1980-1989

1. [Blackwell et al. 1980]
   Terrain effects on heat flow. Ref from [Pollack and Huang 2000]
   

2. [Trenberth and Paolino 1980]
   NCAR dataset of SLP. Assessment of quality and homogeneity.

3. [Barnet 1981]
   Predictability of seasonal air temperatures over North America using linear methods (similar to CCA). Predictors are sst and slp from tropical and mid latitude pacific over 78 years. SST can be used to predict air temperature in advance over areas of North America (one and more seasons in advance). Hindcast results were supproted by independent forecast experiment. Tropical sst and slp outperformed midlatitude variables. El Nino relation in description. Predictability related to tropical east/west gradients. Skill dependent on season and region. Potential predictability compared to relative predictability. See [Barnett and Preisendorfer 1987]
   Seasonal prediction, Nino, cca.

4. [Potter 1981]
   Homogeneity test for precipitation series. Detects only one inhomogeneity. Creates a reference series for a 19-station network that did not vary with time by using the mean of all the other 18 stations in his network for each candidate. After the homogeneity test was run on all the estations, he created a new reference series as before but excluding those stations with inhomogeneities.

5. [Horel 1981]
   Apply varimax rotated EOFs to Z500 in N. Hemisphere. Same data as [Wallace and Gutzler 1981]: DJF monthly values for 1962-77. Show example of the use of varimax rotation and advantages. The aim of rotating is not to explain maximum variance but to find teleconections with explained variance concentrated in a few regions. The results show this alternative solution is easier to relate to regional anomaly fields. A drawback is that the alternative solution is not unique. For the winter analysis produces 'seesaws' with the southern center of action in the subtropics. For the summertime the patterns are analogous but displaced northward.
   NAO, EOFs.

6. [Rogers 1981]
   Compares 21st eofs of slp and 500 mb geop. for all seasons. Users daily data averaged to monthly anomalies for the period 1946-1977. 1st eof is related to the NAO. The correlations between surface and 500 mb eofs are higher in winter and spring and autumn and lower in summer. Spring and autumn show a barotropic behavior. Winter shows a 500mb 1st eof similar to st anomalies. The 1st two eofs in winter for 500mb are (?) linear combinations of the slp ones (not independent normal correlations)
   NAO, slp, 500 mb geopotential

7. [Trenberth and Paolino 1981]
   NCAR dataset of SLP. Calculate EOFs of seasonal and annual slp for the period 1925-77. Attempt autorregresive prediction of some pcs with no success. First description of a hemispherical scale pattern related with the NAO (?. see also [Kutzbach 1970]). [Kutzbach 1970] show first description for winter. Here all seasons and annual maps. Southern oscillation is described as a global scale mode of the atmosphere. Some pcs show significant variability at 16 yr, QBO and 6 yr. No lag relationships nor seasonal correlation.
   NAO, AO, SOI, hemispherical patterns.

8. [Wallace and Gutzler 1981]
   Self defined index of monthly differences in 1000-700 mb thickness between Greenland and Oslo.
   Detect anomalies associated to NAO over the Northern Pacific. Strong Icelandic low and weak Aleutian low related to positive thickness anomalies over the eastern USA and NW Europe and negative thickness anomalies over the Labrador and Greenland region.
   Describe zonally symmentric seesaw.
   Describe PNA pattern with relationship to NAO. Evidence of seesaw between the Icelandic and Aleutian lows and of common impact over Eastern USA Description of EA, PNA, WA, WP and EU as teleconection at 500 mb. Description of NAO and PNA at sea level pressure.
   Bo para descripcion de patrons en HN. Mapas de teleconexion.

9. [North et al. 1982]
   Weighting of latitude
   Weigting, methods

10. [Rogers and van Loon 1982]
    Modes of circulation in the SH
    SH, AAO

11. [Hellerman and Rosenstein 1983]
    Climatology of wind stress and wind stress curl.
    Wind stress curl.

12. [Livezey and Chen 1983]
    Significance of patterns using Monte Carlo techniques. Effective sample size, degrees of freedom.
    Monte Carlo, degrees of freedom, significance, spatial analysis

13. [Salstein et al. 1983]
    Use EOF to study the variability of annual water vapor and water vapor transport fields in the Norther Hemisphere during 1958-73. The distribution of water vapor is mostly determined by T and thus shows a latitudinal arrangement. Variability in wter vapor shows separation of the tropics into two regions of different sign centered in wester Africa and western Pacific. Time series show a steep trend (change of regime) maybe related with changes in the Hadley cell. Zonal transport dominates over the meridional one. Zonal transport shows westerly fluxes prevail in the midlatitudes and esasterly in the tropics (except for Asia due to the Monzoon). Meridional transport tends to be less organized. In midlatitudes northward fluxes prevail and southward in the tropics.1st EOF of zonal transport shows a similar pattern to that of the water vapor EOF. Zonal trnasport depends more on the amount of water vapor than on zonal winds. The pc shows corresponding bahavior (negative trend). Meridional transport does not show the two regimes and maybe is more dependetn on meridional wind. Ecuatorial areas south of Africa and Asia seem to contribute more to the Hadley cell. From a complex eof analysis the authors point areas of divergence and convergence that change with time and thus heating or cooling the surface.
    Water vapor. Transport. Winds.

14. [Kim et al. 1984]
    Referenced and discussed by [Wigley et al. 1990].
    Downscaling.

15. [Mitchell 1984]
    Physical processes by which certain gasses influence cliamte. Sumarizes constituents and origin and illustrates GCMs to date.
    GCM, climate change, CO2, pollutants.

16. [Peixoto and Oort 1984]
    A revview of our present understanding o the gobal climate system, consisting of the atmosphere, hydrosphere, cryosphere, lithosphere, and biosphere and their complex interactins and feedbacks in given from the point of view of a physicist.
    Climate variability, climate change, physics of cliamte.

17. [Rogers 1984]
    Use available z500, ST and SLP. Try to see connection between NAO and SO. Both influence strongly East USA but there is no apparent influence of SO on the NAO trends. Both, SO and NAO have a clear signature spreaded over most of the Northern Hemisphere.
    NAO, SO, z500, ST, SLP.

18. [Trenberth and Shin 1984]
    Complex EOF are applied to seasonal SLP data 1925-29. Three modes are described. The first one shows zonallity and has elements of the NAO, PNA and SO. Evidence for the origins of the QBO periodicity has not been found.
    NAO, PNA, SO, QBO, SLP.

19. [Trenberth 1984a]
    Statistical description of methods to compute autocorrelations and filtering of seasonality in the data, length of the time series, degrees of freedom.
    Autocorrelations, persistence, seasonality.

20. [Trenberth 1984b]
    Statistical description of methods to compute signal to noise ratio, potential predictability, degrees of freedom.
    Predictability, signal to noise ratio

21. [Rogers 1985]
    Shows surface and circulation changes that took place before and during the 1920's. Warming in the Arctic sea and Greenland lags Europe. Westerlies strengthened through the late 19th century. Explains the simultaneous or out of phase behavior of Greenland and Northern Europe by changes in the size and position of the Icelantic low. Very descriptive.
    NAO, 1920's, Greenland, Europe.

22. [Alexanderson 1986]
    Studies relative homogeneity in 56 Swedish precipitation data. Explains SNHT. Supports correcting annual data against monthly page 672.
    Relative homogeneity, precipitation, quality

23. [Enting 1986]
    Unifies interpretation of several filtering techniques from the point of view of digital filtering.
    Filtering, Atmospheric constituents, sigma filters, refs. to [Bloomfield 1976], pag 149.

24. [Jones et al. 1986a]
    Description of a new version of the UKMO Surface temperature dataset. Corrections, uncertainties,trends
    UKMO, SLP,dataset, Northern hemisphere

25. [Jones et al. 1986b]
    Description of a new version of the UKMO ST dataset for trends in the SH. Corrections, uncertainties.
    UKMO, SLP,dataset, Southern Hemisphere

26. [Lachenbruch and Marshall 1986]
    Analysis of geothermal data from the north slope of Alaska with climate change orientation. Cite from [Pollack and Huang 2000].
    UKMO, SLP,dataset, Southern Hemisphere

27. [Marks 1984]
    'Believes are not automatically updated by the best evidence available, but have an active life of their own and fight tenaciously for their own surival. They tell us what to read, what to listen to, who to trust and how to rationalize contrary information'.
    Psicology

28. [Richman 1986]
    Paper pro-rotated eof. Unrotated eofs exhibit some bad sensitivity to some aspects which hamper their ability to isolate individual modes of variation: domain shape dependence, subdomain instability, sampling problems and inadequate portrayal of physical relationships embeded in the original correlation matrix. Mathematical description of eof calculation, varimax, oblimin, harris kaiser II, promax, dapper and orthogonal procustex rotation.
    EOF, PCA, factor analysis, rotation of principal components.

29. [Warrilow et al. 1986]
    Reference to soil model in ECHAM4.
    ECHAM4

30. [Wieringa 1986]
    A two-laycr boundary layer model is presented for regional assessment of seasonal-average surface wind climatc from stalion data. In the surface layer the wind measurements are objectively corrected for imperfect siation exposure. using gustiness-derived roughness values for each azimuth sector. This produces at 60m a mesoscale wind spccd, Urn,representative for a SxSkm’ area block. The regional averaging of Urnin the Ekman layer is done by geostrophic similarity methods, referring to a mesoscale block roughness obtained from maps through terrain classification. All wind data for the flat Netherlands are analysed, using a 30-year series of annual windiness indiccs for time-span adjustment of non-simultaneous data series. The evaluation leads to objectivcly interpolated seasonal surface wind maps. supplemented by wind distributions and extreme dcsign wind spccds. Possible extension of the methodology to complex terrain is discussed.
    Surface wind, surface roughness

31. [Barnston and Livezey 1987]
    Comprehensive classification and seasonal characterization of patterns in 700 mb height.
    NAO is the only patter unambiguosly found in 12 months.
    Catalogo de patrons en HN.

32. [Barnett and Preisendorfer 1987]
    Uses canonical correlation analysis to predict surface temperatures in 33 stations over the USA. As predictors they use slp, sst and temperature itself lagged two to three months in advance. Good discussion on predictive skill and the initial rationale of seasonal forecasts.
    Canonical correlation analysis, cca, crosvalidation, .

33. [Jolliffe 1987]
    Comments on [Richman 1986]. Basically it seems like a language problem. PCAs are made for purposes which contradict Richman arguments
    EOF, PCA, factor analysis, rotation of principal components.

34. [Kalstein et al. 1987]
    Comparison of three clustering techniques for use in synoptic climatological classification: the average linkage, the centroid and ward's method. Clear description of methods and comments. The author arguments against rotating pcas. Classifies weather types in the clustering techniques making use of unrotated pcas for 1 site. For this site the authors use 7 variables and 4 meassurements per day. Component scores resulting from the 4 components of the pca each show 28 values per day (7 variables by 4 meassurements). There is a matrix of 620 (days) by 4 component scores. This matrix is grouped according to the 3 clustering procedures. The best performing method is the average linkage.
    PCA, cluster analysis, average linkage, ward's, centroid. Synoptic classification.

35. [Kalnicky 1987]
    Seasonality, changes in the seasonal cycle using complex demodulation. CO2 concentrations in Mauna Loa
    Seasonality, climate change, clomplex demodulation.
36. [Kalnicky 1987]
    Seasonality, changes in the seasonal cycle using factor analysis
    Seasonality, climate change.

37. [Karl and Williams 1987]
    Homogeneity test dependent of metadata.
    Homogeneity. Temperature and precipitation

38. [Lamb and Peppler 1987]
    Concept of NAO. Description of spatial distribution and implications for precipitation in Morocco.
    NAO, precipitation in Morocco.

39. [Jones 1987]
    Errors in grided SLP data in de Arctic before 1931. The current believe of a semi-permanent Arctic high caused values to be 4 to 6 mb too high. Also errors in northwestern USA. Errors are corrected using independent station data. He uses linear regression with EOF analysis.
    Reference for SLP data set (UKMO). SLP.

40. [Moses et al. 1987]
    Focuses on characteristics of NAO reversals in relation to blocking episodes and reduced westerlies over northern Europe. Blocking anticyclones in NE Atlantic are the preferred region for strong blocking in NH.
    Characterizes precipitation and temperature chantes in the times of reversal.
    Algunhas graficas interesantes, pero non aporta moito novo, salvo punto de vista de blocking.

41. [Nicholls 1987]
    Suggests use of cca for teleconnection assessment. Some cca description in appendix. Good biblio account. Example of application to some southern oscillation time series. Shows potential to capture time signal.
    Canonical Correlation Analysis, teleconnections, SOI.

42. [Palutikov et al. 1987]
    Modern applications of wind energy include water pumping and, for supply of electricity, grid-connected wind turbines and wind/diesel stand-alone systems. In Britain, wind energy has been found to be particularly suited to isolated communities where the costs of transporting diesel fuel are such that it is prohibitively expensive to provide a constant source of electricity. The impact of long-term climate variability on wind energy production has been almost totally neglected in wind energy studies. The 1898–1954 Southport windspeed record is analyzed and it is shown that the annual mean varies between 7.3 and 5.2 m s−2. Depending on the turbine characteristics, this can represent a 50% reduction in output. A principal components analysis (PCA) was performed on forty-two monthly windspeed records for the years 1962–81 and the scores of the first component (PC1) were used to analyze temporal variability in the wind field. It was found that the period 1962–81 has three phases of alternating high-low-high winds over Britain. The time series of windspeed PC1 scores is shown to be highly correlated with the PC1 scores of a PCA of the Lamb Catalogue, an index of the atmospheric circulation systems affecting Britain. In meteorological terms, the relative frequency of westerly and anticyclonic conditions is related to the strength of the wind field. By correlation of station wind speeds with an index of westerly/anticyclonic frequency, it is shown that west coast stations are strongly affected by the relative frequency of these two weather types, whereas the relationship at east coast stations is much weaker.
    wind power

43. [Richman 1987]
    Again coming to the old reasonins as in [Richman 1986]. This answers to [Jolliffe 1987]. PCA do not provide stable results. Rotated PCA do and bear important information with realistic representation of variability modes.
    EOF, PCA, factor analysis, rotation of principal components.

44. [Sollow 1987]
    Description of two phase regression model. Application to climate change
    Climate change, change points, two phase regression

45. [Woodruf et al. 1987]
    COADs dataset. Description of the development of the dataset
    COADS

46. [Barring 1988]
    Maximum likehood factor analysis on 73 daily precipitation time series from Kenya for 1971 to 1885. Interesting pre quality control of data and transformations to attempt disminishing the influence of skewness. Uses 3 criteria for selection of factor analysis components in S-mode (with correlation matrix) and direct oblimin rotation for selection of simple structures. Results are dependent on the number of components selected and also on the method of rotation. It is important that, as a criteria, the final regions bear some relationship to physical or geographical areas. Clustering is achieved through labelling each site with the number of the component that provides the highest loading.
    Factor analysis,daily precipitation, Kenya, regionalization, cluster analysis.

47. [Jones and Wigley 1988]
    Reconstruction and analysis of sea level pressure over the Antarctica.
    SLP antarctica

48. [Kidson 1988]
    Modes of variability SH
    SH, SAM

49. [Kidson and Trenberth 1988]
    Sampling methods. Sample sizes necesary for appropriate estimating climate parameters and statistics.
    Sample sizes. Missing data. Sampling

50. [Parker and Folland 1988]
    Descriptive on some examples about climate variability. Influence of ocean to moderate weather in winter.
    Pretencioso, non hay mensaxe novo.

51. [Preisendorfer 1988]
    Book on Principal component analyis in meteorology and oceanography
    EOFs

52. [Harrison 1989]
    Climatology of wind stress and wind stress curl
    Wind stress, curl, ocean

53. [Kushnir and Wallace 1989]
    Description of atmo. circulation patterns at 500 mb height. North Atlantic Oscillation, PNA and WA are most important. Good catalogue of maps.
    NAO, 500 mb, PNA, WA.

54. [Mason 1989]
    Good description of greenhouse effect and radiative balance. GCMs...
    GCM, climate change, CO2, pollutants.

55. [Mitchell 1989]
    Remake of [Mitchell 1984]. Good description of greenhouse effect and radiative balance. Also pollutants...
    GCM, climate change, CO2, pollutants.