Search Results
You are looking at 1 - 10 of 154 items for
- Author or Editor: Arun Kumar x
- Refine by Access: All Content x
Abstract
In recent years, there has been a steady increase in the emphasis on routine seasonal climate predictions and their potential for enhancing societal benefits and mitigating losses related to climate extremes. It is also suggested by the users, as well as by the producers of climate predictions, that for informed decision making, real-time seasonal climate predictions should be accompanied by a corresponding level of skill estimated from a sequence of the past history of forecasts. In this paper it is discussed whether conveying skill information to the user community can indeed deliver the promised benefits or whether issues inherent in the estimates of seasonal prediction skill may still lead to potential misinterpretation of the information content associated with seasonal predictions. Based on the analysis of atmospheric general circulation model simulations, certain well-known, but often underappreciated, issues inherent in the estimates of seasonal prediction skill from the past performance of seasonal forecasts are highlighted. These include the following: 1) the stability of estimated skill depends on the length of the time series over which seasonal forecasts are verified, leading to scenarios where error bars on the estimated skill could be of the same magnitude as the skill itself; 2) a single estimate of skill obtained from the verification over a given forecast time series, because of variation in the signal-to-noise ratio from one year to another, is generally not representative of seasonal prediction skill conditional to sea surface temperature anomalies on a case-by-case basis. These issues raise questions on the interpretation, presentation, and utilization of skill information for seasonal prediction efforts and present opportunities for increased dialogue and the exploration of ways for their optimal utilization by decision makers.
Abstract
In recent years, there has been a steady increase in the emphasis on routine seasonal climate predictions and their potential for enhancing societal benefits and mitigating losses related to climate extremes. It is also suggested by the users, as well as by the producers of climate predictions, that for informed decision making, real-time seasonal climate predictions should be accompanied by a corresponding level of skill estimated from a sequence of the past history of forecasts. In this paper it is discussed whether conveying skill information to the user community can indeed deliver the promised benefits or whether issues inherent in the estimates of seasonal prediction skill may still lead to potential misinterpretation of the information content associated with seasonal predictions. Based on the analysis of atmospheric general circulation model simulations, certain well-known, but often underappreciated, issues inherent in the estimates of seasonal prediction skill from the past performance of seasonal forecasts are highlighted. These include the following: 1) the stability of estimated skill depends on the length of the time series over which seasonal forecasts are verified, leading to scenarios where error bars on the estimated skill could be of the same magnitude as the skill itself; 2) a single estimate of skill obtained from the verification over a given forecast time series, because of variation in the signal-to-noise ratio from one year to another, is generally not representative of seasonal prediction skill conditional to sea surface temperature anomalies on a case-by-case basis. These issues raise questions on the interpretation, presentation, and utilization of skill information for seasonal prediction efforts and present opportunities for increased dialogue and the exploration of ways for their optimal utilization by decision makers.
Abstract
The expected value for various measures of skill for seasonal climate predictions is determined by the signal-to-noise ratio. The expected value, however, is only realized for long verification time series. In practice, the verifications for specific seasons—for example, forecasts for the December–February seasonal mean—seldom exceed a sample size of 30. The estimates of skill measure based on small verification time series, because of sampling errors, can have large departures from their expected value. An analysis of spread in the estimates of skill measures with the length of verification time series and for different signal-to-noise ratios is made. The analysis is based on the Monte Carlo approach and skill measures for deterministic, categorical, and probabilistic forecasts are considered. It is shown that the behavior of spread for various skill measures can be very different and it is not always the largest for the small values of signal-to-noise ratios.
Abstract
The expected value for various measures of skill for seasonal climate predictions is determined by the signal-to-noise ratio. The expected value, however, is only realized for long verification time series. In practice, the verifications for specific seasons—for example, forecasts for the December–February seasonal mean—seldom exceed a sample size of 30. The estimates of skill measure based on small verification time series, because of sampling errors, can have large departures from their expected value. An analysis of spread in the estimates of skill measures with the length of verification time series and for different signal-to-noise ratios is made. The analysis is based on the Monte Carlo approach and skill measures for deterministic, categorical, and probabilistic forecasts are considered. It is shown that the behavior of spread for various skill measures can be very different and it is not always the largest for the small values of signal-to-noise ratios.
Abstract
For the dynamic normal-mode initialization (DNMI) procedure used in the initialization of the limited-area models, it is shown that it does not succeed in initializing higher vertical modes. An improved initialization procedure where the DNMI is performed in the vertical mode space is next formulated. A comparison of the two initialization schemes demonstrates that with the new procedure the initialization of the higher vertical modes can be better accomplished. The need for a better initialization method for limited-area models is prompted by the fact that if the assimilation of the observed estimates of the convective heating toward enhancing the analysis of the divergence is to be achieved, higher vertical modes have to be properly initialized. It is shown that the improved DNMI procedure does succeed in the assimilation of the convective heating information.
Abstract
For the dynamic normal-mode initialization (DNMI) procedure used in the initialization of the limited-area models, it is shown that it does not succeed in initializing higher vertical modes. An improved initialization procedure where the DNMI is performed in the vertical mode space is next formulated. A comparison of the two initialization schemes demonstrates that with the new procedure the initialization of the higher vertical modes can be better accomplished. The need for a better initialization method for limited-area models is prompted by the fact that if the assimilation of the observed estimates of the convective heating toward enhancing the analysis of the divergence is to be achieved, higher vertical modes have to be properly initialized. It is shown that the improved DNMI procedure does succeed in the assimilation of the convective heating information.
Abstract
Seasonal prediction skill of SSTs from coupled models has considerable spatial variations. In the tropics, SST prediction skill in the tropical Pacific clearly exceeds prediction skill over the Atlantic and Indian Oceans. Such skill variations can be due to spatial variations in observing system used for forecast initializations or systematic errors in the seasonal prediction systems, or they could be a consequence of inherent properties of the coupled ocean–atmosphere system leaving a fingerprint on the spatial structure of SST predictability. Out of various alternatives, the spatial variability in SST prediction skill is argued to be a consequence of inherent characteristics of climate system. This inference is supported based on the following analyses. SST prediction skill is higher over the regions where coupled air–sea interactions (or Bjerknes feedback) are inferred to be stronger. Coupled air–sea interactions, and the longer time scales associated with them, imprint longer memory and thereby support higher SST prediction skill. The spatial variability of SST prediction skill is also consistent with differences in the ocean–atmosphere interaction regimes that distinguish between whether ocean drives the atmosphere or atmosphere drives the ocean. Regions of high SST prediction skill generally coincide with regions where ocean forces the atmosphere. Such regimes correspond to regions where oceanic variability is on longer time scales compared to regions where atmosphere forces the ocean. Such regional differences in the spatial characteristics of ocean–atmosphere interactions, in turn, also govern the spatial variations in SST skill, making spatial variations in skill an intrinsic property of the climate system and not an artifact of the observing system or model biases.
Abstract
Seasonal prediction skill of SSTs from coupled models has considerable spatial variations. In the tropics, SST prediction skill in the tropical Pacific clearly exceeds prediction skill over the Atlantic and Indian Oceans. Such skill variations can be due to spatial variations in observing system used for forecast initializations or systematic errors in the seasonal prediction systems, or they could be a consequence of inherent properties of the coupled ocean–atmosphere system leaving a fingerprint on the spatial structure of SST predictability. Out of various alternatives, the spatial variability in SST prediction skill is argued to be a consequence of inherent characteristics of climate system. This inference is supported based on the following analyses. SST prediction skill is higher over the regions where coupled air–sea interactions (or Bjerknes feedback) are inferred to be stronger. Coupled air–sea interactions, and the longer time scales associated with them, imprint longer memory and thereby support higher SST prediction skill. The spatial variability of SST prediction skill is also consistent with differences in the ocean–atmosphere interaction regimes that distinguish between whether ocean drives the atmosphere or atmosphere drives the ocean. Regions of high SST prediction skill generally coincide with regions where ocean forces the atmosphere. Such regimes correspond to regions where oceanic variability is on longer time scales compared to regions where atmosphere forces the ocean. Such regional differences in the spatial characteristics of ocean–atmosphere interactions, in turn, also govern the spatial variations in SST skill, making spatial variations in skill an intrinsic property of the climate system and not an artifact of the observing system or model biases.
Abstract
Using extensive hindcasts from seasonal prediction systems participating in the North American Multi-Model Ensemble (NMME), possible causes for low skill in predicting seasonal mean precipitation over California during December–February (DJF) are investigated. The analysis focuses on investigating two possibilities for low prediction skill: role model biases or inherent predictability limits. The motivation for the analysis was the seasonal prediction during DJF 2015/16 that called for enhanced probability for above normal precipitation over southern California (which was consistent with expected conditions during an extreme El Niño) while the observed precipitation was below normal. Based on various analysis approaches and using hindcast datasets from multiple seasonal prediction systems, we build up the evidence that low skill in predicting seasonal mean precipitation over California is likely to be due to inherent predictability associated with a low signal-to-noise (SNR) regime. For the same set of seasonal prediction systems, the precipitation variability over California is contrasted with that over the southeast United States where prediction skill, as well as the SNR, is higher. The discussion also notes that building a knowledge base that goes beyond the well-known response to ENSO (based on the linear regression or composite techniques) has proven to be difficult and a systematic approach to reaching resolution to some of the overarching questions is required, and toward that end, a pathway is suggested.
Abstract
Using extensive hindcasts from seasonal prediction systems participating in the North American Multi-Model Ensemble (NMME), possible causes for low skill in predicting seasonal mean precipitation over California during December–February (DJF) are investigated. The analysis focuses on investigating two possibilities for low prediction skill: role model biases or inherent predictability limits. The motivation for the analysis was the seasonal prediction during DJF 2015/16 that called for enhanced probability for above normal precipitation over southern California (which was consistent with expected conditions during an extreme El Niño) while the observed precipitation was below normal. Based on various analysis approaches and using hindcast datasets from multiple seasonal prediction systems, we build up the evidence that low skill in predicting seasonal mean precipitation over California is likely to be due to inherent predictability associated with a low signal-to-noise (SNR) regime. For the same set of seasonal prediction systems, the precipitation variability over California is contrasted with that over the southeast United States where prediction skill, as well as the SNR, is higher. The discussion also notes that building a knowledge base that goes beyond the well-known response to ENSO (based on the linear regression or composite techniques) has proven to be difficult and a systematic approach to reaching resolution to some of the overarching questions is required, and toward that end, a pathway is suggested.
Abstract
In this study the influence of snow on atmospheric seasonal mean variability in the extratropical latitudes during boreal winter was studied. The motivation for this analysis was to understand the characteristics of low-frequency atmospheric variability in the extratropical latitudes, and to assess if the interannual variations in snow could lead to potential predictability on seasonal timescales. The influence of snow on atmospheric variability was assessed from a suite of atmospheric general circulation model (GCM) simulations where snow depth amount was either prescribed to a seasonally varying climatology, or was allowed to evolve during the model integration. Further, the influence of snow variability was contrasted with the influence of interannual variability in sea surface temperatures (SSTs) on the atmospheric flow.
A systematic influence of snow variability on the atmospheric seasonal mean variability was found. For example, for the GCM simulations in which snow amount and its extent were allowed to evolve freely, the interannual variability of surface air temperature was found to be larger. The influence of snow variability, however, was confined to the lower troposphere, and little change in the interannual variability of upper-tropospheric circulation, for example, 200-hPa heights, occurred. This bottom-up vertical structure of the influence of snow on the atmospheric variability was in contrast to the top-down influence of tropical SST variability on the extratropical flow.
The cause for the enhancement of atmospheric variability in the lower troposphere was argued to be related to the dependence of surface albedo on snow depth amount. This dependence was such that the interaction between the atmospheric variability and the underlying snow could be viewed as a positive feedback process whereby surface temperature anomalies amplify even further.
Abstract
In this study the influence of snow on atmospheric seasonal mean variability in the extratropical latitudes during boreal winter was studied. The motivation for this analysis was to understand the characteristics of low-frequency atmospheric variability in the extratropical latitudes, and to assess if the interannual variations in snow could lead to potential predictability on seasonal timescales. The influence of snow on atmospheric variability was assessed from a suite of atmospheric general circulation model (GCM) simulations where snow depth amount was either prescribed to a seasonally varying climatology, or was allowed to evolve during the model integration. Further, the influence of snow variability was contrasted with the influence of interannual variability in sea surface temperatures (SSTs) on the atmospheric flow.
A systematic influence of snow variability on the atmospheric seasonal mean variability was found. For example, for the GCM simulations in which snow amount and its extent were allowed to evolve freely, the interannual variability of surface air temperature was found to be larger. The influence of snow variability, however, was confined to the lower troposphere, and little change in the interannual variability of upper-tropospheric circulation, for example, 200-hPa heights, occurred. This bottom-up vertical structure of the influence of snow on the atmospheric variability was in contrast to the top-down influence of tropical SST variability on the extratropical flow.
The cause for the enhancement of atmospheric variability in the lower troposphere was argued to be related to the dependence of surface albedo on snow depth amount. This dependence was such that the interaction between the atmospheric variability and the underlying snow could be viewed as a positive feedback process whereby surface temperature anomalies amplify even further.
Abstract
In the Atmospheric Model Intercomparison Project (AMIP) simulations the sea surface temperatures (SSTs) are specified and the oceanic evolution consistent with air–sea interaction is not included. This omission could lead to errors in the atmospheric response to SSTs. At the same time, the AMIP experimental setup is well suited for investigating many aspects of climate variability (e.g., the attribution of the interannual atmospheric variability) and continues to be extensively used. As coupled El Niño–Southern Oscillation (ENSO) SST variability is a dominant factor in determining the predictable component of the observed interannual atmospheric variability, the difference in the atmospheric response to ENSO SSTs between AMIP and coupled simulations is investigated. The results indicate that the seasonal atmospheric response to ENSO between coupled and uncoupled integrations is similar, and the inclusion of oceanic evolution consistent with air–sea interaction does not play a dominant role. The analysis presented in this paper is one step toward assessing differences in atmospheric response to SSTs in coupled and uncoupled simulations, and is required to correctly interpret the results of AMIP simulations.
Abstract
In the Atmospheric Model Intercomparison Project (AMIP) simulations the sea surface temperatures (SSTs) are specified and the oceanic evolution consistent with air–sea interaction is not included. This omission could lead to errors in the atmospheric response to SSTs. At the same time, the AMIP experimental setup is well suited for investigating many aspects of climate variability (e.g., the attribution of the interannual atmospheric variability) and continues to be extensively used. As coupled El Niño–Southern Oscillation (ENSO) SST variability is a dominant factor in determining the predictable component of the observed interannual atmospheric variability, the difference in the atmospheric response to ENSO SSTs between AMIP and coupled simulations is investigated. The results indicate that the seasonal atmospheric response to ENSO between coupled and uncoupled integrations is similar, and the inclusion of oceanic evolution consistent with air–sea interaction does not play a dominant role. The analysis presented in this paper is one step toward assessing differences in atmospheric response to SSTs in coupled and uncoupled simulations, and is required to correctly interpret the results of AMIP simulations.
Abstract
Based on the variability of heat content in the upper 300 m of the ocean (HC300), the feasibility of defining an index of Pacific decadal oscillation (PDO) is explored. The motivation for defining the PDO index on HC300 stems from the following considerations: (i) a need to accentuate lower-frequency variations in the monitoring of PDO and (ii) to take into account variations in the temperatures associated with the PDO that extend throughout the upper ocean (and are modulated by the seasonal cycle of mixed layer variability). It is demonstrated that an HC300-based definition is better suited to encapsulate these characteristics in the PDO variability. The variability in an HC300-based definition is also contrasted with the traditional definition of the PDO based on SSTs.
Abstract
Based on the variability of heat content in the upper 300 m of the ocean (HC300), the feasibility of defining an index of Pacific decadal oscillation (PDO) is explored. The motivation for defining the PDO index on HC300 stems from the following considerations: (i) a need to accentuate lower-frequency variations in the monitoring of PDO and (ii) to take into account variations in the temperatures associated with the PDO that extend throughout the upper ocean (and are modulated by the seasonal cycle of mixed layer variability). It is demonstrated that an HC300-based definition is better suited to encapsulate these characteristics in the PDO variability. The variability in an HC300-based definition is also contrasted with the traditional definition of the PDO based on SSTs.
Abstract
Faced with the scenario when prediction skill is low, particularly in conjunction with long-range predictions, a commonly proposed solution is that an increase in ensemble size will rectify the issue of low skill. Although it is well known that an increase in ensemble size does lead to an increase in prediction skill, the general scope of this supposition, however, is that low prediction skill is not a consequence of constraints imposed by inherent predictability limits, but an artifact of small ensemble sizes, and further, increases in ensemble sizes (that are often limited by computational resources) are the major bottlenecks for improving long-range predictions. In proposing that larger ensemble sizes will remedy the issue of low skill, a fact that is not well appreciated is that for scenarios with high inherent predictability, a small ensemble size is sufficient to realize high predictability, while for scenarios with low inherent predictability, much larger ensemble sizes are needed to realize low predictability. In other words, requirements on ensemble size (to realize the inherent predictability) and inherent predictability are complementary variables. A perceived need for larger ensembles, therefore, may also imply the presence of low predictability.
Abstract
Faced with the scenario when prediction skill is low, particularly in conjunction with long-range predictions, a commonly proposed solution is that an increase in ensemble size will rectify the issue of low skill. Although it is well known that an increase in ensemble size does lead to an increase in prediction skill, the general scope of this supposition, however, is that low prediction skill is not a consequence of constraints imposed by inherent predictability limits, but an artifact of small ensemble sizes, and further, increases in ensemble sizes (that are often limited by computational resources) are the major bottlenecks for improving long-range predictions. In proposing that larger ensemble sizes will remedy the issue of low skill, a fact that is not well appreciated is that for scenarios with high inherent predictability, a small ensemble size is sufficient to realize high predictability, while for scenarios with low inherent predictability, much larger ensemble sizes are needed to realize low predictability. In other words, requirements on ensemble size (to realize the inherent predictability) and inherent predictability are complementary variables. A perceived need for larger ensembles, therefore, may also imply the presence of low predictability.
