Predictor Selection for CNN-based Statistical Downscaling of Monthly Precipitation

Convolutional neural networks(CNNs) have been widely studied and found to obtain favorable results in statistical downscaling to derive high-resolution climate variables from large-scale coarse general circulation models(GCMs).However, there is a lack of research exploring the predictor selection for CNN modeling. This paper presents an effective and efficient greedy elimination algorithm to address this problem. The algorithm has three main steps: predictor importance attribution, predictor removal, and CNN retraining, which are performed sequentially and iteratively. The importance of individual predictors is measured by a gradient-based importance metric computed by a CNN backpropagation technique, which was initially proposed for CNN interpretation. The algorithm is tested on the CNN-based statistical downscaling of monthly precipitation with 20 candidate predictors and compared with a correlation analysisbased approach. Linear models are implemented as benchmarks. The experiments illustrate that the predictor selection solution can reduce the number of input predictors by more than half, improve the accuracy of both linear and CNN models,and outperform the correlation analysis method. Although the RMSE(root-mean-square error) is reduced by only 0.8%,only 9 out of 20 predictors are used to build the CNN, and the FLOPs(Floating Point Operations) decrease by 20.4%. The results imply that the algorithm can find subset predictors that correlate more to the monthly precipitation of the target area and seasons in a nonlinear way. It is worth mentioning that the algorithm is compatible with other CNN models with stacked variables as input and has the potential for nonlinear correlation predictor selection.

Statistical Downscaling Retrieval of Land Surface Temperature in an Area with Complex Landforms in the Eastern Qinling Mountains of China Based on Sentinel-2/3 Satellite Data

The study of land surface temperature(LST)is of great significance for ecosystem monitoring and ecological environmental protection in the Qinling Mountains of China.In view of the contradicting spatial and temporal resolutions in extracting LST from satellite remote sensing(RS)data,the areas with complex landforms of the Eastern Qinling Mountains were selected as the research targets to establish the correlation between the normalized difference vegetation index(NDVI)and LST.Detailed information on the surface features and temporal changes in the land surface was provided by Sentinel-2 and Sentinel-3,respectively.Based on the statistically downscaling method,the spatial scale could be decreased from 1000 m to 10 m,and LST with a Sentinel-3 temporal resolution and a 10 m spatial resolution could be retrieved.Comparing the 1 km resolution Sentinel-3 LST with the downscaling results,the 10 m LST downscaling data could accurately reflect the spatial distribution of the thermal characteristics of the original LST image.Moreover,the surface temperature data with a 10 m high spatial resolution had clear texture and obvious geomorphic features that could depict the detailed information of the ground features.The results showed that the average error was 5 K on April 16,2019 and 2.6 K on July 15,2019.The smaller error values indicated the higher vegetation coverage of summer downscaling result with the highest level on July 15.

Statistical Downscaling for Multi-Model Ensemble Prediction of Summer Monsoon Rainfall in the Asia-Pacific Region Using Geopotential Height Field

The 21-yr ensemble predictions of model precipitation and circulation in the East Asian and western North Pacific （Asia-Pacific） summer monsoon region （0°-50°N, 100° 150°E） were evaluated in nine different AGCM, used in the Asia-Pacific Economic Cooperation Climate Center （APCC） multi-model ensemble seasonal prediction system. The analysis indicates that the precipitation anomaly patterns of model ensemble predictions are substantially different from the observed counterparts in this region, but the summer monsoon circulations are reasonably predicted. For example, all models can well produce the interannual variability of the western North Pacific monsoon index （WNPMI） defined by 850 hPa winds, but they failed to predict the relationship between WNPMI and precipitation anomalies. The interannual variability of the 500 hPa geopotential height （GPH） can be well predicted by the models in contrast to precipitation anomalies. On the basis of such model performances and the relationship between the interannual variations of 500 hPa GPH and precipitation anomalies, we developed a statistical scheme used to downscale the summer monsoon precipitation anomaly on the basis of EOF and singular value decomposition （SVD）. In this scheme, the three leading EOF modes of 500 hPa GPH anomaly fields predicted by the models are firstly corrected by the linear regression between the principal components in each model and observation, respectively. Then, the corrected model GPH is chosen as the predictor to downscale the precipitation anomaly field, which is assembled by the forecasted expansion coefficients of model 500 hPa GPH and the three leading SVD modes of observed precipitation anomaly corresponding to the prediction of model 500 hPa GPH during a 19-year training period. The cross-validated forecasts suggest that this downscaling scheme may have a potential to improve the forecast skill of the precipitation anomaly in the South China Sea, western North Pacific and the East Asia Pacific regions, where the anomaly correlation coefficient （ACC） has been improved by 0.14, corresponding to the reduced RMSE of 10.4% in the conventional multi-model ensemble （MME） forecast.

Using Statistical Downscaling to Quantify the GCM-Related Uncertainty in Regional Climate Change Scenarios: A Case Study of Swedish Precipitation

There are a number of sources of uncertainty in regional climate change scenarios. When statistical downscaling is used to obtain regional climate change scenarios, the uncertainty may originate from the uncertainties in the global climate models used, the skill of the statistical model, and the forcing scenarios applied to the global climate model. The uncertainty associated with global climate models can be evaluated by examining the differences in the predictors and in the downscaled climate change scenarios based on a set of different global climate models. When standardized global climate model simulations such as the second phase of the Coupled Model Intercomparison Project （CMIP2） are used, the difference in the downscaled variables mainly reflects differences in the climate models and the natural variability in the simulated climates. It is proposed that the spread of the estimates can be taken as a measure of the uncertainty associated with global climate models. The proposed method is applied to the estimation of global-climate-model-related uncertainty in regional precipitation change scenarios in Sweden. Results from statistical downscaling based on 17 global climate models show that there is an overall increase in annual precipitation all over Sweden although a considerable spread of the changes in the precipitation exists. The general increase can be attributed to the increased large-scale precipitation and the enhanced westerly wind. The estimated uncertainty is nearly independent of region. However, there is a seasonal dependence. The estimates for winter show the highest level of confidence, while the estimates for summer show the least.

Statistical Downscaling of Summer Temperature Extremes in Northern China

Two approaches of statistical downscaling were applied to indices of temperature extremes based on percentiles of daily maximum and minimum temperature observations at Beijing station in summer during 1960-2008. One was to downscale daily maximum and minimum temperatures by using EOF analysis and stepwise linear regression at first, then to calculate the indices of extremes; the other was to directly downseale the percentile-based indices by using seasonal large-scale temperature and geo-potential height records. The cross-validation results showed that the latter approach has a better performance than the former. Then, the latter approach was applied to 48 meteorological stations in northern China. The cross- validation results for all 48 stations showed close correlation between the percentile-based indices and the seasonal large-scale variables. Finally, future scenarios of indices of temperature extremes in northern China were projected by applying the statistical downsealing to Hadley Centre Coupled Model Version 3 （HadCM3） simulations under the Representative Concentration Pathways 4.5 （RCP 4.5） scenario of the Fifth Coupled Model Inter-comparison Project （CMIP5）. The results showed that the 90th percentile of daily maximum temperatures will increase by about 1.5℃, and the 10th of daily minimum temperatures will increase by about 2℃ during the period 2011- 35 relative to 1980-99.

Seasonal Prediction of June Rainfall over South China:Model Assessment and Statistical Downscaling

The performances of various dynamical models from the Asia-Pacific Economic Cooperation（APEC） Climate Center（APCC） multi-model ensemble（MME） in predicting station-scale rainfall in South China（SC） in June were evaluated.It was found that the MME mean of model hindcasts can skillfully predict the June rainfall anomaly averaged over the SC domain.This could be related to the MME＇s ability in capturing the observed linkages between SC rainfall and atmospheric large-scale circulation anomalies in the Indo-Pacific region.Further assessment of station-scale June rainfall prediction based on direct model output（DMO） over 97 stations in SC revealed that the MME mean outperforms each individual model.However,poor prediction abilities in some in-land and southeastern SC stations are apparent in the MME mean and in a number of models.In order to improve the performance at those stations with poor DMO prediction skill,a station-based statistical downscaling scheme was constructed and applied to the individual and MME mean hindcast runs.For several models,this scheme can outperform DMO at more than 30 stations,because it can tap into the abilities of the models in capturing the anomalous Indo-Paciric circulation to which SC rainfall is considerably sensitive.Therefore,enhanced rainfall prediction abilities in these models should make them more useful for disaster preparedness and mitigation purposes.

Temporal Statistical Downscaling of Precipitation and Temperature Forecasts Using a Stochastic Weather Generator

Statistical downscaling is based on the fact that the large-scale climatic state and regional/local physiographic features control the regional climate. In the present paper, a stochastic weather generator is applied to seasonal precipitation and temperature forecasts produced by the International Research Institute for Climate and Society （IRI）. In conjunction with the GLM （generalized linear modeling） weather generator, a resampling scheme is used to translate the uncertainty in the seasonal forecasts （the IRI format only specifies probabilities for three categories： below normal, near normal, and above normal） into the corresponding uncertainty for the daily weather statistics. The method is able to generate potentially useful shifts in the probability distributions of seasonally aggregated precipitation and minimum and maximum temperature, as well as more meaningful daily weather statistics for crop yields, such as the number of dry days and the amount of precipitation on wet days. The approach is extended to the case of climate change scenarios, treating a hypothetical return to a previously observed drier regime in the Pampas.

A Hybrid Statistical-Dynamical Downscaling of Air Temperature over Scandinavia Using the WRF Model

An accurate simulation of air temperature at local scales is crucial for the vast majority of weather and climate applications.In this work,a hybrid statistical–dynamical downscaling method and a high-resolution dynamical-only downscaling method are applied to daily mean,minimum and maximum air temperatures to investigate the quality of localscale estimates produced by downscaling.These two downscaling approaches are evaluated using station observation data obtained from the Finnish Meteorological Institute over a near-coastal region of western Finland.The dynamical downscaling is performed with the Weather Research and Forecasting(WRF)model,and the statistical downscaling method implemented is the Cumulative Distribution Function-transform(CDF-t).The CDF-t is trained using 20 years of WRF-downscaled Climate Forecast System Reanalysis data over the region at a 3-km spatial resolution for the central month of each season.The performance of the two methods is assessed qualitatively,by inspection of quantile-quantile plots,and quantitatively,through the Cramer-von Mises,mean absolute error,and root-mean-square error diagnostics.The hybrid approach is found to provide significantly more skillful forecasts of the observed daily mean and maximum air temperatures than those of the dynamical-only downscaling(for all seasons).The hybrid method proves to be less computationally expensive,and also to give more skillful temperature forecasts(at least for the Finnish near-coastal region).

Projecting future precipitation change across the semi-arid Borana lowland,southern Ethiopia

Climate change caused by past,current,and future greenhouse gas emissions has become a major concern for scientists in the field in many countries and regions of the world.This study modelled future precipitation change by downscaling a set of large-scale climate predictor variables(predictors)from the second generation Canadian Earth System Model(CanESM2)under two Representative Concentration Pathway(RCP)emission scenarios(RCP4.5 and RCP8.5)in the semi-arid Borana lowland,southern Ethiopia.The Statistical DownScaling Model(SDSM)4.2.9 was employed to downscale and project future precipitation change in the middle(2036-2065;2050s)and far(2066-2095;2080s)future at the local scale.Historical precipitation observations from eight meteorological stations stretching from 1981 to 1995 and 1996 to 2005 were used for the model calibration and validation,respectively,and the time period of 1981-2018 was considered and used as the baseline period to analyze future precipitation change.The results revealed that the surface-specific humidity and the geopotential height at 500 hPa were the preferred large-scale predictors.Compared to the middle future(2050s),precipitation showed a much greater increase in the far future(2080s)under both RCP4.5 and RCP8.5 scenarios at all meteorological stations(except Teletele and Dillo stations).At Teltele station,the projected annual precipitation will decrease by 26.53%(2050s)and 39.45%(2080s)under RCP4.5 scenario,and 34.99%(2050s)and 60.62%(2080s)under RCP8.5 scenario.Seasonally,the main rainy period would shift from spring(March to May)to autumn(September to November)at Dehas,Dire,Moyale,and Teltele stations,but for Arero and Yabelo stations,spring would consistently receive more precipitation than autumn.It can be concluded that future precipitation in the semi-arid Borana lowland is predicted to differ under the two climate scenarios(RCP4.5 and RCP8.5),showing an increasing trend at most meteorological stations.This information could be helpful for policymakers to design adaptation plans in water resources management,and we suggest that the government should give more attention to improve early warning systems in drought-prone areas by providing dependable climate forecast information as early as possible.

Downscaling GCMs Using the Smooth Support Vector Machine Method to Predict Daily Precipitation in the Hanjiang Basin

General circulation models （GCMs） are often used in assessing the impact of climate change at global and continental scales. However, the climatic factors simulated by GCMs are inconsistent at comparatively smaller scales, such as individual river basins. In this study, a statistical downscaling approach based on the Smooth Support Vector Machine （SSVM） method was constructed to predict daily precipitation of the changed climate in the Hanjiang Basin. NCEP/NCAR reanalysis data were used to establish the statistical relationship between the larger scale climate predictors and observed precipitation. The relationship obtained was used to project future precipitation from two GCMs （CGCM2 and HadCM3） for the A2 emission scenario. The results obtained using SSVM were compared with those from an artificial neural network （ANN）. The comparisons showed that SSVM is suitable for conducting climate impact studies as a statistical downscaling tool in this region. The temporal trends projected by SSVM based on the A2 emission scenario for CGCM2 and HadCM3 were for rainfall to decrease during the period 2011–2040 in the upper basin and to increase after 2071 in the whole of Hanjiang Basin.

Analysis and prediction of reference evapotranspiration with climate change in Xiangjiang River Basin, China

Reference evapotranspiration （ETo） is often used to estimate actual evapotranspiration in water balance studies. In this study, the present and future spatial distributions and temporal trends of ETo in the Xiangjiang River Basin （XJRB） in China were analyzed. ETo during the period from 1961 to 2010 was calculated with historical meteorological data using the FAO Penman-Monteith （FAO P-M） method, while ETo during the period from 2011 to 2100 was downscaled from the Coupled Model Intercomparison Project Phase 5 （CMIP5） outputs under two emission scenarios, representative concentration pathway 4.5 and representative concentration pathway 8.5 （RCP45 and RCP85）, using the statistical downscaling model （SDSM）. The spatial distribution and temporal trend of ETo were interpreted with the inverse distance weighted （IDW） method and Mann-Kendall test method, respectively. Results show that： （1） the mean annual ETo of the XJRB is 1 006.3 mm during the period from 1961 to 2010, and the lowest and highest values are found in the northeast and northwest parts due to the high latitude and spatial distribution of climatic factors, respectively; （2） the SDSM performs well in simulating the present ETo and can be used to predict the future ETo in the XJRB; and （3） CMIP5 predicts upward trends in annual ETo under the RCP45 and RCP85 scenarios during the period from 2011 to 2100. Compared with the reference period （1961-1990）, ETo increases by 9.8%, 12.6%, and 15.6% under the RCP45 scenario and 10.2%, 19.1%, and 27.3% under the RCP85 scenario during the periods from 2011 to 2040, from 2041 to 2070, and from 2071 to 2100, respectively. The predicted increasing ETo under the RCP85 scenario is greater than that under the RCP45 scenario during the period from 2011 to 2100.

CLDASSD:Reconstructing Fine Textures of the Temperature Field Using Super-Resolution Technology

Before 2008,the number of surface observation stations in China was small.Thus,the surface observation data were too sparse to effectively support the High-resolution China Meteorological Administration’s Land Assimilation System(HRCLDAS)which ultimately inhibited the output of high-resolution and high-quality gridded products.This paper proposes a statistical downscaling model based on a deep learning algorithm in super-resolution to research the above problem.Specifically,we take temperature as an example.The model is used to downscale the 0.0625°×0.0625°,2-m temperature data from the China Meteorological Administration’s Land Data Assimilation System(CLDAS)to 0.01°×0.01°,named CLDASSD.We performed quality control on the paired data from CLDAS and HRCLDAS,using data from 2018 and 2019.CLDASSD was trained on the data from 31 March 2018 to 28 February 2019,and then tested with the remaining data.Finally,extensive experiments were conducted in the Beijing-Tianjin-Hebei region which features complex and diverse geomorphology.Taking the HRCLDAS product and surface observation data as the"true values"and comparing them with the results of bilinear interpolation,especially in complex terrain such as mountains,the root mean square error(RMSE)of the CLDASSD output can be reduced by approximately 0.1℃,and its structural similarity(SSIM)was approximately 0.2 higher.CLDASSD can estimate detailed textures,in terms of spatial distribution,with greater accuracy than bilinear interpolation and other sub-models and can perform the expected downscaling tasks.

Effect of future climate change on the water footprint of major crops in southern Tajikistan

Danghara,a major food production area in southern Tajikistan,is currently suffering from the impact of rapid climate change and intensive human activities.Assessing the future impact of climate change on crop water requirements(CWRs)for the current growing period and defining the optimal sowing date to reduce future crop water demand are essential for local/regional water and food planning.Therefore,this study attempted to analyze possible future climate change effects on the water requirements of major crops using the statistical downscaling method in the Danghara District to simulate the future temperature and precipitation for two future periods(2021-2050 and 2051-2080),under three representative concentration pathways(RCP2.6,RCP4.5,and RCP8.5)according to the CanESM2 global climate model.The water footprint(WFP)of major crops was calculated as a measure of their CWRs.The increased projection of precipitation and temperature probably caused an increase in the main crop’s WFP for the current growing period,which was mainly due to the green water(GW)component in the long term and a decrease in the blue water(BW)component during the second future period,except for cotton,where all components were predicted to remain stable.Under three scenarios for the two future potato and winter wheat decreased from 5.7%to 4.8%and 3.4%to 2.2%,respectively.Although the WFP of cotton demonstrated a stable increase,according to the optimal sowing date,adecrease in irrigation demand or Bw was expected.The results of our study might be useful fordeveloping a new strategy related to irrigation systems and could help to find a balance betweenwater and food for environmental water demands and human use.

Hydrological Impact Assessment of Climate Change on Lake Tana’s Water Balance, Ethiopia

The aim of this study is to evaluate the hydrological impacts of climate change on the water balance of Lake Tana in Ethiopia. Impact assessments are by downscaled General Circulation Model (GCM) output and hydrological modeling. For A2 and B2 emission scenarios, precipitation, maximum and minimum temperature estimates from the HadCM3 GCM were used. GCM output was downscaled using the Statistical DownScaling Model (SDSM 4.2). Impact analyses were applied for three future time periods: early, mid and late 21st century. Over-lake evaporation is estimated by Hardgrave’s method, and over-lake precipitation is estimated by inverse distance weighing interpolation, whereas inflows from gauged and ungauged catchments are simulated by the HBV hydrological model. Findings indicate increases in maximum and minimum temperature on annual base for both emission scenarios. The projection of mean annual over lake precipitation for both A2 and B2 emission scenarios shows increasing pattern for 21st century in comparison to the baseline period. The increase of mean annual precipitation for A2 emission scenario is 9% (112 mm/year), 10% (125 mm/year) and 11% (137 mm/year) for the three future periods respectively. B2 emission scenario mean annual precipitation shows increase by 9% (111 mm/year), 10% (122 mm/year) and 10% (130 mm/year) respectively for the three future periods. Findings indicate consistent increases of lake storage for all three future periods for both A2 and B2 emission scenarios.

Adaptive Statistical Spatial Downscaling of Precipitation Supported by High-Resolution Atmospheric Simulation Data for Mountainous Areas of Nepal

In complex terrain regions, it is very challenging to obtain high accuracy and resolution precipitation data that are required in land hydrological studies. In this study, an adaptive precipitation downscaling method is proposed based on the statistical downscaling model MicroMet. A key input parameter in the MicroMet is the precipitation adjustment factor(PAF) that shows the elevation dependence of precipitation. Its value is estimated conventionally based on station observations and suffers sparse stations in high altitudes. This study proposes to estimate the PAF value and its spatial variability with precipitation data from high-resolution atmospheric simulations and tests the idea in Nepal of South Himalayas, where rainfall stations are relatively dense. The result shows that MicroMet performs the best with the PAF value estimated from the simulation data at the scale of approximately 1.5 degrees. Not only the value at this scale is qualitatively consistent with early knowledge obtained from intensive observations, but also the downscaling performance with this value is better than or comparable to that with the PAF estimated from dense station data. Finally, it is shown that the PAF estimation, although critical, cannot replace the importance of increasing input station density for downscaling.

Assessment of total and extreme precipitation over central Asia via statistical downscaling: Added value and multi-model ensemble projection

Central Asia(CA)is highly sensitive and vulnerable to changes in precipitation due to global warming,so the projection of precipitation extremes is essential for local climate risk assessment.However,global and regional climate models often fail to reproduce the observed daily precipitation distribution and hence extremes,especially in areas with complex terrain.In this study,we proposed a statistical downscaling(SD)model based on quantile delta mapping to assess and project eight precipitation indices at 73 meteorological stations across CA driven by ERA5 reanalysis data and simulations of 10 global climate models(GCMs)for present and future(2081-2100)periods under two shared socioeconomic pathways(SSP245 and SSP585).The reanalysis data and raw GCM outputs clearly underestimate mean precipitation intensity(SDII)and maximum 1-day precipitation(RX1DAY)and overestimate the number of wet days(R1MM)and maximum consecutive wet days(CWD)at stations across CA.However,the SD model effectively reduces the biases and RMSEs of the modeled precipitation indices compared to the observations.Also it effectively adjusts the distributional biases in the downscaled daily precipitation and indices at the stations across CA.In addition,it is skilled in capturing the spatial patterns of the observed precipitation indices.Obviously,SDII and RX1DAY are improved by the SD model,especially in the southeastern mountainous area.Under the intermediate scenario(SSP245),our SD multi-model ensemble projections project significant and robust increases in SDII and total extreme precipitation(R95PTOT)of 0.5 mm d^(-1) and 19.7 mm,respectively,over CA at the end of the 21st century(2081-2100)compared to the present values(1995-2014).More pronounced increases in indices R95PTOT,SDII,number of very wet days(R10MM),and RX1DAY are projected under the higher emission scenario(SSP585),particularly in the mountainous southeastern region.The SD model suggested that SDII and RX1DAY will likely rise more rapidly than those projected by previous model simulations over CA during the period 2081-2100.The SD projection of the possible future changes in precipitation and extremes improves the knowledge base for local risk management and climate change adaptation in CA.

Projected heat wave increasing trends over China based on combined dynamical and multiple statistical downscaling methods

Extensive investigations on the projection of heat waves(HWs)were conducted on the basis of coarse-resolution global climate models(GCMs).However,these investigations still fail to characterise the future changes in HWs regionally over China.PRECIS dynamical downscaling with a horizontal resolution of 25 km×25 km was employed on the basis of GCM-HadCM3 to provide reliable projections on HWs over the Chinese mainland,and six statistical downscaling methods were used for bias correction under RCP4.5 and RCP8.5 scenarios.The multi-method ensemble(MME)of the top three dynamical downscaling methods with good performance was used to project future changes.Results showed that PRECIS primarily replicated the detailed spatiotemporal pattern of HWs.However,PRECIS overestimated the HWs in the Northwest and Southeast and expanded the areas of HWs in the Northeast and Southwest.Three statistical downscaling methods(quantile mapping,CDF-t and quantile delta mapping)demonstrated good performance in improving PRECIS simulation for reproducing HWs.By contrast,parametric-based trend-preserving approaches such as scaled distribution mapping and ISI-MIP are outperformed by the three aforementioned methods in downscaling HWs,particularly in the high latitudes of China.Based on MME projections,at the end of the 21st century,the national average of the number of HW days each year,the length of the longest HW event in the year and the extreme maximum temperature in HW will increase by 3 times,1 time and 1.3℃,respectively,under the RCP4.5 scenario,whilst that under the RCP8.5 scenario will increase by 8 times,3 times and 3.7℃,respectively,relative to 1986-2005.The Northwest is regionally projected to suffer long and hot HWs,whilst the South and Southeast will experience frequent consecutive HWs.Thus,HWs projected by the combined dynamical and statistical downscaling method are highly reliable in projecting HWs over China.

A Deep Learning Method for Statistical Downscaling of CLDAS Relative Humidity with Different Sources of Data:Sensitivity Analysis

High-resolution relative humidity(RH)data are essential in studies of climate change and in numerical meteorological forecasting.However,because high-resolution meteorological grid data require a large number of stations,the sparse distribution of ground meteorological stations in China before 2008 has limited the development of long-term and high-resolution RH products in the China Meteorological Administration’s Land Assimilation System(CLDAS)dataset.To retrieve high-quality and high-resolution RH data before 2008,we propose a statistical downscaling model(SDM)based on a generative adversarial network(GAN)to transform the original RH data from a resolution of0.05°to 0.01°.The GAN-based SDM(GSDM)is trained with the RH of the CLDAS(0.05°)dataset after 2008 as its input,and the RH of the high-resolution CLDAS(HRCLDAS,0.01°)dataset after 2008 as its target for training.The2-m air temperature data from the HRCLDAS dataset are also included in the input,and the station observations of RH are incorporated in the target for training.To select the optimum data combination for the model,we compared three methods:(1)incorporating without auxiliary data(GSDM),(2)incorporating air temperature as an additional input(GSDM_T),and(3)incorporating air temperature as an additional input and the RH data at stations as an additional target for training(GSDM_TO).Taking the Beijing–Tianjin–Hebei region as an example,we trained the GSDM by using data from 2018 and tested the model performance in 2019.The experimental results showed that the GSDM_TO algorithm achieved the lowest root-mean-square error(3.85%),followed by the GSDM_T(4.01%)and GSDM(4.95%)algorithms.The proposed models showed a competitive performance and captured more local details of the RH fields than other deep learning models and traditional bilinear interpolation.In general,the GSDM_TO algorithm using a combination of different sources of data(air temperature and observed RH)achieved the best results among the various deep learning approaches,indicating that more auxiliary data and more accurate observations are beneficial in downscaling.This may be helpful for the statistical downscaling of other meteorological data.

Assessment of Future Climate Change Scenario in Halaba District, Southern Ethiopia

Climate change is one environmental threat that poses great challenges to the future development prospects of Ethiopia. The study used the statistically downscaled daily data in 30-years intervals from the second generation of the Earth System Model (CanESM2) under two Representative Concentration Pathways (RCPs): RCP 4.5 and RCP 8.5 for three future time slices;near-term (2010-2039), mid-century (2040-2069) and end-century (2071-2099) were generated. The observed data of maximum and minimum temperature and precipitation are a good simulation with the modeled data during the calibration and validation periods using the correlation coefficient (R2), the Nash-Sutcliffe efficiency (NSE), and the Root Mean Square Error (RMSE). The projected annual minimum and maximum temperatures are expected to increase by 0.091°C, 0.517°C, and 0.73°C and 0.072°C, 0.245°C, and 0.358°C in the 2020s, 2050s, and 2080s under the intermediate scenario, respectively. Under RCP8.5, the annual minimum and maximum temperatures are expected to increase by 0.192°C, 0.409°C, and 0.708°C, 0.402°C, 4.352°C, and 8.750°C in the 2020s, 2050s, and 2080s, respectively. Besides, the precipitation is expected to increase under intermediate and high emission scenarios by 1.314%, 7.643%, and 12.239%, and 1.269%, 10.316% and 26.298% in the 2020s, 2050s, and 2080s, respectively. Temperature and precipitation are projected to increase in total amounts under all-time slices and emissions pathways. In both emission scenarios, the greatest changes in maximum temperature, minimum temperature, and precipitation are predicted by the end of the century. This implies climate smart actions in development policies and activities need to consider locally downscale expected climatic changes.

Projection of China's Near- and Long-Term Climate in a New High-Resolution Daily Downscaled Dataset NEX-GDDP

The projection of China＇s near- and long-term future climate is revisited with a new-generation statistically down- scaled dataset, NEX-GDDP （NASA Earth Exchange Global Daily Downscaled Projections）. This dataset presents a high-resolution seamless climate projection from 1950 to 2100 by combining observations and GCM results, and re- markably improves CMIP5 hindcasts and projections from large scale to regional-to-local scales with an unchanged long-term trend. Three aspects are significantly improved： （1） the climatology in the past as compared against the ob- servations; （2） more reliable near- and long-term projections, with a modified range of absolute value and reduced inter-model spread as compared to CMIP5 GCMs; and （3） much added value at regional-to-local scales compared to GCM outputs. NEX-GDDP has great potential to become a widely-used high-resolution dataset and a benchmark of modem climate change for diverse earth science communities.