Most climatological studies provide detailed information concerning homogenisation problems and indicating some helpful procedures. However, they mainly deal with temperature or precipitation data. There are not too many analyses depicting the homogenisation problems of the air pressure data. The most important ones were published recently (Bärring et al., 1999, Slonosky et al., 1999, Jones et al., 1999) and they univocally show that the subject is not schematic. It is worth to add that the pressure data and field are quite important in climatology mainly to reconstruct the atmospheric circulation and its patterns. It especially concerns the oldest pressure records.
In the frame of "Visby", the international project two Polish long-term daily series of air pressure were performed i.e. for Warsaw and for Cracow. Both of them have continuous data since 1826, and with some breaks date even from 1779 and 1792, respectively. However, some problems concerning homogenization have appeared. It especially concerns the Warsaw data where basic difficulties are connected with localisation changes including the height of the barometer. Long Cracow series of pressure, which were elaborated by J. Trepinska (1988, 1997), was not included to the European climatologies (Jones et al., 1999, Slonosky et al., 1999) although it is situated strictly at the grid point 50oN and 20oE (Jones et al., 1987, 1996). A few shorter series are also available in Poland for Wroclaw, Poznan, Bydgoszcz and Pulawy. By comparison with long series from Lvov (Ukraina) and Vilnius (Lithuania), and recently elaborated data for Lund in southern Sweden (Bärring et al., 1999), it is possible to reconstruct the more real pressure data series for Central Europe. Such data are very important for the construction of regional circulation indices for the Central Europe (Niedzwie 1996, Ustrnul 1997) similar to North Atlantic Oscillation (Hurrel 1995, Jones et al., 1997, CRU, 2000) and zonal index (Kozuchowski 1993).

Homogenisation of the mentioned series has been divided into two separate phases: the first and principal concerned the mean monthly values, the second was done for the daily data. During the first part few steps have been done. The first one concerned general data checking including the outer control. During that phase some specific limits have been established and data has been checked. These limits contained thresholds of the possible maximum and minimum values as well as the same values for the changes from term to term and from day to day. For example, the special attention was given to cases when the pressure dropped up or down over 10 hPa between the two following readings. Similar thresholds have been applied to changes between following days (i.e. over 20 hPa). If there were some doubts the comparison with the equivalent values at the second station (Cracow or Warsaw) was performed. In the next phase, checking of data due to the instrumental corrections and gravity has been provided. The special attention has to be paid to the scales applied. In old data series the pressure very often is expressed in different units. For recalculation into hectopascals (hPa) the following relations were used:
1 mm Hg = 1.333224 hPa;
1 inch Hg = 33.8639 hPa;
1''' (Paris line) = 3.0075 hPa.
Thus in the end of that step all data have been converted into one common and contemporary scale i.e. hectopascals.

Then the pressure reduction to the sea level due to the barometer's altitude and outdoor temperature has been done. It was the most difficult because in many times we did not have the precious height of the barometer and sometimes it appeared problematic. Information about that height is crucial for further processing. For reduction of the station pressure (Ps) measured at the barometer altitude (Hb) to the sea level pressure (Pm) we apply the formula used in the Polish Meteorological Service:
   Pm = exp(lnPs+2.30259Hb/(18400(1+0.003667(t+0.0025Hb))));
   where t is the air temperature and pressure is expressed in hPa.
In some cases when we have a break in the pressure measurements at the station, it is possible to reconstruct the missing value. It can be done on the basis of sea level pressure, which is possible to obtain by interpolation between the isobars on the synoptic map for the fixed term values or on the monthly maps of isobars for monthly means. For such purposes we used the same formula as below:
   Ps = exp(lnPm-2.30259Hb/(18400(1+0.003667(t+0.0025Hb))));
The next phase consists in interstation homogenization. It is based on the neighbouring station and/or the gridded values available for Europe since 1780 (Jones et al., 1996). For such purposes the difference series were compared and tested by Standard Normal Homogeneity Test - SNHT (Alexandersson, 1986). For quick analysis and testing of climatological data the software package AnClim prepared by P. Štepánek (2000) seems to be quite convenient and useful. The mentioned test indicated in easy way the difference and inhomogeneity existing between the Warsaw and Cracow data in 1871 (Fig. 1, 2). Until now we do not realise the reason of it. There are no special notes in the available metadata. Simultaneously, we discovered that the difference in 1891, which reached 0.5 mm Hg was caused by the lack of the gravity correction. The next phase consisted of corrections of daily values when some errors have been detected in the monthly values.

In pressure studies, special attention must be paid to the extreme values. They provide information on the existing inconsistency in the series of pressure data. The best solution is the comparison of these data with the pressure fields presented on synoptic maps. But the last ones date since 1873 for Europe (Tägliche Wetterbericht, 1873-1975, Europäischer Wetterbericht, 1976-2000, Historical Weather Maps, 1899-1950). For earlier periods the reconstruction of synoptic maps will be helpful for checking even the small errors and also for filling some gaps. However, for this purpose the analysis of all existing, even short period series, of the pressure data must be done. Experience of J.A. Kington (1980, 1988) with synoptic maps reconstruction for Western Europe for the end of 18-th century is quite useful.
Another possibility for checking the inconsistency in pressure data is the comparison od station data with the nearest grid point series prepared by P. Jones et al. (1996). We present one example for Cracow data (Fig. 3) which is situated strictly in the grid point. During the analysed period, few substages exist: 1780-1830, 1831-1870, 1871-1890, 1891-1950 and 1951-1980. These inconsistencies must be explained by the detailed studies with the use of some other neighbouring stations. For example, in the period 1871-1890 pressure values in Cracow seems to be too low and after 1950 unsolved trend of these differences can be observed.

Pressure data can be applied to determine the main circulation features over the analysed area. It is possible to construct them for the each grid box of the Northern Hemisphere (Ustrnul, 1997). We try to check such possibility by calculation of the two simple indices for Poland based on the 4 grid points originated from Jones et al. (1996) series. Pressure difference between 50°N and 55°N expresses the intensity of the zonal airflow in the way similar to the North Atlantic Oscillation Index. We called it Polish Zonal Oscillation Index (Fig. 4). In the same way pressure difference between 30°E and 10°E indicates the intensity of the meridional airflow - Polish Meridional Oscillation Index (Fig. 5). Figures 4 and 5 present the large variability of the mentioned indices during the instrumental period. The special attention should be paid to the high values of the Zonal index which happened in the last decade of the 19^th century and recently. Simultaneously the lowest value of the Meridional index was notified around 1880.

The air pressure homogenisation experience with the long-term series allows to distinguish few phases which are necessary in the adequate studies. They are: (I) general data checking including the outer control, (II) checking of data due to the instrumental corrections and gravity (III) pressure reduction to the sea level due to the barometer's altitude and outdoor temperature (IV) interstation homogenization based on the neighbouring station and/or the gridded vales. These four steps seem to be the basis for each typical long-term pressure series. Of course in some singular cases some other steps could be also required.

The study was partially prepared in the frame of the International Programme 'Climate variability in the Baltic region: data sources and environmental applications' co-ordinated by Prof. Lars Bärring and sponsored by the Swedish Institute.

Alexandersson H., 1986, A homogeneity test applied to precipitation data, Journal of Climatology, 6, 6, 661-675.
Bärring L., Jönsson P., Achberger C., Ekström M., Alexsandersson H., 1999, The Lund instrumental record of meteorological observations: reconstruction of monthly sea-level pressure 1780-1997, International Journal of Climatology, 19, p. 1427-1443.
CRU, 2000, North Atlantic Oscillation (NAO), Climate Research Unit, University of East Anglia, home page: http://www.cru.uea.ac.uk/cru/data/nao.htm, and:
http://www.cru.uea.ac.uk/~timo/projpages/nao_update.htm
Europäischer Wetterbericht, 1976-2000, Deutscher Wetterdienst, Offenbach a. Main. (earlier: Tägliche Wetterbericht).
Falarz M., 1997, Ekstremalne wahania cisnienia atmosferycznego w Krakowie na tle cyrkulacji atmosfery (Extreme variations of pressure in Cracow on the background of atmospheric circulation), Sympozjum Jubileuszowe Polskiego Towarzystwa Geofizycznego, "Ekstremalne Zjawiska Meteorologiczne, Hydrologiczne i Oceanograficzne", 12-14 listopad 1997, Materialy, Warszawa, p. 32-37 (in Polish).
Gorczynski W., 1916, O rozkladzie geograficznym i zmiennosci sredniej cisnienia w Polsce i Europie (About the geographical distribution and variability of pressure in Poland and Europe), Sprawozdania Towarzystwa Naukowego Warszawskiego, IX, 6, p.609-626 (in Polish).
Gorczynski W., 1917, O cisnieniu powietrza w Polsce i w Europie - Pression atmosphériques en Pologne et en Europe (About the pressure in Poland and Europe), Drukarnia i Litografja Jan Cotta, Warszawa, 265 pp. (in Polish and French).
Heino R., 1994, Climate of Finland during the period of meteorological observations, Finnish Meteorological Institute, Contributions 12, Helsinki, 209 pp.
Historical Weather Maps, 1899-1950. Daily Synoptic Series. Northern Hemisphere Sea Level. US Weather Service.
Hurrel J.W., 1995, Decadal trends in the North Atlantic Oscillation - regional temperatures and precipitation, Science, Vol. 269, No. 5224, p. 676-679.
Jones P.D., Wigley T.M.L., Briffa K.R., 1987, Monthly mean pressure reconstructions for Europe (1780-1980) and North America (1858-1980), DOE Technical Report No. TR037, U.S. Department of Energy, Carbon Dioxide Research Division, Washington, D.C.
Jones P.D., Wigley T.M.L., Briffa K.R., 1996, Dataset NDP025: Monthly mean pressure reconstructions for Europe (1780-1980) and North America (1858-1980),     CDIAC: http://cdiac.esd.ornl.gov/authors/jones.html
Jones P.D., Jónsson T., Wheeler D., 1997, Extension to the North Atlantic Oscillation using early instrumental pressure observations from Gibraltar and South-West Iceland, International Journal of Climatology, 17, 13, p. 1433-1450.
Jones P.D. (et al.), 1999, Monthly mean pressure reconstructions for Europe for the 1780-1995 period, Int. J. Climatol, 19, p. 347-364.
Kington J.A., 1980, Daily weather mapping from 1781, Clim. Change, 3, p. 7-36.
Kington J.A., 1988, The weather of the 1780s over Europe, Cambridge University Press, Cambridge, 166 pp.
Kozuchowski, K., 1993, Variations in hemispheric zonal index since 1899 and its relationship with air temperature, International Journal of Climatology, Vol. 13, p. 853-864.
Niedzwiedz T., 1996, Long-term variability of the zonal circulation index above the Central Europe, Early Meteorological Instrumental Records in Europe - Methods and Results, Zeszyty Naukowe UJ - Prace Geograficzne, Kraków, 102, p. 213-219.
Niedzwiedz T., Ustrnul Z., 1997, Experience with the homogeneization of the climate data sets in Poland. Proceedings of the First Seminar for Homogeneization of Surface Climatological Data, Budapest, Hungary, 6-12 October 1996, Hungarian Meteorological Service, p. 121-130.
Peterson T.C. (et al.), 1998, Homogeneity adjustments of in situ atmospheric climate data: a review, Int. J. Climatol., 18, p. 1493-1517.
Slonosky V.C., Jones P.D., Davies T.D., 1999, Homogenization techniques for European monthly mean surface pressure series, Journal of Climate, 12, 8, p. 2658-2672.
Štepánek P., 2000, AnClim - Software for Homogenisation and Statistical Analysis of Climatological Time Series, Department of Geography, Masaryk University, Brno     http://www.sci.muni.cz/~pest/.
Tägliche Wetterbericht, 1873-1975, Deutsche Wetterdienst.
Trenberth K.E., Paolino D.A., 1980, The Northern Hemisphere sea-level pressure data set: trends, errors and discontinuities, Mon. Wea. Rev., 108, p. 855-872.
Trepinska J., 1988, Wieloletni przebieg cisnienia i temperatury powietrza w Krakowie na tle ich zmiennosci w Europie (Long-term variation of pressure and temperature in Cracow on the backgrount of its variability in Europe), Rozprawy Habilitacyjne UJ, Nr. 140, Kraków, 169 pp. (in Polish).
Trepinska J., 1997, Wieloletni przebieg srednich miesiecznych wartosci cisnienia powietrza w Krakowie (Long-term variation of mean monthly pressure values in Cracow), 1792-1995, [in:] Wahania klimatu w Krakowie (J. Trepinska ed.), IGUJ Krakow, p. 131-161 (in Polish).
Ustrnul Z., 1997, Zmiennosc cyrkulacji atmosfery na polkuli polnocnej w XX wieku (Variability of atmospheric circulation on the Northern Hemisphere during XX Century), Materialy Badawcze, Seria: Meteorologia, 27, IMGW, Warszawa, 208 pp. (in Polish, Eng. summary).
Young K.C., 1993, Detecting and removing inhomogeneities from long-term monthly sea-level pressure time series, Journal of Climate, 6, p. 1205-1220.