Predicting Local Climate Change in Snow-Covered Regions with High-Resolution Regional Climate Models
Year: 2022
Type: dissertation
Abstract: Global warming has become apparent in the current climate and affected the extreme events such as heavy rainfall.In the future, global warming will continue and accelerate depending on the emission scenario.In order to take countermeasures against global warming, it is important to assess the future changes in regional climates under the much warmer condition.The Meteorological Research Institute (MRI) has provided several datasets based on future climate simulations using the Meteorological Research Institute atmospheric general circulation model (MRI-AGCM) and the Non-Hydrostatic Regional Climate Model (NHRCM).These datasets can be used for studying changes in meteorological phenomena in the warmer climate.The author has involved in the designing and conducting future climate simulation using NHRCM for creating the datasets.Also, the author has analyzed the datasets and clarified the mechanism of future changes in air temperature, precipitation, wind, and snow cover in Japan.It is important to know the response of surface air temperature (SAT) and precipitation to global warming on a regional scale over Japan.The database for policy making on future climate change (d4PDF) is a dataset of large ensemble climate experiments and used for analyzing future climate changes in SAT and precipitation.However, the d4PDF only assumes +2K and +4K climate compared to the pre-industrial period.We conducted additional ensemble experiment when the global mean temperature increases by 1.5 K using the NHRCM with a 20 km grid spacing, which is a subset of the d4PDF.The reproducibility of the current climate experiment of d4PDF is sufficient to study the future climate change over Japan, and the dynamic downscaling using NHRCM from the global climate model improves the frequency of strong daily precipitation.Comparing the present-climate, +1.5K +2K, and +4K experiments, the SAT over Japan will increase linearly and faster than the global mean SAT.As warming progresses, the contrast in SST increase over time will become more pronounced.The increase in temperature due to global warming in the annual minimum SAT is larger than that in the average SAT and also that in the annual maximum SAT.Winter precipitation decreases and increases linearly in western and eastern Japan, respectively, reflecting the weakening of the winter monsoon.The annual maximum of daily precipitation (R1d) increases almost linearly with the rise in global SAT, whereas annual precipitation remains almost unchanged.The number of days with strong precipitation will increase as temperatures rise.The number of days without precipitation also increases with increasing temperature.The pattern of precipitation change in the Japanese Islands is highly correlated with that of change in annual mean precipitation and seasonal precipitation excluding summer.The pattern of precipitation change was shown to be responsive to the defined future trend pattern of sea surface temperature.The number of days of snow cover decreases with increasing global temperature at a rate greater than a linear trend.When changes in the global mean temperature range from +1.5K to +2K, R1d increases by 2.7% in the Japanese Islands, although it varies with region.In northern Japan, where the increase in SAT is ii large, the increase in R1d is 5%.The model with 20 km horizontal grid spacing is a coarse resolution for the representation of snow depth because this resolution cannot adequately represent the elevation of each point, making it difficult to project snow depth, which is highly dependent on elevation.NHRCM with 5 km horizontal resolution was used to simulate snow depth in Japan and project future changes in snow depth.The simulated snow depths had biases even at 5 km resolution, and bias correction was necessary to accurately project the future snow depths.Therefore, we developed a new method of bias correction that can produce accurate snow depth and can be easily automated on a computer.Three classification methods of regional frequency analysis were tested in nine regions of Japan; the classification method based on the second order of L-moment (L-cv) was the best bias correction method among those tested.We confirmed that this bias correction is a useful method for projecting future climate, considering estimates from the test sample.In the future climate, the snow depth was projected to decrease by about 50 cm, and the average snow depth in Japan was projected to be about 30 cm.The decrease in maximum snow depth was projected to be larger in Nagano and Gifu and smaller in Hokkaido than in other regions.The 5 km horizontal resolution model had a snow cover bias and was used for future projections after correcting for data from AMeDAS sites.The 2 km resolution model can reproduce the terrain better and represent convection, which improve the reproduction of weather elements and snow cover.We use the NHRCM with 2 km horizontal resolution to investigate the effect of reduced snow cover on near-surface wind speeds over complex terrain in Japan under a global warming scenario.On synoptic scales, global warming weakens the winter monsoon, which in turn weakens near-surface winds around Japan.Global warming due to climate change reduces Japan's snowpack.Reductions in snow cover drive localized acceleration of rising surface temperatures and prevent the formation of a stable atmospheric boundary layer.When the stable boundary layer collapses, the wind speed and temperature near the surface increase due to the enhancement of momentum and heat exchange with the upper atmosphere.This effect is stronger at low elevations.However, this situation does not occur at high elevations where there is no stable layer under current climate conditions, despite increased ground warming due to reduced snow cover under global warming conditions.An originally stable near-surface atmospheric layer under current climate conditions can therefore influence future changes in wind speed and surface air temperature in snow-covered areas.This mechanism is also confirmed by current climatic observations, which show that wind speeds in the absence of the winter monsoon are stronger on days when there is no snow cover rather than on days with snow cover.This study, using a regional climate model, has contributed to our understanding of climate change in Japan.The datasets used in these studies were provided to the "JMA Global Warming Projection No. 9" published by the Japan Meteorological Agency.It has also been used for flood control planning by the Ministry of Land, Infrastructure, Transport and Tourism.In "Climate Change in Japan 2020" iii published by the Ministry of Education, Culture, Sports, Science and Technology and the Japan Meteorological Agency, in addition to the utility of the dataset, some of the research results were published.Biases in regional climate model simulations has been reduced significantly with the progress of horizontal resolution.Reducing biases and improving the reproducibility are issues that need to be solved continuously because biases still exist.In addition, it is important to proceed with calculations with increasing ensembles to address uncertainties due to models and SSTs.
Author: Nosaka, Masaya
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
Keywords: regional climate change, snow-covered regions, high-resolution regional climate modeling, d4PDF, global warming, surface air temperature, precipitation change, daily precipitation extremes, winter monsoon, snow cover, Japan, dry days, dynamical downscaling, emission scenarios, heavy rainfall
SDG: Climate action
Open Access status: green
OpenAlex ID: W7207963424
