Search Results
You are looking at 1 - 10 of 26 items for
- Author or Editor: John R. Lanzante x
- Refine by Access: All Content x
Abstract
The 10–30 day variability of extratropical, cold season (November–March) 700 mb geopotential height data of the Northern Hemisphere was studied through the use of rotated complex principal component (RCPC) analysis. The intramonthly modes (IMMs), which result from RCPC analysis of 36 years of 10–30 day bandpass filtered data, were examined. In order to assess seasonality separate analyses were done for three subseasons within the cold season.
Tests of sensitivity (to the number of eigenvectors rotated) and robustness (to the deletion of part of the sample) were conducted to ensure the stability of the RCPC analysis. A Monte Carlo procedure was used to objectively identify episodes of occurrence. Based on the episodes a sequence of ten composite maps was constructed for each mode to depict the evolution over a typical lifecycle. The significance of these lifecycle composite maps was tested using a Monte Carlo procedure.
The average periods of the IMMs are 16–18 days; almost all occurrences fall in the range 13–22 days. For objectively defined episodes (which occur about 10%–30% of the time) each IMM explains about 30%–45% of the 10–30 day bandpass variance averaged over those gridpoints deemed significant at the 1% level (which cover about 20%–40% of the grid area). Since a number of distinct IMMs were found their collective impact is considerable for intramonthly time scales.
Initial interpretation of RCPC loading maps was difficult due to the superposition of phenomena and non-idealized evolution. Composite maps which depict the evolution of each IMM over a typical lifecycle were found to be invaluable for interpretation.
Three classes of IMMs were found. The first class contains several modes which each involve a high latitude transient disturbance (zonal wavenumber 1 or 2) and an oscillating standing wave. The distinction between the modes is mostly in the location of the standing wave pattern. The wavenumber 1 transient is probably the “16-day wave” or (1,3) Rossby normal mode that has been identified in the atmosphere by others; however, the strong association of these transients with regional standing waves has not previously been documented. It is speculated that regional index cycle-like variations in the winds associated with oscillation of the standing wave may excite the transient component. The second class consists of only one mode, which is to a first approximation, an oscillating dipole concentrated in the Atlantic sector; it is speculated to be the result of a regional, baroclinic zonal index cycle. The third class contains several distinct regional midlatitude wave trains; the two most important are located over Eurasía and North America. In a broad sense, IMMs represent favored “modes of evolution” or “paths through phase space”, which the atmosphere follows on the 10–30 day time scale.
Vertical structures of the IMMs were found to be generally consistent with those of other studies of large-scale atmospheric motions. Vertical tilt was found to be large over the continents (especially central and eastern portions) while over the oceans, the IMMs were found to be nearly equivalent barotropic.
About a third of the IMMs were found to be associated with a particular low frequency state. These states are configured in the form of two well-known atmospheric teleconnection patterns. They are such that weaker than normal midlatitude westerlies occur during episodes of some IMMs.
Abstract
The 10–30 day variability of extratropical, cold season (November–March) 700 mb geopotential height data of the Northern Hemisphere was studied through the use of rotated complex principal component (RCPC) analysis. The intramonthly modes (IMMs), which result from RCPC analysis of 36 years of 10–30 day bandpass filtered data, were examined. In order to assess seasonality separate analyses were done for three subseasons within the cold season.
Tests of sensitivity (to the number of eigenvectors rotated) and robustness (to the deletion of part of the sample) were conducted to ensure the stability of the RCPC analysis. A Monte Carlo procedure was used to objectively identify episodes of occurrence. Based on the episodes a sequence of ten composite maps was constructed for each mode to depict the evolution over a typical lifecycle. The significance of these lifecycle composite maps was tested using a Monte Carlo procedure.
The average periods of the IMMs are 16–18 days; almost all occurrences fall in the range 13–22 days. For objectively defined episodes (which occur about 10%–30% of the time) each IMM explains about 30%–45% of the 10–30 day bandpass variance averaged over those gridpoints deemed significant at the 1% level (which cover about 20%–40% of the grid area). Since a number of distinct IMMs were found their collective impact is considerable for intramonthly time scales.
Initial interpretation of RCPC loading maps was difficult due to the superposition of phenomena and non-idealized evolution. Composite maps which depict the evolution of each IMM over a typical lifecycle were found to be invaluable for interpretation.
Three classes of IMMs were found. The first class contains several modes which each involve a high latitude transient disturbance (zonal wavenumber 1 or 2) and an oscillating standing wave. The distinction between the modes is mostly in the location of the standing wave pattern. The wavenumber 1 transient is probably the “16-day wave” or (1,3) Rossby normal mode that has been identified in the atmosphere by others; however, the strong association of these transients with regional standing waves has not previously been documented. It is speculated that regional index cycle-like variations in the winds associated with oscillation of the standing wave may excite the transient component. The second class consists of only one mode, which is to a first approximation, an oscillating dipole concentrated in the Atlantic sector; it is speculated to be the result of a regional, baroclinic zonal index cycle. The third class contains several distinct regional midlatitude wave trains; the two most important are located over Eurasía and North America. In a broad sense, IMMs represent favored “modes of evolution” or “paths through phase space”, which the atmosphere follows on the 10–30 day time scale.
Vertical structures of the IMMs were found to be generally consistent with those of other studies of large-scale atmospheric motions. Vertical tilt was found to be large over the continents (especially central and eastern portions) while over the oceans, the IMMs were found to be nearly equivalent barotropic.
About a third of the IMMs were found to be associated with a particular low frequency state. These states are configured in the form of two well-known atmospheric teleconnection patterns. They are such that weaker than normal midlatitude westerlies occur during episodes of some IMMs.
Abstract
A long historical record (∼100 years) of monthly sea surface temperature anomalies from the Comprehensive Ocean–Atmosphere Data Set was used to examined the lag relationships between different locations in the global Tropics. Application of complex principal component (CPC) analysis revealed that the leading mode captures ENSO-related quasi-cyclical warming and cooling in the tropical Pacific Ocean. The dominant features of this mode indicate that SST anomalies in the eastern Pacific lead those of the central Pacific. However, a somewhat weaker aspect of this mode also indicates that SST anomalies in the tropical Indian and western tropical North Atlantic Oceans vary roughly in concert with each other but lag behind those in the central and eastern Pacific. The stability of these lag relationships is indicated by the fact that the leading mode is quite similar in three different 30-year time periods.
In order to further examine these relationships some simple indexes were formed as the average over several grid points in each of the four key areas suggested by the CPC analyses. Several different types of analyses including lag correlation, checking the correspondence between extrema, and visual examination of time series plots were used to confirm the relationships implied by the CPC spatial patterns. By aggregating the lag correlations over the three 30-year time periods and performing a Monte Carlo simulation the relationships were found to be statistically significant at the 1% level. Reasonable agreement in the pattern of lag correlations was found using a different SST dataset.
Without aggregation of the lag correlations (i.e., considering each 30-year period separately) the area in the Pacific and Indian were consistently well related, but those involving the North Atlantic were more variable. The weaker correlations involving the Atlantic Ocean underscore the more tenuous nature of this remote relationship. While major ENSO-related swings in tropical Pacific SST are often followed by like variations in a portion of the Atlantic, there are times when there is either no obvious association or one of opposite sign. It may be that while ENSO variability tends to have an impact in the Atlantic, more localized factors can override this tendency. This may explain some of the contradictory statements found in the literature regarding such remote associations.
In comparing the findings of this project with some studies that utilize very recent data (since about 1982) some discrepancies were noted. In particular, some studies have reported evidence of 1) an inverse relationship between SST anomalies in the tropical Pacific and those in the eastern tropical South Atlantic and 2) the appearance of ENSO-related SST anomalies in the central tropical Pacific prior to those in the eastern tropical Pacific. From a historical perspective both of these characteristics are unusual. Thus, the recent time period may merit special attention. However, it is important to stress that caution should be exercised in generalizing findings based only on this recent time period.
Abstract
A long historical record (∼100 years) of monthly sea surface temperature anomalies from the Comprehensive Ocean–Atmosphere Data Set was used to examined the lag relationships between different locations in the global Tropics. Application of complex principal component (CPC) analysis revealed that the leading mode captures ENSO-related quasi-cyclical warming and cooling in the tropical Pacific Ocean. The dominant features of this mode indicate that SST anomalies in the eastern Pacific lead those of the central Pacific. However, a somewhat weaker aspect of this mode also indicates that SST anomalies in the tropical Indian and western tropical North Atlantic Oceans vary roughly in concert with each other but lag behind those in the central and eastern Pacific. The stability of these lag relationships is indicated by the fact that the leading mode is quite similar in three different 30-year time periods.
In order to further examine these relationships some simple indexes were formed as the average over several grid points in each of the four key areas suggested by the CPC analyses. Several different types of analyses including lag correlation, checking the correspondence between extrema, and visual examination of time series plots were used to confirm the relationships implied by the CPC spatial patterns. By aggregating the lag correlations over the three 30-year time periods and performing a Monte Carlo simulation the relationships were found to be statistically significant at the 1% level. Reasonable agreement in the pattern of lag correlations was found using a different SST dataset.
Without aggregation of the lag correlations (i.e., considering each 30-year period separately) the area in the Pacific and Indian were consistently well related, but those involving the North Atlantic were more variable. The weaker correlations involving the Atlantic Ocean underscore the more tenuous nature of this remote relationship. While major ENSO-related swings in tropical Pacific SST are often followed by like variations in a portion of the Atlantic, there are times when there is either no obvious association or one of opposite sign. It may be that while ENSO variability tends to have an impact in the Atlantic, more localized factors can override this tendency. This may explain some of the contradictory statements found in the literature regarding such remote associations.
In comparing the findings of this project with some studies that utilize very recent data (since about 1982) some discrepancies were noted. In particular, some studies have reported evidence of 1) an inverse relationship between SST anomalies in the tropical Pacific and those in the eastern tropical South Atlantic and 2) the appearance of ENSO-related SST anomalies in the central tropical Pacific prior to those in the eastern tropical Pacific. From a historical perspective both of these characteristics are unusual. Thus, the recent time period may merit special attention. However, it is important to stress that caution should be exercised in generalizing findings based only on this recent time period.
Abstract
This paper reviews the considerations in evaluating the skill and significance of screening multiple linear regression (SMLR) models. Formulations and procedures are given along with relevant references to prior studies. Topics discussed include predictor selection, serial correlation, artificial skill, true skill, and Monte Carlo significance testing. New results with wide applicability in the assessment of SMLR model skill and significance are presented in graphical form. However, the results are restricted to situations involving predictors which are independent of one another and are serially uncorrelated. The methodology presented is suggested for use in both model evaluation and experimental design.
Abstract
This paper reviews the considerations in evaluating the skill and significance of screening multiple linear regression (SMLR) models. Formulations and procedures are given along with relevant references to prior studies. Topics discussed include predictor selection, serial correlation, artificial skill, true skill, and Monte Carlo significance testing. New results with wide applicability in the assessment of SMLR model skill and significance are presented in graphical form. However, the results are restricted to situations involving predictors which are independent of one another and are serially uncorrelated. The methodology presented is suggested for use in both model evaluation and experimental design.
Abstract
Climate studies often involve comparisons between estimates of some parameter derived from different observed and/or model-generated datasets. It is common practice to present estimates of two or more statistical quantities with error bars about each representing a confidence interval. If the error bars do not overlap, it is presumed that there is a statistically significant difference between them. In general, such a procedure is not valid and usually results in declaring statistical significance too infrequently. Simple examples that demonstrate the nature of this pitfall, along with some formulations, are presented. It is recommended that practitioners use standard hypothesis testing techniques that have been derived from statistical theory rather than the ad hoc approach involving error bars.
Abstract
Climate studies often involve comparisons between estimates of some parameter derived from different observed and/or model-generated datasets. It is common practice to present estimates of two or more statistical quantities with error bars about each representing a confidence interval. If the error bars do not overlap, it is presumed that there is a statistically significant difference between them. In general, such a procedure is not valid and usually results in declaring statistical significance too infrequently. Simple examples that demonstrate the nature of this pitfall, along with some formulations, are presented. It is recommended that practitioners use standard hypothesis testing techniques that have been derived from statistical theory rather than the ad hoc approach involving error bars.
Abstract
Measurements from radiosonde temperatures have been used in studies that seek to identify the human influence on climate. However, such measurements are known to be contaminated by artificial inhomogeneities introduced by changes in instruments and recording practices that have occurred over time. Some simple diagnostics are used to compare vertical profiles of temperature trends from the observed data with simulations from a GCM driven by several different sets of forcings. Unlike most earlier studies of this type, both raw (i.e., fully contaminated) as well as adjusted observations (i.e., treated to remove some of the contamination) are utilized. The comparisons demonstrate that the effect of observational data adjustment can be as important as the inclusion of some major climate forcings in the model simulations. The effects of major volcanic eruptions critically influence temperature trends, even over a time period nearly four decades in length.
In addition, it is seen that the adjusted data show consistently better agreement than the unadjusted data, with simulations from a climate model for 1959–97. Particularly noteworthy is the fact that the adjustments supply missing warming in the tropical upper troposphere that has been attributed to model error in a number of earlier studies.
Finally, an evaluation of the fidelity of the model’s temperature response to major volcanic eruptions is conducted. Although the major conclusions of this study are unaffected by shortcomings of the simulations, they highlight the fact that even using a fairly long period of record (∼40 yr), any such shortcomings can have an important impact on trends and trend comparisons.
Abstract
Measurements from radiosonde temperatures have been used in studies that seek to identify the human influence on climate. However, such measurements are known to be contaminated by artificial inhomogeneities introduced by changes in instruments and recording practices that have occurred over time. Some simple diagnostics are used to compare vertical profiles of temperature trends from the observed data with simulations from a GCM driven by several different sets of forcings. Unlike most earlier studies of this type, both raw (i.e., fully contaminated) as well as adjusted observations (i.e., treated to remove some of the contamination) are utilized. The comparisons demonstrate that the effect of observational data adjustment can be as important as the inclusion of some major climate forcings in the model simulations. The effects of major volcanic eruptions critically influence temperature trends, even over a time period nearly four decades in length.
In addition, it is seen that the adjusted data show consistently better agreement than the unadjusted data, with simulations from a climate model for 1959–97. Particularly noteworthy is the fact that the adjustments supply missing warming in the tropical upper troposphere that has been attributed to model error in a number of earlier studies.
Finally, an evaluation of the fidelity of the model’s temperature response to major volcanic eruptions is conducted. Although the major conclusions of this study are unaffected by shortcomings of the simulations, they highlight the fact that even using a fairly long period of record (∼40 yr), any such shortcomings can have an important impact on trends and trend comparisons.
Abstract
No Abstract.
Abstract
No Abstract.
Abstract
An investigation of singularities associated with the semiannual cycle of Northern Hemisphere (25–90°N) 700 mb heights was conducted using 33 years of data. Extension of the harmonic analyses of Lanzante (1983) to a larger domain revealed the considerable importance of the semiannual cycle to the seasonal progression of heights over subtropical Asia; this is undoubtedly a reflection of the Asiatic monsoon. From a global perspective, the semiannual cycle varies in such a fashion that this Asiatic region is out of phase with a region stretching from northeastern Siberia to the Gulf of Alaska, as well as the subtropical Atlantic and eastern Pacific. Furthermore, it is suggested that the semiannual cycle is reflected regionally in the persistence of height anomalies as reported by van den Dool and Livezey (1984).
Indices for nine regions of relatively large explained variance and uniform phase angle of the second harmonic of heights were derived in order to quantify interannual variations in the semiannual cycle. Frequency distribution of parameters, based on 25-day and seasonal aggregates of the daily indices were found to be largely Gaussian. It is concluded that (for the time averages considered) there is no indication that the indices come from two populations representing either the occurrence or nonoccurrence of an amplified semiannual cycle. Finally, interannual relationships among the nine indices were investigated through auto- and cross-correlation analyses; the results were largely negative.
Abstract
An investigation of singularities associated with the semiannual cycle of Northern Hemisphere (25–90°N) 700 mb heights was conducted using 33 years of data. Extension of the harmonic analyses of Lanzante (1983) to a larger domain revealed the considerable importance of the semiannual cycle to the seasonal progression of heights over subtropical Asia; this is undoubtedly a reflection of the Asiatic monsoon. From a global perspective, the semiannual cycle varies in such a fashion that this Asiatic region is out of phase with a region stretching from northeastern Siberia to the Gulf of Alaska, as well as the subtropical Atlantic and eastern Pacific. Furthermore, it is suggested that the semiannual cycle is reflected regionally in the persistence of height anomalies as reported by van den Dool and Livezey (1984).
Indices for nine regions of relatively large explained variance and uniform phase angle of the second harmonic of heights were derived in order to quantify interannual variations in the semiannual cycle. Frequency distribution of parameters, based on 25-day and seasonal aggregates of the daily indices were found to be largely Gaussian. It is concluded that (for the time averages considered) there is no indication that the indices come from two populations representing either the occurrence or nonoccurrence of an amplified semiannual cycle. Finally, interannual relationships among the nine indices were investigated through auto- and cross-correlation analyses; the results were largely negative.
Abstract
The local association between ocean and atmosphere was examined statistically by correlating the anomalous sea surface temperature (SST) gradient and the anomalous geostrophic wind at the 700 mb level using 30 years (1949–78) of monthly data. Under the assumption that anomalous oceanic thermal gradients are transmitted to the lower troposphere via anomalous fluxes of latent and sensible heat, and by applying the thermal wind relationship, a significant positive correlation is expected. This analysis is an extension of earlier work by Harnack and Broccoli and includes results for both the Atlantic and Pacific, for both zonal and meridional components, for lags as well as contemporaneous associations, and includes an examination of spatial variability. The major findings are: 1) the expected association is found in both oceans and for both components, although it is somewhat stronger in the Pacific and when relating the zonal wind to the meridional SST gradient, 2) the best association is found in the zonal band of 35–45°N, although some seasonal variability is experienced in the Pacific, 3) the lag relationships are significant only at zero lag or with atmosphere leading ocean and 4) the effect of the association is enhanced by time averaging (over 3 months).
Abstract
The local association between ocean and atmosphere was examined statistically by correlating the anomalous sea surface temperature (SST) gradient and the anomalous geostrophic wind at the 700 mb level using 30 years (1949–78) of monthly data. Under the assumption that anomalous oceanic thermal gradients are transmitted to the lower troposphere via anomalous fluxes of latent and sensible heat, and by applying the thermal wind relationship, a significant positive correlation is expected. This analysis is an extension of earlier work by Harnack and Broccoli and includes results for both the Atlantic and Pacific, for both zonal and meridional components, for lags as well as contemporaneous associations, and includes an examination of spatial variability. The major findings are: 1) the expected association is found in both oceans and for both components, although it is somewhat stronger in the Pacific and when relating the zonal wind to the meridional SST gradient, 2) the best association is found in the zonal band of 35–45°N, although some seasonal variability is experienced in the Pacific, 3) the lag relationships are significant only at zero lag or with atmosphere leading ocean and 4) the effect of the association is enhanced by time averaging (over 3 months).
Abstract
Associations between Pacific and Atlantic sea surface temperature (SST) and the 700 mb circulation were studied using 30 years (1949-78) of gridded monthly data. The monthly data were grouped into four (nonstandard) seasons for the analysis. The linear correlation between the two fields (SST and 700 mb heights) was analyzed by finding the eigenvalue solution of the mean product matrix of the cross correlations. This technique, introduced by Prohaska, was modified through the use of a varimax rotation. The result is a set of pairs of patterns (one pattern for each medium) which explains part of the linear correlation between the two fields. The number of significant modes (pairs) was determined by a Monte Carlo simulation.
One well-known atmospheric teleconnections pattern, referred to as the Pacific/North American (PNA) pattern by Wallace and Gutzler, was found in all seasons; however, seasonal differences were noted. During the warm season, the PNA pattern manifests itself as the Great Plains (GP) pattern. The OP pattern results in dry/hot (wet/cool) conditions over much of the central and eastern United States. Another, the Northeast (NE) pattern, has a center along the coast of the eastern United States and can result in dry/cool (wet/warm) weather in the Northeast. Additionally, the North Atlantic Oscillation was identified during both the coldand warm seasons.
In the examination of lag associations, significant relationships in which the atmosphere leads the ocean were found for all seasons (and many modes) in the Pacific, but only for the case of January-March 700 mb heights leading SST (by one to two months) in the Atlantic. In the case of the ocean leading the atmosphere (by one to two months), significant results were found only for Pacific SST leading JanuaryMarch 100 mb heights.
Abstract
Associations between Pacific and Atlantic sea surface temperature (SST) and the 700 mb circulation were studied using 30 years (1949-78) of gridded monthly data. The monthly data were grouped into four (nonstandard) seasons for the analysis. The linear correlation between the two fields (SST and 700 mb heights) was analyzed by finding the eigenvalue solution of the mean product matrix of the cross correlations. This technique, introduced by Prohaska, was modified through the use of a varimax rotation. The result is a set of pairs of patterns (one pattern for each medium) which explains part of the linear correlation between the two fields. The number of significant modes (pairs) was determined by a Monte Carlo simulation.
One well-known atmospheric teleconnections pattern, referred to as the Pacific/North American (PNA) pattern by Wallace and Gutzler, was found in all seasons; however, seasonal differences were noted. During the warm season, the PNA pattern manifests itself as the Great Plains (GP) pattern. The OP pattern results in dry/hot (wet/cool) conditions over much of the central and eastern United States. Another, the Northeast (NE) pattern, has a center along the coast of the eastern United States and can result in dry/cool (wet/warm) weather in the Northeast. Additionally, the North Atlantic Oscillation was identified during both the coldand warm seasons.
In the examination of lag associations, significant relationships in which the atmosphere leads the ocean were found for all seasons (and many modes) in the Pacific, but only for the case of January-March 700 mb heights leading SST (by one to two months) in the Atlantic. In the case of the ocean leading the atmosphere (by one to two months), significant results were found only for Pacific SST leading JanuaryMarch 100 mb heights.
