Browse

You are looking at 1 - 10 of 2,721 items for :

  • Weather and Forecasting x
  • User-accessible content x
Clear All
Akshay Deoras, Kieran M. R. Hunt, and Andrew G. Turner

Abstract

This study analyzes the prediction of Indian monsoon low pressure systems (LPSs) on an extended time scale of 15 days by models of the Subseasonal-to-Seasonal (S2S) prediction project. Using a feature-tracking algorithm, LPSs are identified in 11 S2S models during a common reforecast period of June–September 1999–2010, and then compared with 290 and 281 LPSs tracked in ERA-Interim and MERRA-2 reanalysis datasets. The results show that all S2S models underestimate the frequency of LPSs. They are able to represent transits, genesis, and lysis of LPSs; however, large biases are observed in the Australian Bureau of Meteorology, China Meteorological Administration (CMA), and Hydrometeorological Centre of Russia (HMCR) models. The CMA model exhibits large LPS track position error and the intensity of LPSs is overestimated (underestimated) by most models when verified against ERA-Interim (MERRA-2). The European Centre for Medium-Range Weather Forecasts and Met Office models have the best ensemble spread–error relationship for the track position and intensity, whereas the HMCR model has the worst. Most S2S models are underdispersive—more so for the intensity than the position. We find the influence of errors in the LPS simulation on the pattern of total precipitation biases in all S2S models. In most models, precipitation biases increase with forecast lead time over most of the monsoon core zone. These results demonstrate the potential for S2S models at simulating LPSs, thereby giving the possibility of improved disaster preparedness and water resources planning.

Open access
Toshichika Iizumi, Yonghee Shin, Jaewon Choi, Marijn van der Velde, Luigi Nisini, Wonsik Kim, and Kwang-Hyung Kim

Abstract

Forecasting global food production is of growing importance in the context of globalizing food supply chains and observed increases in the frequency of climate extremes. The National Agriculture and Food Research Organization–Asia-Pacific Economic Cooperation Climate Center (NARO-APCC) Crop Forecasting Service provides yield forecasts for global cropland on a monthly basis using seasonal temperature and precipitation forecasts as the main inputs, and 1 year of testing the operation of the service was recently completed. Here we evaluate the forecasts for the 2019 yields of major commodity crops by comparing with the reported yields and forecasts from the European Commission’s Joint Research Centre (JRC) and the U.S. Department of Agriculture (USDA). Forecasts for maize, wheat, soybean, and rice were evaluated for 20 countries located in the Northern Hemisphere, including 39 crop-producing states in the United States, for which 2019 reported yields were already publicly available. The NARO-APCC forecasts are available several months earlier than the JRC and USDA forecasts. The skill of the NARO-APCC forecasts was good in absolute terms, but the forecast errors in the NARO-APCC forecasts were almost always larger than those of the JRC and USDA forecasts. The forecast errors in the JRC and USDA forecasts decreased as the harvest approached, whereas those in the NARO-APCC forecasts were rather stable over the season, with some exceptions. Although this feature seems to be a disadvantage, it may turn into an advantage if skillful forecasts are achievable in the earlier stages of a season. We conclude by discussing relative advantages and disadvantages and potential ways to improve global yield forecasting.

Open access
Jingzhuo Wang, Jing Chen, Hanbin Zhang, Hua Tian, and Yining Shi

Abstract

Ensemble forecasting is a method to faithfully describe initial and model uncertainties in a weather forecasting system. Initial uncertainties are much more important than model uncertainties in the short-range numerical prediction. Currently, initial uncertainties are described by the ensemble transform Kalman filter (ETKF) initial perturbation method in Global and Regional Assimilation and Prediction Enhanced System–Regional Ensemble Prediction System (GRAPES-REPS). However, an initial perturbation distribution similar to the analysis error cannot be yielded in the ETKF method of the GRAPES-REPS. To improve the method, we introduce a regional rescaling factor into the ETKF method (we call it ETKF_R). We also compare the results between the ETKF and ETKF_R methods and further demonstrate how rescaling can affect the initial perturbation characteristics as well as the ensemble forecast skills. The characteristics of the initial ensemble perturbation improve after applying the ETKF_R method. For example, the initial perturbation structures become more reasonable, the perturbations are better able to explain the forecast errors at short lead times, and the lower kinetic energy spectrum as well as perturbation energy at the initial forecast times can lead to a higher growth rate of themselves. Additionally, the ensemble forecast verification results suggest that the ETKF_R method has a better spread–skill relationship, a faster ensemble spread growth rate, and a more reasonable rank histogram distribution than ETKF. Furthermore, the rescaling has only a minor impact on the assessment of the sharpness of probabilistic forecasts. The above results all suggest that ETKF_R can be effectively applied to the operational GRAPES-REPS.

Open access
Evan A. Kalina, Isidora Jankov, Trevor Alcott, Joseph Olson, Jeffrey Beck, Judith Berner, David Dowell, and Curtis Alexander

Abstract

The High-Resolution Rapid Refresh Ensemble (HRRRE) is a 36-member ensemble analysis system with 9 forecast members that utilizes the Advanced Research version of the Weather Research and Forecasting (ARW-WRF) dynamic core and the physics suite from the operational Rapid Refresh/High-Resolution Rapid Refresh deterministic modeling system. A goal of HRRRE development is a system with sufficient spread among members, comparable in magnitude to the random error in the ensemble mean, to represent the range of possible future atmospheric states. HRRRE member diversity has traditionally been obtained by perturbing the initial and lateral boundary conditions of each member, but recent development has focused on implementing stochastic approaches in HRRRE to generate additional spread. These techniques were tested in retrospective experiments and in the May 2019 Hazardous Weather Testbed Spring Experiment (HWT-SE). Results show a 6%–25% increase in the ensemble spread in 2-m temperature, 2-m mixing ratio, and 10-m wind speed when stochastic parameter perturbations are used in HRRRE (HRRRE-SPP). Case studies from HWT-SE demonstrate that HRRRE-SPP performed similar to or better than the operational High-Resolution Ensemble Forecast system, version 2 (HREFv2), and the nonstochastic HRRRE. However, subjective evaluations provided by HWT-SE forecasters indicated that overall, HRRRE-SPP predicted lower probabilities of severe weather (using updraft helicity as a proxy) compared to HREFv2. A statistical analysis of the performance of HRRRE-SPP and HREFv2 from the 2019 summer convective season supports this claim, but also demonstrates that the two systems have similar reliability for prediction of severe weather using updraft helicity.

Open access
Julia Jeworrek, Gregory West, and Roland Stull

Abstract

Physics parameterizations in the Weather Research and Forecasting (WRF) Model are systematically varied to investigate precipitation forecast performance over the complex terrain of southwest British Columbia (BC). Comparing a full year of modeling data from over 100 WRF configurations to station observations reveals sensitivities of precipitation intensity, season, location, grid resolution, and accumulation window. The choice of cumulus and microphysics parameterizations is most important. The WSM5 microphysics scheme yields competitive verification scores when compared to more sophisticated and computationally expensive parameterizations. Although the scale-aware Grell–Freitas cumulus parameterization performs better for summertime convective precipitation, the conventional Kain–Fritsch parameterization better simulates wintertime frontal precipitation, which contributes to the majority of the annual precipitation in southwest BC. Finer grid spacings have lower relative biases and a more realistic spread in precipitation intensity distribution, yet higher relative standard deviations of their errors—they produce finer spatial differences and local extrema. Finer resolutions produce the best fraction of correct-to-incorrect forecasts across all precipitation intensities, whereas the coarser 27-km domain yields the highest hit rates and equitable threat scores. Verification metrics improve greatly with longer accumulation windows—hourly precipitation values are prone to double-penalty issues, while longer accumulation windows compensate for timing errors but lose information about short-term precipitation intensities. This study provides insights regarding WRF precipitation performance in complex terrain across a wide variety of configurations, using metrics important to a range of end users.

Open access
William R. Burrows and Curtis J. Mooney

Abstract

Blizzard conditions occur regularly in the Canadian Arctic, with high impact on travel and life there. These extreme conditions are challenging to forecast for this vast domain because the observation network is sparse and remote sensing coverage is limited. To establish occurrence statistics we analyzed METeorological Aerodrome Reports (METARs) from Canadian Arctic stations between October and May 2014-2018. Blizzard conditions occur most frequently in open tundra east and north of the boreal forest boundary, with highest frequency found on the northwest side of Hudson Bay and over flat terrain in central Baffin Island. Except in sheltered locations, the reported cause of reduced visibility is blowing snow without precipitating snow in about one-half to two-thirds of METARs made by a human observer, even higher at some stations.

We produce three products that forecast blizzard conditions from post-processed NWP model output. The blizzard potential (BP), generated from expert’s rules, is intended for warning well in advance of areas where blizzard conditions may develop. A second product (BH) stems from regression equations for the probability of visibility ≤ 1 km in blowing snow and/or concurrent snow derived by Baggaley and Hanesiak (2005). A third product (RF), generated with the Random Forest ensemble classification algorithm, makes a consensus YES/NO forecast for blizzard conditions. We describe the products, provide verification, and show forecasts for a significant blizzard event. Receiver Operator Characteristic curves and critical success index scores show RF forecasts have greater accuracy than BP and BH forecasts at all lead times.

Open access
Akira Yamazaki, Takemasa Miyoshi, Jun Inoue, Takeshi Enomoto, and Nobumasa Komori

Abstract

An ensemble-based forecast sensitivity to observations (EFSO) diagnosis has been implemented in an atmospheric general circulation model–ensemble Kalman filter data assimilation system to estimate the impacts of specific observations from the quasi-operational global observing system on weekly short-range forecasts. It was examined whether EFSO reasonably approximates the impacts of a subset of observations from specific geographical locations for 6-hour forecasts, and how long the 6-hour observation impacts can be retained during the 7-day forecast period. The reference for these forecasts was obtained from 12 data denial experiments in each of which a subset of three radiosonde observations launched from a geographical location was excluded. The 12 locations were selected from three latitudinal bands comprising (i) four Arctic regions, (ii) four midlatitude regions in the Northern Hemisphere, and (iii) four tropical regions during the Northern Hemisphere winter of 2015/16. The estimated winter-averaged EFSO-derived observation impacts well corresponded to the 6-hour observation impacts obtained by the data denials and EFSO could reasonably estimate the observation impacts by the data denials on short-range (6-hour to 2-day) forecasts. Furthermore, during the medium-range (4-day to 7-day) forecasts, it was found that the Arctic observations tend to seed the broadest impacts and their short-range observation impacts could be projected to beneficial impacts in Arctic and midlatitude North American areas. The midlatitude area located just downstream of dynamical propagation from the Arctic toward the midlatitudes. Results obtained by repeated Arctic data-denial experiments were found to be generally common to those from the non-repeated experiments.

Open access
Regula Keller, Jan Rajczak, Jonas Bhend, Christoph Spirig, Stephan Hemri, Mark A. Liniger, and Heini Wernli

Abstract

Statistical postprocessing is applied in operational forecasting to correct systematic errors of numerical weather prediction models (NWP) and to automatically produce calibrated local forecasts for end-users. Postprocessing is particularly relevant in complex terrain, where even state-of-the-art high-resolution NWP systems cannot resolve many of the small-scale processes shaping local weather conditions. In addition, statistical postprocessing can also be used to combine forecasts from multiple NWP systems. Here we assess an ensemble model output statistics (EMOS) approach to produce seamless temperature forecasts based on a combination of short-term ensemble forecasts from a convection-permitting limited-area ensemble and a medium-range global ensemble forecasting model. We quantify the benefit of this approach compared to only postprocessing the high-resolution NWP. The multi-model EMOS approach (’Mixed EMOS’) is able to improve forecasts by 30% with respect to direct model output from the high-resolution NWP. A detailed evaluation of Mixed EMOS reveals that it outperforms either one of the single-model EMOS versions by 8-12%. Temperature forecasts at valley locations profit in particular from the model combination. All forecast variants perform worst in winter (DJF), however calibration and model combination improves forecast quality substantially. In addition to increasing skill as compared to single model postprocessing, it also enables to seamlessly combine multiple forecast sources with different time horizons (and horizontal resolutions) and thereby consolidates short-term to medium-range forecasting time horizons in one product without any user-relevant discontinuity.

Open access
Theodore W. Letcher, Sandra L. LeGrand, and Christopher Polashenski

Abstract

Blowing snow presents substantial risk to human activities by causing severe visibility degradation and snow drifting. Furthermore, blowing snow presents a weather forecast challenge since it is not generally simulated in operational weather forecast models. In this study, we apply a physically based blowing snow model as a diagnostic overlay to output from a reforecast WRF simulation of a significant blowing snow event that occurred over the northern Great Plains of the United States during the winter of 2019. The blowing snow model is coupled to an optics parameterization that estimates the visibility reduction by blowing snow. This overlay is qualitatively evaluated against false color satellite imagery from the GOES-16 operational weather satellite and available surface visibility observations. The WRF-simulated visibility is substantially improved when incorporating blowing snow hydrometeors. Furthermore, the model-simulated plume of blowing snow roughly corresponds to the blowing snow plumes visible in the satellite imagery. Overall, this study illustrates how a blowing snow diagnostic model can aid weather forecasters in making blowing snow visibility forecasts, and demonstrates how the model can be evaluated against satellite imagery.

Open access
Florian Dupuy, Olivier Mestre, Mathieu Serrurier, Valentin Kivachuk Burdá, Michaël Zamo, Naty Citlali Cabrera-Gutiérrez, Mohamed Chafik Bakkay, Jean-Christophe Jouhaud, Maud-Alix Mader, and Guillaume Oller

Abstract

Cloud cover provides crucial information for many applications such as planning land observation missions from space. It remains, however, a challenging variable to forecast, and numerical weather prediction (NWP) models suffer from significant biases, hence, justifying the use of statistical postprocessing techniques. In this study, ARPEGE (Météo-France global NWP) cloud cover is postprocessed using a convolutional neural network (CNN). CNN is the most popular machine learning tool to deal with images. In our case, CNN allows the integration of spatial information contained in NWP outputs. We use a gridded cloud cover product derived from satellite observations over Europe as ground truth, and predictors are spatial fields of various variables produced by ARPEGE at the corresponding lead time. We show that a simple U-Net architecture (a particular type of CNN) produces significant improvements over Europe. Moreover, the U-Net outclasses more traditional machine learning methods used operationally such as a random forest and a logistic quantile regression. When using a large number of predictors, a first step toward interpretation is to produce a ranking of predictors by importance. Traditional methods of ranking (permutation importance, sequential selection, etc.) need important computational resources. We introduced a weighting predictor layer prior to the traditional U-Net architecture in order to produce such a ranking. The small number of additional weights to train (the same as the number of predictors) does not impact the computational time, representing a huge advantage compared to traditional methods.

Open access