Seasonal Prediction Skill Research Articles

Using Convolutional Neural Network to Emulate Seasonal Tropical Cyclone Activity

AbstractIt has been widely recognized that tropical cyclone (TC) genesis requires favorable large‐scale environmental conditions. Based on these linkages, numerous efforts have been made to establish an empirical relationship between seasonal TC activities and large‐scale environmental favorability in a quantitative way, which lead to conceptual functions such as the TC genesis index. However, due to the limited amount of reliable TC observations and complexity of the climate system, a simple analytic function may not be an accurate portrait of the empirical relationship between TCs and their ambiences. In this research, we use convolution neural networks (CNNs) to disentangle this complex relationship. To circumvent the limited amount of seasonal TC observation records, we implement transfer‐learning technique to train ensemble of CNNs first on suites of high‐resolution climate model simulations with realistic seasonal TC activities and large‐scale environmental conditions, and then on a state‐of‐the‐art reanalysis from 1950 to 2019. The trained CNNs can well reproduce the historical TC records and yields significant seasonal prediction skills when the large‐scale environmental inputs are provided by operational climate forecasts. Furthermore, by inputting the ensemble CNNs with 20th century reanalysis products and Phase 6 of the Coupled Model Intercomparison Project (CMIP6) simulations, we investigated TC variability and its changes in the past and future climates. Specifically, our ensemble CNNs project a decreasing trend of global mean TC activity in the future warming scenario, which is consistent with our future projections using high‐resolution climate model.

A Statistical Prediction Model for Summer Precipitation in China Based on TSD Method and EOF Modes’ Time Coefficients

It is a challenge to improve the skill of seasonal precipitation prediction, because there are many factors affecting summer precipitation in China, which are found on different time scales and have complex interactions with each other. For these reasons, we establish a prediction model with the time-scale decomposition (TSD) method to investigate whether the TSD has an improving effect on the prediction skill of summer precipitation in China. Using this statistical model, the predictors and predictands will be separated into interannual and interdecadal time scales, after which Empirical Orthogonal Function (EOF) decomposition is performed on these two components, and their time coefficients are predicted, respectively. The hindcast cross-validation results show that the model without TSD has prediction skills only in some regions of East China and South China. Compared with the model without TSD, surprisingly, the model with TSD can significantly improve the prediction performance in more regions in China, such as Xinjiang Province and Northeast China. The anomaly correlation coefficients (ACC) between hindcast precipitation with TSD and observation are higher in most years than that without TSD. The results of the independent sample test show that the forecast model with TSD has a stable and gratifying prediction skill, and the averaged ACC is increased by more than 0.1.

Late-fall satellite-based soil moisture observations show clear connections to subsequent spring streamflow

Because runoff production is more efficient over wetter soils, and because soil moisture has an intrinsic memory, soil moisture information can potentially contribute to the accuracy of streamflow predictions at seasonal leads. In this work, we use surface (0–5 cm) soil moisture retrievals obtained with the National Aeronautics and Space Administration’s Soil Moisture Active Passive satellite instrument in conjunction with streamflow measurements taken within 236 intermediate-scale (2000–10,000 km2) unregulated river basins in the conterminous United States to show that late-fall satellite-based surface soil moisture estimates are indeed strongly connected to subsequent springtime streamflow. We thus show that the satellite-based soil moisture retrievals, all by themselves, have the potential to produce skillful seasonal streamflow predictions several months in advance. In poorly instrumented regions, they could perform better than reanalysis soil moisture products in this regard.

A new approach for seasonal prediction using the coupled model CFSv2 with special emphasis on Indian Summer Monsoon

AbstractPredicting Indian Summer Monsoon (ISM) is a challenging task due to its complexity and nonlinear interactions. Any improvement in the seasonal prediction skill of models would highly benefit a large population and the economy. In this context, three sets of hindcast experiments are carried out for 9 months each, for the period 1993–2019, using the National Centers for Environmental Prediction‐Climate Forecast System version 2 (NCEP‐CFSv2). The experiments differ from each other in the way they are initialized: one is initialized in February (FebIC), and the second in May (MayIC), whereas in the third one (Exp), ocean is initialized in February and allowed to evolve but the atmosphere reinitialized every month up to May. So, hindcasts of Exp are from the pre‐evolved ocean with realistic atmospheric initial conditions of May. The representation of mean tropical Indo‐Pacific Sea surface temperature (SST), Walker circulation, mean monsoon circulation and moisture transport to Indian landmass are better represented in Exp. The ISM rainfall prediction (interannual) skill improved in Exp as compared to FebIC and MayIC over central India, Indian land mass and extended monsoon region. The initialization strategy adopted in Exp reduces the model initial shocks especially in the upper ocean heat content and SST over the Indo‐Pacific region, thereby offering a cost‐effective alternative approach for reduction of the initial shock. The well‐known cold tongue SST bias over the equatorial Pacific in MayIC is reduced significantly in Exp with improved monsoon teleconnections and the overdependence of ISM rainfall on El Niño Southern Oscillation in MayIC is also reduced in Exp.

Quantifying the Benefits of Altimetry Assimilation in Seasonal Forecasts of the Upper Ocean

AbstractSatellite altimetry measurements of sea surface height provide near‐global ocean state observations on sub‐monthly time scales, which are not always utilized by seasonal climate forecasting systems. As early as the mid‐1990s, attempts were made to assimilate altimetry observations to initialize climate models. These experiments demonstrated improved ocean forecasting skill, especially compared to experiments that did not assimilate subsurface ocean temperature information. Nowadays, some operational climate forecasting models utilize altimetry in their assimilation systems, whereas others do not. Here, we assess the impact of altimetry assimilation on seasonal prediction skill of ocean variables in two climate forecasting systems that are from the European Centre for Medium‐Range Weather Forecasts (SEAS5) and the Australian Bureau of Meteorology (ACCESS‐S). We show that assimilating altimetry improves the initialization of subsurface ocean temperatures, as well as seasonal forecasts of monthly variability in upper‐ocean heat content and sea level. Skill improvements are largest in the subtropics, where there are typically less subsurface ocean observations available to initialize the forecasts. In the tropics, there are no noticeable improvements in forecast skill. The positive impact of altimetry assimilation on forecast skill related to the subsurface ocean does not seem to affect predictions of sea surface temperature. Whether this is because current forecasting systems are close to the potential predictability limit for the ocean surface, or perhaps altimetry observations are not fully exploited, remains a question. In summary, we find that utilizing altimetry observations improves the overall global ocean forecasting skill, at least for upper‐ocean heat content and sea level.

Improving Seasonal Prediction of Summer Precipitation in the Middle–Lower Reaches of the Yangtze River Using a TU-Net Deep Learning Approach

Abstract The two-step U-Net model (TU-Net) contains a western North Pacific subtropical high (WNPSH) prediction model and a precipitation prediction model fed by the WNPSH predictions, oceanic heat content, and surface temperature. The data-driven forecast model provides improved 4-month lead predictions of the WNPSH and precipitation in the middle and lower reaches of the Yangtze River (MLYR), which has important implications for water resources management and precipitation-related disaster prevention in China. When compared with five state-of-the-art dynamical climate models including the Climate Forecast System of Nanjing University of Information Science and Technology (NUIST-CFS1.0) and four models participating in the North American Multi-Model Ensemble (NMME) project, the TU-Net produces comparable skills in forecasting 4-month lead geopotential height and winds at the 500- and 850-hPa levels. For the 4-month lead prediction of precipitation over the MLYR region, the TU-Net has the best correlation scores and mean latitude-weighted RMSE in each summer month and in boreal summer [June–August (JJA)], and pattern correlation coefficient scores are slightly lower than the dynamical models only in June and JJA. In addition, the results show that the constructed TU-Net is also superior to most of the dynamical models in predicting 2-m air temperature in the MLYR region at a 4-month lead. Thus, the deep learning-based TU-Net model can provide a rapid and inexpensive way to improve the seasonal prediction of summer precipitation and 2-m air temperature over the MLYR region. Significance Statement The purpose of this study is to examine the seasonal predictive skill of the western North Pacific subtropical high anomalies and summer rainfall anomalies over the middle and lower reaches of the Yangtze River region by means of deep learning methods. Our deep learning model provides a rapid and inexpensive way to improve the seasonal prediction of summer precipitation as well as 2-m air temperature. The work has important implications for water resources management and precipitation-related disaster prevention in China and can be extended in the future to predict other climate variables as well.

Current status and progress in the seasonal prediction of the Asian summer monsoon

The Asian summer monsoon (ASM) has a considerable impact on human lives in the most populated region in the world. Thus, its seasonal prediction is a high-profile application in Earth Science. However, the prediction skill of the regional ASM variability has long been limited due to a formidable difficulty in accurately simulating the complex interactions of the atmosphere-ocean variability and its remote influence on regional climate in numerical models. This study updates the current status and assesses progress in the ASM seasonal prediction performance. This study evaluated the seasonal prediction skill of two generations of models in hindcast data archived by the WCRP Climate-system Historical Forecast Project (CHFP) and Copernicus Climate Change Service (C3S). A special focus was put on the representation of the predominant teleconnections associated with the ENSO and Indian Ocean variability. It was found that the latest seasonal prediction systems (C3S) generally outperform previous-generation systems (CHFP) in terms of the reproducibility of the observed precipitation climatology and the prediction skill of the interannual variability of seasonal precipitation over the ASM region. Furthermore, the results suggested that the improvement of the prediction skill of the ASM likely stems from the improved representation of the monsoon climatology and teleconnections in the models. These analyses highlight the steady progress of the atmosphere-ocean coupled modelling and promise future improvements in the seasonal ASM prediction.

Skillful Seasonal Prediction of Typhoon Track Density Using Deep Learning

Tropical cyclones (TCs) seriously threaten the safety of human life and property especially when approaching a coast or making landfall. Robust, long-lead predictions are valuable for managing policy responses. However, despite decades of efforts, seasonal prediction of TCs remains a challenge. Here, we introduce a deep-learning prediction model to make skillful seasonal prediction of TC track density in the Western North Pacific (WNP) during the typhoon season, with a lead time of up to four months. To overcome the limited availability of observational data, we use TC tracks from CMIP5 and CMIP6 climate models as the training data, followed by a transfer-learning method to train a fully convolutional neural network named SeaUnet. Through the deep-learning process (i.e., heat map analysis), SeaUnet identifies physically based precursors. We show that SeaUnet has a good performance for typhoon distribution, outperforming state-of-the-art dynamic systems. The success of SeaUnet indicates its potential for operational use.

Seasonal Prediction Skill and Biases in GloSea5 Relating to the East Asia Winter Monsoon

Seasonal Prediction Skill and Biases in GloSea5 Relating to the East Asia Winter Monsoon

Mixed Layer Temperature Budget in the Arabian Sea During Winter 2019 and Spring 2019: The Role of Diapycnal Heat Flux

AbstractSea surface temperatures (SSTs) simulated by almost all models in Coupled Model Inter‐comparison Project phase five are consistently colder (∼−1°C to −4°C) than the observation in the northern Arabian Sea (AS) during winter and spring. These biases significantly weaken the seasonal and extended summer monsoon prediction skills and climate change projection's reliability in this region. To understand the relative contribution of diapycnal heat flux (Jh) compared to other terms in the mixed layer (ML) temperature (MLT) budget during these two seasons, we use time series of vertical profiles of microstructure shear measurements collected in the northeastern AS (NEAS) at 18.4°N and 67.4°E during January 10–17, 2019 (W19) and May 7–21, 2019 (S19). It is found that the vertical processes term and net surface heat flux together determine the bulk of MLT tendency during S19 and W19, and the contribution of the horizontal advection and Jh is relatively smaller. The mean value of Jh (∼−2 W m−2) at the base of the ML shows a comparable magnitude during W19 and S19, and the median values of diapycnal diffusivity (Kρ) at the ML base are not significantly different between W19 (∼3.1 × 10−6 m2 s−1) and S19 (∼5.2 × 10−6 m2 s−1). Besides, Kρ and Jh values were estimated using different Kρ‐Richardson number (Ri) based interior ocean parameterization schemes at the ML base in the NEAS overestimated with respect to observation. The alleged role of overestimation of Kρ and Jh in parameterization schemes on model simulation of cold SST bias in the AS is also discussed.

Skillful seasonal prediction of summer wildfires over Central Asia

Skillful seasonal prediction of summer wildfires over Central Asia

A Physics‐Based Empirical Model for the Seasonal Prediction of the Central China July Precipitation

AbstractJuly is the rainy peak month of central China, with a large interannual variation of local precipitation often causing serious droughts and floods. The seasonal prediction of the central China July precipitation (CCJP) is an important but still challenging task. Here, we suggest several robust seasonal predictors for the CCJP, including the preceding winter intensity of El Niño‐South Oscillation (ENSO), the winter‐to‐spring decaying rate of ENSO signals in the central Pacific, as well as the spring tropical and subpolar North Atlantic sea surface temperature anomalies. A physics‐based empirical model is then developed to predict the CCJP by using the principal component regression of the aforementioned seasonal predictors. In our statistical model, the seasonal prediction skill of the CCJP is high, with the cross‐validated reforecast skill at 0.81 during 1993–2021. This suggests a skillful seasonal prediction of the CCJP, with potentially enormous benefits for the local society and economy.

Predictability of China winter temperature under different intensities of La Niña events in seasonal climate forecast models

AbstractIn this study, we focus on the features and predictability of winter 2‐m air temperature (t2m) over China under different intensities of La Niña events. The seasonal and intraseasonal prediction skills of winter t2m over China are examined by four state‐of‐art seasonal climate forecast models. The results show that the China winter t2m anomaly is sensitive to the intensity of La Niña. Strong and moderate (S/M) La Niña events are often accompanied by cold winter over China, while a positive anomaly of mean t2m over China frequently occurs in the weak La Niña events. The intensity difference of the East Asian winter monsoon (EAWM) results in this discrepancy. All models have a slightly higher anomaly correlation coefficient (ACC) for the S/M La Niña than that for the weak La Niña in leading 1 month (LT1). Both individual models and MME exhibit higher skills of anomaly sign consistency rate (ASCR) for the S/M La Niña, which implies that models predict t2m anomaly over China more accurately in the S/M La Niña events. Furthermore, the winter t2m over China presents larger intraseasonal variability for the S/M La Niña than that for the weak La Niña. Models can capture the t2m anomaly over China more accurately when the temperature varies gently with small intraseasonal variability during winter for the S/M La Niña.

Evaluation of Dynamical Seasonal Prediction Skills for Tropical Cyclone Activity over the South China Sea in FGOALS-f2

Based on 35-year (1981–2015) ensemble (24 members) hindcasts of the IAP/LASG global seasonal prediction system named FGOALS-f2 V1.0 (FGOALS-f2), the tropical cyclone (TC) seasonal prediction skills over the South China Sea (SCS) during the TC peak season (July–November) are evaluated. Starting the prediction from June 20th, FGOALS-f2 can well capture the seasonal mean characteristics for both the genesis location and track of TCs over the SCS. For seasonal anomalous TC numbers, FGOALS-f2 underestimates the maximum and minimum of the TC number compared to the observation. The temporal correlation coefficients (TCCs) between FGOALS-f2 and the observation are 0.39 for the TC number and 0.51 for accumulated cyclone energy (ACE) over the SCS, respectively, which are both above the 95% significant level. Additionally, FGOALS-f2 has acceptable prediction skill for the seasonal mean number of TCs landing on three areas (coastal southeastern China, Indochina Peninsula, and Philippines) surrounding the SCS. The skillful prediction of SCS TCs could be ascribed to the well-predicted tropical anomaly of sea surface temperature (SSTA), TC and El Niño-Southern Oscillation (TC-ENSO) relations, and Genesis potential index (GPI).

Why Seasonal Prediction of California Winter Precipitation Is Challenging

Abstract Despite an urgent demand for reliable seasonal prediction of precipitation in California (CA) due to the recent recurrent and severe drought conditions, our predictive skill for CA winter precipitation remains limited. October hindcasts by the coupled dynamical models typically show a correlation skill of about 0.3 for CA winter (November–March) precipitation. In this study, an attempt is made to understand the underlying processes that limit seasonal prediction skill for CA winter precipitation. It is found that only about 25% of interannual variability of CA winter precipitation can be attributed to influences by El Niño–Southern Oscillation (ENSO). Instead, the year-to-year CA winter precipitation variability is primarily due to circulation anomalies independent from ENSO, featuring a circulation center over the west coast United States as a portion of a short Rossby wave train pattern over the North Pacific. Analyses suggest that dynamical models show nearly no skill in predicting these ENSO-independent circulation anomalies, thus leading to limited predictive skill for CA winter precipitation. Low predictability of these ENSO-independent circulation anomalies is further demonstrated by a large ensemble of atmospheric-only climate model simulations. While low predictability of the ENSO-independent circulation anomalies could be due to chaotic internal atmospheric processes over the mid- to high latitudes, possible underexploited predictability sources for CA precipitation in models are also discussed. This study pinpoints an urgent need for improved understanding of the formation mechanisms of ENSO-independent circulation anomalies over the U.S. West Coast for a breakthrough in seasonal prediction of CA winter precipitation.

Dynamic-LSTM hybrid models to improve seasonal drought predictions over China

Accurate drought prediction is essential for drought resilience and water resources management. The skill of seasonal drought prediction from dynamical and statistical models has limitations. This study combines dynamical models and deep learning to construct hybrid (dynamical-statistical) models. We use the random forest to extract typical grids based on the geopotential height, sea-level pressure, and 2-m temperature. The long short-term memory (LSTM) is used to construct the statistical models, with atmospheric variables as predictors and the 3-month Standardized Precipitation Index (SPI3) as the predictand. The hindcasts of atmospheric variables from the European Centre for Medium-Range Weather Forecasts (ECMWF) SEAS5 model are processed as predictors to force the statistical models. The hybrid model, constructed using the dynamical models and the LSTM, is named dynamic-LSTM (“D-LSTM”). The results suggest that the LSTM models are of information-extraction capability and robustness. When the lead time exceeds one month, the prediction skills are significantly improved by the D-LSTM models, especially in the East, Northwest, Southwest, and Tibet. In most regions, the D-LSTM models are more skillful across all seasons for lead times exceeding 30 days and are reliable in predicting droughts in spring and summer when the ECMWF SEAS5 loses skills at the seasonal scale. Furthermore, the D-LSTM models are more accurate in drought onset prediction.

Seasonal predictions of sea surface height in BCC-CSM1.1m and their modulation by tropical climate dominant modes

Sea surface height (SSH) variation has profound impacts on ecological system and socioeconomical activities, and thus skillful prediction for SSH on seasonal to interannual scale is quite desirable. This study evaluates the seasonal prediction skills of global SSH based on outputs from Beijing Climate Center Climate System Model (BCC-CSM1.1m) seasonal forecasting system and further investigates the predictability source of SSH. Compared with the damped persistence forecast, superiority of dynamic prediction of this model is mainly found over tropical Pacific and Indian Ocean. Prediction skills depend on initiated month and length of lead time and also feature the spring prediction barrier related to El Niño-Southern Oscillation (ENSO), which are clearly modulated by ENSO and Indian Ocean Dipole (IOD), respectively. During El Niño/La Niña, there are higher skills of SSH prediction over tropical Pacific and when IOD event occurred, higher skills initiated in boreal autumn emerge over the tropical Indian Ocean. What's more, SSH over tropical western Pacific is significantly correlated with ENSO/annual cycle combination mode. Further analysis indicates that ENSO/IOD is the main source of predictability for SSH over tropical Pacific/Indian Ocean. We demonstrate that the performance of SSH prediction in the model is largely dependent on the capability of representing such observed relationships and predicting ENSO/IOD. The considerable skill in predicting large-scale SSH variation on seasonal scale indicates that BCC-CSM1.1m could join as a member of ensemble of multiple global models and contribute to improved seasonal sea level prediction.

On the seasonal prediction and predictability of winter surface Temperature Swing Index over North America

The rapid day-to-day temperature swings associated with extratropical storm tracks can cause cascading infrastructure failure and impact human outdoor activities, thus research on seasonal prediction and predictability of extreme temperature swings is of huge societal importance. To measure the extreme surface air temperature (SAT) variations associated with the winter extratropical storm tracks, a Temperature Swing Index (TSI) is formulated as the standard deviation of 24-h-difference-filtered data of the 6-hourly SAT. The dominant term governing the TSI variability is shown to be proportional to the product of eddy heat flux and mean temperature gradient. The seasonal prediction skill of the winter TSI over North America was assessed using Geophysical Fluid Dynamics Laboratory's new seasonal prediction system. The locations with skillful TSI prediction show a geographic pattern that is distinct from the pattern of skillful seasonal mean SAT prediction. The prediction of TSI provides additional predictable climate information beyond the traditional seasonal mean temperature prediction. The source of the seasonal TSI prediction can be attributed to year-to-year variations of the El Niño-Southern Oscillation (ENSO), North Pacific Oscillation (NPO), and Pacific/North American (PNA) teleconnection. Over the central United States, the correlation skill of TSI prediction reaches 0.75 with strong links to observed ENSO, NPO, and PNA, while the skill of seasonal SAT prediction is relatively low with a correlation of 0.36. As a first attempt of diagnosing the combined predictability of the first-order (the seasonal mean) and second-order (TSI) statistics for SAT, this study highlights the importance of the eddy-mean flow interaction perspective for understanding the seasonal climate predictability in the extra tropics. These results point toward providing skillful prediction of higher-order statistical information related to winter temperature extremes, thus enriching the seasonal forecast products for the research community and decision makers.

Weakening of the Atlantic Niño variability under global warming

The Atlantic Niño is one of the most important patterns of interannual tropical climate variability, but how climate change will influence this pattern is not well known due to large climate model biases. Here we show that state-of-the-art climate models robustly predict a weakening of Atlantic Niños in response to global warming, mainly due to a decoupling of subsurface and surface temperature variations as the upper equatorial Atlantic Ocean warms. This weakening is predicted by most (>80%) models in the Coupled Model Intercomparison Project Phases 5 and 6 under the highest emission scenarios. Our results indicate a reduction in variability by the end of the century by 14%, and as much as 24–48% when accounting for model errors using a simple emergent constraint analysis. Such a weakening of Atlantic Niño variability will potentially impact climate conditions and the skill of seasonal predictions in many regions.

Skillful Seasonal Prediction of North American Summertime Heat Extremes

Abstract This study shows that the frequency of North American summertime (June–August) heat extremes is skillfully predicted several months in advance in the newly developed Geophysical Fluid Dynamics Laboratory (GFDL) Seamless System for Prediction and Earth System Research (SPEAR) seasonal forecast system. Using a statistical optimization method, the average predictability time, we identify three large-scale components of the frequency of North American summer heat extremes that are predictable with significant correlation skill. One component, which is related to a secular warming trend, shows a continent-wide increase in the frequency of summer heat extremes and is highly predictable at least 9 months in advance. This trend component is likely a response to external radiative forcing. The second component is largely driven by the sea surface temperatures in the North Pacific and North Atlantic and is significantly correlated with the central U.S. soil moisture. The second component shows largest loadings over the central United States and is significantly predictable 9 months in advance. The third component, which is related to the central Pacific El Niño, displays a dipole structure over North America and is predictable up to 4 months in advance. Potential implications for advancing seasonal predictions of North American summertime heat extremes are discussed.