Machine-Learning-Based Downscaling of Hourly ERA5-Land Air Temperature over Mountainous Regions
نویسندگان
چکیده
In mountainous regions, the scarcity of air temperature (Ta) measurements is a major limitation for hydrological and crop monitoring. An alternative to in situ could be downscale reanalysis Ta data provided at high-temporal resolution. However, relatively coarse spatial resolution these products (i.e., 9 km ERA5-Land) unlikely directly representative actual local patterns. To address this issue, study presents new downscaling strategy hourly ERA5-Land with three-step procedure. First, ERA5 corrected its original by using reference derived from elevation grid an estimate over area Environmental Lapse Rate (ELR). Such correction trained several machine learning techniques, including Multiple Linear Regression (MLR), Support Vector (SVR), Extreme Gradient Boosting (Xgboost), as well ancillary (daily mean, standard deviation, ELR, elevation). Next, algorithms are run correct Ta, used derive updated ELR (without measurements). Third, disaggregate 30-meter SRTM’s Digital Elevation Model (DEM). The effectiveness method assessed across northern part High Atlas Mountains central Morocco through (1) k-fold cross-validation against five years (2016 2020) readings (2) comparison classical methods based on constant ELR. Our results indicate significant enhancement distribution Ta. By comparing our model, which included Xgboost, SVR, MLR, ELR-based approach, we were able decrease regional root mean square error approximately 3 ∘C 1.61 ∘C, 1.75 1.8 reduce bias −0.5 null, increase coefficient determination 0.88 0.97, 0.96, 0.96 respectively.
منابع مشابه
Improving Land Surface Temperature Retrievals over Mountainous Regions
Algorithms for Land Surface Temperature (LST) retrieval from infrared measurements are usually sensitive to the amount of water vapor present in the atmosphere. The Satellite Application Facilities on Climate Monitoring and Land Surface Analysis (CM SAF and LSA SAF) are currently compiling a 25 year LST Climate data record (CDR), which uses water vapor information from ERA-Int reanalysis. Howev...
متن کاملIndirect downscaling of hourly precipitation based on atmospheric circulation and temperature
The main source of information on future climate conditions are global circulation models (GCMs). While the various GCMs agree on an increase of surface temperature, the predictions for precipitation exhibit high spread among the models, especially in shorter-than-daily temporal resolution. This paper presents a method to predict regional distributions of the hourly rainfall depth based on dail...
متن کاملDownscaling Snow Cover Fraction Data in Mountainous Regions Based on Simulated Inhomogeneous Snow Ablation
High-resolution snow distributions are essential for studying cold regions. However, the temporal and spatial resolutions of current remote sensing snow maps remain limited. Remotely sensed snow cover fraction (SCF) data only provide quantitative descriptions of snow area proportions and do not provide information on subgrid-scale snow locations. We present a downscaling method based on simulat...
متن کاملComparison of Different Machine Learning Approaches for Monthly Satellite-Based Soil Moisture Downscaling over Northeast China
Although numerous satellite-based soil moisture (SM) products can provide spatiotemporally continuous worldwide datasets, they can hardly be employed in characterizing fine-grained regional land surface processes, owing to their coarse spatial resolution. In this study, we proposed a machine-learning-based method to enhance SM spatial accuracy and improve the availability of SM data. Four machi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Atmosphere
سال: 2023
ISSN: ['2073-4433']
DOI: https://doi.org/10.3390/atmos14040610