emission scenario, the carbon dioxide concentration is expected to increase to 615–875 ppmv and the global temperature will rise by 2.4–3.4 K relative to the preindustrial value. In the 20th–21st centuries, a general growth of the buildup of carbon dioxide in the atmosphere and ocean and its reduction in terrestrial ecosystems can be expected. In general, by the end of the 21st century, the more aggressive emission scenarios are characterized by a smaller climate-carbon cycle feedback parameter, a lower sensitivity of climate to a single increase in the atmospheric concentration of carbon dioxide, a larger fraction of anthropogenic emissions stored in the atmosphere and the ocean, and a smaller fraction of emissions in terrestrial ecosystems.     

Impact of subgrid-scale vegetation heterogeneity on the simulation of carbon-cycle characteristics

The carbon cycle module of the global climate model developed at the Obukhov Institute of Atmospheric Physics, Russian Academy of Sciences (IAP RAS CM) has been extended by implementing the subgrid-scale heterogeneity (SH) of plant functional types (PFTs). It is found that subgrid-scale PFT heterogeneity enhances the photosynthesis intensity and increases vegetation and soil carbon stocks in grass-dominated regions. In forest-dominated regions, photosynthesis is suppressed and vegetation and soil carbon stocks are diminished. Regionally, accounting for subgrid-scale vegetation heterogeneity may lead to twofold changes in these variables. On the whole, accounting for subgrid-scale PFT heterogeneity enhances (suppresses) the carbon flux in regions where it is directed from terrestrial ecosystems to the atmosphere (from the atmosphere to terrestrial ecosystems).     

Coupled simulation of climate and vegetation dynamics

Experiments are performed with the climate model of the Institute of Numerical Mathematics, Russian Academy of Sciences (INM RAS), integrated jointly with the Lund-Potsdam-Jena dynamic global vegetation model (LPJ-DGVM). It is shown that the coupled model reproduces the distribution of basic plant functional types around the world quite well. In simulations of climate for the 21st century, this model predicts changes in dominant plant types and in the total area occupied by vegetation regionally. However, it does not reproduce the significant inverse influence that vegetation succession has on the simulated climate.     

Estimating the efficiency of mitigating and preventing global warming with scenarios of controlled aerosol emissions into the stratosphere

Ensemble numerical experiments with the climate model of intermediate complexity developed at the A.M. Obukhov Institute of Atmospheric Physics of the Russian Academy of Sciences (IAP RAS CM) are conducted to estimate the efficiency of controlled climate forcing (geoengineering) due to stratospheric sulfate aerosol (SSA) emissions in order to compensate for global warming under the SRES A1B anthropogenic emission scenario. Full (or even excessive) compensation for the expected anthropogenic warming in the model is possible with sufficiently intense geoengineering. For ensemble members with values of the governing parameters corresponding to those obtained for the Mt. Pinatubo eruption, global warming is reduced by no more than 0.46 K in the second half of the 21st century, with a residual rise in the global surface temperature T _g comparative to 1961–1990 of 1.0–1.2 K by 2050 and 1.9–2.2 K by 2100. The largest reduction in global warming (with the other parameters of the numerical experiment being equal) is found not for a meridional distribution of SSA concentration peaked at low latitudes (despite the largest (in magnitude) global compensation instantaneous radiative forcing), but for a uniform horizontal aerosol distribution and for a distribution with the SSA concentration maximum in the middle and subpolar latitudes of the Northern Hemisphere. The efficiency of geoengineering in terms of T _g in the second half of the 21st century between the most efficient and the least efficient meridional distributions of stratospheric aerosols differs by as much as one-third, depending on the values of other governing parameters. For meridional distributions of SSA concentration, which produce the largest deceleration of global warming, such a deceleration is regionally most pronounced over high- and subpolarlatitude land areas and in the Arctic. In particular, this is expressed in the smallest reduction in the sea-ice extent and permafrost area under climate warming in the model. The compensation forcing also decelerates a general increase in global annual precipitation P _g during warming. The relative deceleration in precipitation increase is most pronounced in land regions outside the tropics, where a significant deficit in precipitation is currently observed. After the theoretical completion of geoengineering in the first or second decade, its temperature effect vanishes with an abrupt acceleration of global and regional surface warming. For individual members of the ensemble experiment, the global temperature change in this period is five times as large as that in the experiment without geoengineering and ten times as large regionally (in northeastern Siberia).     

Validating and assessing the sensitivity of the climate model with an ocean general circulation model developed at the Institute of Atmospheric Physics,Russian Academy of Sciences

A new version of the Institute of Atmospheric Physics, Russian Academy of Sciences (IAP RAS), climate model (CM) has been developed using an ocean general circulation model instead of the statistical-dynamical ocean model applied in the previous version. The spatial resolution of the new ocean model is 3° in latitude and 5° in longitude, with 25 unevenly spaced vertical levels. In the previous version of the oceanic model, as in the atmospheric model, the horizontal resolution was 4.5° in latitude and 6° in longitude, with four vertical levels (the upper quasi-homogeneous layer, seasonal thermocline, abyssal ocean, and bottom friction layer). There is no correction for the heat and momentum fluxes between the atmosphere and ocean in the new version of the IAP RAS CM. Numerical experiments with the IAP RAS CM have been performed under current initial and boundary conditions, as well as with an increasing concentration of atmospheric carbon dioxide. The main simulated atmospheric and oceanic fields agree quite well with observational data. The new version’s equilibrium temperature sensitivity to atmospheric CO₂ doubling was found to be 2.9 K. This value lies in the mid-range of estimates (2–4.5 K) obtained from simulations with state-of-the-art models of different complexities.     

北极夏季海冰反照率的观测和数值模拟试验

在中国第3次北极科学考察浮冰站开展了积雪/海冰反照率观测.本文对观测结果进行了分析,并结合一维高分辨雪/冰模式(HIGHTSI)对3个常用的反照率参数化方案在天气尺度的表现进行了评估.观测期间测站反照率变化范围0.75～0.85,其天气尺度变化同天气和表面冰、雪状况紧密相关,降雪和吹雪过程可改变表面积雪厚度及水平分布,...     

Optical properties and surface energy flux of spring fast ice in the Arctic

Over the past decades, sea ice in the polar regions has been significantly affecting local and even hemispheric climate through a positive ice albedo feedback mechanism. The role of fast ice, as opposed to drift ice, has not been well-studied due to its relatively small coverage over the earth. In this paper, the optical properties and surface energy balance of land fast ice in spring are studied using in situ observations in Barrow, Alaska. The results show that the albedo of the fast ice varied between 0.57 and 0.85 while the transmittance increased from 1.3×10?3 to 4.1×10?3 during the observation period. Snowfall and air temperature affected the albedo and absorbance of sea ice, but the transmittance had no obvious relationship with precipitation or snow cover. Net solar shortwave radiation contributes to the surface energy balance with a positive 99.2% of the incident flux, with sensible heat flux for the remaining 0.8%. Meanwhile, the ice surface loses energy through the net longwave radiation by 18.7% of the total emission, while the latent heat flux accounts for only 0.1%. Heat conduction is also an important factor in the overall energy budget of sea ice, contributing 81.2% of the energy loss. Results of the radiative transfer model reveal that the spectral transmittance of the fast ice is determined by the thickness of snow and sea ice as well as the amount of inclusions. As major inclusions, the ice biota and particulates have a significant influence on the magnitude and distribution of the spectral transmittance. Based on the radiative transfer model, concentrations of chlorophyll and particulate in the fast ice are estimated at 5.51 mg/m2 and 95.79 g/m2, which are typical values in the spring in Barrow.     

城市热岛效应的卫星遥感分析

利用MODIS资料研究了2004年4月南京城市热岛特征及其影响因子,结合地表覆盖类型分析了植被归一化指数（Normalized Difference Vegetation Index,NDVI）、地表温度（Ts）、地表反照率（α）的城乡差异及其相互关系,探讨了城市热岛（Urban Heat Island,UHI）效应形成的机制。结果表明：南京城区存在着明显的城市热岛效应;城市平均Ts比乡村高约10．83％;城市NDVI和α分别比乡村低约为62％和18.75％;NDVI与Ts呈负相关,相关系数为-0．73,而NDVI与α之间关系与波段有关;城乡植被覆盖差异是造成UHI的主要原因,其次是地表反照率,通过提高植被覆盖率和地表反照率可以减小城市热岛效应。     

CMIP5模式对南海SST的模拟和预估

分析了32个CMIP5模式对南海历史海表温度(SST)的模拟能力和不同排放情景下未来SST变化的预估。通过检验各气候模式对南海历史SST增温趋势和均方差的模拟,发现大部分模式都能较好地模拟出南海20世纪历史SST的基本特征和变化规律,但也有部分模式的模拟存在较大偏差。尽管这些模拟偏差较大的模式对SST多模式集合平均的影响不大,但会增加未来情景预估的不确定性。剔除15个模式后,分析了南海SST在RCP26、RCP45和RCP85三种排放情景下的变化趋势,发现在未来百年呈明显的增温趋势,多模式集合平均的增温趋势分别为0.42、1.50和3.30℃/(100a)。这些增温趋势在空间上变化不大,但随时间并不是均匀变化的。在前两种排放情景下,21世纪前期的增温趋势明显强于后期,而在RCP85情景下,21世纪后期的增温趋势强于前期。     

基于Icepack海冰柱模式的融池反照率模拟研究

基于Icepack一维海冰柱模式,以2014年中国第6次北极科学考察长期冰站ICE06的3个融池的辐射参量和气象参量的连续观测作为大气强迫数据,对融池反照率及相关参量进行了模拟。本文引入观测的融池深度及海冰厚度作为初始条件,通过考虑融池覆盖率的作用,改进了平整冰融池参数化方案中海冰干舷的计算,修正了冰上可允许的最大融池深度,成功实现了对融池参数变化的模拟;同时,还修正了入射辐射分量比例系数与对应反照率分量权重系数不一致的问题。标准试验中,模拟的3个融池的反照率与观测结果之间的平均误差分别为0.01、0.05和0.13;入射辐射比例的敏感性试验结果表明,当可见光辐射比例增大8%时,融池反照率的模拟结果增大了6%～8%;融池表面再冻结试验的结果显示,当再冻结冰层厚度小于2 cm时,模拟冰面反照率的增加不足0.006,由此引起的表面能量收支减少了约1.1 W/m2。本文研究指出,准确的入射辐射比例对于改善北极海冰反照率模拟是必要的;并指出目前模式仍存在融池表面再冻结参数化、热收支计算、表面吹雪效应等有待解决的问题。