Tag: Climate Change Downscaling Techniques

Abstract

The diurnal cycle of rainfall in the Lake Victoria Basin of East Africa results from the super positioning of regional circulations driven by lake/land temperature differences and topography in the presence of the large-scale flow. Analysis of a triple-nested regional model simulation with a convective-permitting inner domain shows how these elements combine to produce the observed regionality of precipitation, including diurnal cycling, in boreal fall. A single diurnal rainfall peak occurs throughout the basin, but the time of maximum rainfall varies within the basin. The rainy period over the lake begins with precipitation over the northern part of the lake near 02Z (0500 LT), so it is not simply nighttime rain driven by lake/land breezes. Onset of the rainy period occurs only when low geopotential heights over the relatively warm lake cause a southward branch of the Turkana Jet to form. The formation of the jet depends on nighttime cooling over the Eastern Rift Mountains, which acts to direct the large-scale, moist flow around the topography. Topography also plays a role in the daytime rainy period over land in the Lake Victoria Basin. Moist divergence over the lake supports convergence and precipitation over the shore regions. Precipitation rates are twice the magnitude over the eastern shore compared with the western shore because daytime warming of the high elevations of the Eastern Rift Mountains allows the large-scale easterly flow to go over the mountains. This easterly flow converges with the lake/land circulation and doubles precipitation rates over the eastern shore.

Abstract

This study assesses the performance of 37 global climate models (GCMs) from the Coupled Model Inter-comparison Project Phase 6 (CMIP6) in simulating extreme precipitation in Madagascar. In this study, six extreme precipitation indices from the Expert Team on Climate Change Detection and Indices (ETCCDI) were used, namely consecutive dry days (CDD), heavy precipitation days (R10MM), very heavy precipitation days (R20MM), maximum 5-day precipitation (RX5DAY), extremely wet days (R99P), and simple daily intensity (SDII). The performance of the model was evaluated from 1998 to 2014 against the Tropical Rainfall Measuring Mission 3B42 (TRMM) and the Global Precipitation Climatology Project (GPCP). The results show that most of the models in CMIP6 reasonably reproduce the annual precipitation cycle of the study area. The results also suggested that models from CMIP6 tend to underestimate extreme precipitation indices such as CDD, R20MM, SDII, and R99P. However, for R10MM and RX5DAY, the performance of individual models varies remarkably. By looking at the performance metrics, not a single model consistently performed well. Model performance changes with the reference data, the extreme precipitation indices, and the performance metrics. However, the findings of this study indicate that overall, multi-model ensemble mean outperforms most individual models. The study lays the basis for the analysis and projection of future extreme precipitation in Madagascar but also provides scientific evidence for the choice of models.

Abstract

High-resolution weather data is crucial for assessing future climate change impacts on local environments, yet downscaling low-resolution Global Climate Models (GCMs) outputs and addressing associated uncertainty remain significant challenges. In this study, we propose a novel spatial weather generator using generative networks, specifically a numerical conditional deep convolutional generative adversarial network (cDCGAN), as a promising solution. The cDCGAN generates high-resolution weather data from low-resolution GCM outputs and was applied to four case areas in China under four Shared Socio-economic Pathway (SSP) scenarios. The results demonstrate the cDCGAN's accuracy, consistency, and stability, with low uncertainties. The model performs optimally in low-elevation plains and tropical regions. The cDCGAN offers advantages in uncertainty analysis over traditional downscaling methods, serving as a valuable tool for climate change analysis, response estimation, and environmental management decision-making within the spatial statistics domain.

Abstract

The Indo-Gangetic Plain (IGP), the source of grains for around 40% Indian population, is known as the breadbasket of India. The Indian Summer Monsoon Rainfall (ISMR) plays a vital role in the agricultural activities in this region. The rapid urbanization, land use and land cover change have significantly impacted the region’s agriculture, water resources, and socioeconomic facets. The present study has investigated the observed and regional modeling aspects of ISMR characteristics, associated extremes over the IGP, and future perspectives under the high-emission RCP8.5-scenario. Future projections suggest a 10–20% massive decrease during pre-monsoon (March–May) and earlier ISM season months (i.e., June and July). A significant 40–70% decline in mean monsoon rainfall during the June–July months in the near future (NF; 2041–2060) has been projected compared to the historical period (1986–2005). An abrupt increase of 80–170% in mean monsoon rainfall during the post-monsoon (October–December) in the far future (FF; 2080–2099) is also projected. The distribution of projected extreme rainfall events shows a decline in moderate or rather heavy events (5 or more) in NF and FF. Further, an increase in higher rainfall category events such as very heavy (5–10) and extremely heavy rainfall (5 or more) events in NF and FF under the warmer climate is found. However, the changes are less prominent during FF compared to the NF. The mean thresholds for extremely heavy rainfall may increase by 1.9–4.9% during NF and FF. Further, the evolution patterns of various quantities, such as tropospheric temperature gradient (TG), specific humidity, and mean sea level pressure, have been analyzed to understand the physical processes associated with rainfall extremes. The strengthening in TG and enhanced atmospheric moisture content in NF and FF support the intensification in projected rainfall extremes over IGP.

Abstract

We investigate the influence of bias correction of Global Climate Models (GCMs) prior to dynamical downscaling using regional climate models (RCMs), on the change in climate projected. We use 4 GCMs which are bias corrected against ERA-Interim re-analysis as a surrogate truth, and carry out bias corrected and non-bias corrected simulations over the CORDEX Australasia domain using the Weather Research and Forecasting model. Our results show that when considering the effect of bias correction on current and future climate separately, bias correction has a large influence on precipitation and temperature, especially for models which are known to have large biases. However, when considering the change in climate, i.e the \(\Delta\) change (future minus current), we found that while differences between bias-corrected and non-corrected RCM simulations can be substantial (e.g. more than \(1\,^\circ\) C for temperatures) these differences are generally smaller than the models’ inter-annual variability. Overall, averaged across all variables, bias corrected boundary conditions produce an overall reduction in the range, standard deviation and mean absolute deviation of the change in climate projected by the 4 models tested, over 61.5%, 62% and 58% of land area, with a larger reduction for precipitation as compared to temperature indices. In addition, we show that changes in the \(\Delta\) change for DJF tasmax are broadly linked to precipitation changes and consequently soil moisture and surface sensible heat flux and changes in the \(\Delta\) changefor JJA tasmin are linked to downward longwave heat flux. This study shows that bias correction of GCMs against re-analysis prior to dynamical downscaling can increase our confidence in projected future changes produced by downscaled ensembles.

Abstract

This works proposes a probabilistic framework for rainy season onset forecasts over Greater Horn of Africa derived from bias-corrected, long range, multi-model ensemble precipitation forecasts. A careful analysis of the contribution of the different forecast systems to the overall multi-model skill shows that the improvement over the best performing individual model can largely be explained by the increased ensemble size. An alternative way of increasing ensemble size by blending a single model ensemble with climatology is explored and demonstrated to yield better probabilistic forecasts than the multi-model ensemble. Both reliability and skill of the probabilistic forecasts are better for OND onset than for MAM and JJAS onset where forecasts are found to be late biased and have only minimal skill relative to climatology. The insights gained in this study will help enhance operational subseasonal-to-seasonal forecasting in the GHA region.

Abstract

Background

Cross-landscape fuel moisture content is highly variable but not considered in existing fire danger assessments. Capturing fuel moisture complexity and its associated controls is critical for understanding wildfire behavior and danger in emerging fire-prone environments that are influenced by local heterogeneity. This is particularly true for temperate heathland and peatland landscapes that exhibit spatial differences in the vulnerability of their globally important carbon stores to wildfire. Here we quantified the range of variability in the live and dead fuel moisture of Calluna vulgaris across a temperate fire-prone landscape through an intensive fuel moisture sampling campaign conducted in the North Yorkshire Moors, UK. We also evaluated the landscape (soil texture, canopy age, aspect, and slope) and micrometeorological (temperature, relative humidity, vapor pressure deficit, and windspeed) drivers of landscape fuel moisture variability for temperate heathlands and peatlands for the first time.

Results

We observed high cross-landscape fuel moisture variation, which created a spatial discontinuity in the availability of live fuels for wildfire spread (fuel moisture < 65%) and vulnerability of the organic layer to smoldering combustion (fuel moisture < 250%). This heterogeneity was most important in spring, which is also the peak wildfire season in these temperate ecosystems. Landscape and micrometeorological factors explained up to 72% of spatial fuel moisture variation and were season- and fuel-layer-dependent. Landscape factors predominantly controlled spatial fuel moisture content beyond modifying local micrometeorology. Accounting for direct landscape–fuel moisture relationships could improve fuel moisture estimates, as existing estimates derived solely from micrometeorological observations will exclude the underlying influence of landscape characteristics. We hypothesize that differences in soil texture, canopy age, and aspect play important roles across the fuel layers examined, with the main differences in processes arising between live, dead, and surface/ground fuels. We also highlight the critical role of fuel phenology in assessing landscape fuel moisture variations in temperate environments.

Conclusions

Understanding the mechanisms driving fuel moisture variability opens opportunities to develop locally robust fuel models for input into wildfire danger rating systems, adding versatility to wildfire danger assessments as a management tool.

Abstract

Large-scale temperature projections need to be downscaled to river basin scale to facilitate a regional scale climate change impact assessment. A multi-stage statistical downscaling procedure is proposed in the current study, the first stage captures the climate change signals from the simulations of general circulation models (GCMs) by spatially downscaling the monthly GCM simulations. The second stage disaggregates the spatially downscaled monthly series to a daily scale by a weather generator which adds the regional climatic information into the spatially downscaled time series. A distribution-free post-processing shuffling is finally performed to rebuild the intervariable correlation of downscaled temperatures with regional rainfall which is important in reliable projection of streamflow. The procedure is validated by downscaling the maximum and minimum temperatures over the Bharathapuzha catchment in India for the period 1951–2005. The downscaled series of temperature shows Normalised Root Mean Square Error (NRMSE) less than 0.09 and correlation coefficients greater than 0.4. The ability of the procedure in capturing non-stationarity in the climate is also analysed by its performance in different phases of ENSO.

Abstract

Social science often relies on surveys of households and individuals. Dozens of such surveys are regularly administered by the U.S. government. However, they field independent, unconnected samples with specialized questions, limiting research questions to those that can be answered by a single survey. The presented data comprise the fusion onto the American Community Survey (ACS) microdata of select donor variables from the Residential Energy Consumption Survey (RECS) of 2015, the National Household Travel Survey (NHTS) of 2017, the American Housing Survey (AHS) of 2019, and the Consumer Expenditure Survey - Interview (CEI) for the years 2015–2019. This results in an integrated microdataset of household attributes and well-being dimensions that can be analyzed to address research questions in ways that are not currently possible. The underlying statistical techniques, designed under the fusionACS project, are included in an open-source R package, fusionModel, that provides generic tools for the creation, analysis, and validation of fused microdata.

Abstract

Three cumulus and five microphysics parameterization schemes of the Weather Research and Forecasting model (WRF) are the basis for simulating ten specific meteorological events of the Paraíba do Sul River Basin (PSRB) in Southeast Brazil. The cases studied are frontal wave systems, thermodynamic instability, and the South Atlantic Convergence Zone (SACZ). Each parameterization combination generated 15 simulations for each event, resulting in 150 tests. The primary domain has a horizontal resolution of 8.0 km and the nested 2.6 km resolution. Three analysis tools underlie the study: (i) punctual verification of the first 24 h of precipitation forecast, using the Taylor diagram; (ii) verification of the prediction of precipitation using categorical binary variable and (iii) the Model´s ability to reproduce patterns of the spatial distribution of precipitation. The Taylor diagram suggests that the combination of the Morrison Double moment and Multiscale Kain–Fritsch schemes produce the best results. The categorical verification indicates that, for dynamic/convective events, the Morrison Double moment and Multiscale Kain–Fritsch and WRF Double Moment 6–class sets showed the best indices. Some configurations presented reliable results for exclusively convective events, and WRF Single–moment 6–class and Grell–Freitas Ensemble is the best combination. The Morrison Double moment and Multiscale Kain–Fritsch parameterizations yielded the best performance for the spatial distribution. Overall, the schemes tested perform better for the upstream region, i.e., the area of greater water uptake for the basin.