THE ROLE OF HYDRODYNAMIC INSTABILITY IN A WEATHER PREDICTION

 

Marta Mann

Geofizički zavod Prirodoslovno-matematičkog fakulteta

Sveučilišta u Zagrebu

 

Abstract: The aim of this paper is to answer the question why our weather prediction for Croatia failed on June 22, 1990? It deals with the application of the meridional wind profile in estimating the direction of movement of a cyclone in a hydrodynamically unstable atmosphere. Nonlinear meridional variation of zonal flow – barotropic wind shear – has been proved to govern the movement of cyclones and to support their amplification in the vicinity of the inflexional parts of the zonal flow meridional profiles.

 

Key words: hydrodynamic instability, baroclinic and barotropic instability, meridional profiles of zonal flow, barotropic governing


MOUNTAIN DRAG ESTIMATION FROM THE OPERATIONAL SYNOPTIC NETWORK

 

Branka Ivančan-Picek, Vlasta Tutiš

 

Abstract: This paper presents the time series of the three-hourly average pressure drag vectors per unit volume for the Dinaric Alps computed separately from microbarographic and conventional pressure data. Regardless of magnitude, the pressure drag vectors seem to be aligned almost perpendicularly to the main mouton ridge. The pressure drag maxima during SOP are always connected with the Bora periods and the magnitudes of the drag values indicate that during these events there is a major sink of atmospheric momentum over the Dinaric Alpine region.

The main aim of this paper is to provide a means of drag estimation for period in which no microbarographic data are available. It investigates the correlation between the time series of pressure drag computed from microbarographic and synoptic pressure data. The results suggest the possibility of application of the linear regression method in drag estimation for periods without microbarographic data.

 

Key words: mountain drag, pressure drag vector, microbarographic data, synoptic network, correlation coefficients, ALPEX-SOP


APPLICATION OF THE HYDRAULIC THEORY TO NATURALLY OCCURRING BORA FLOW

 

Vesna Jurčec and Dražen Glasnović

 

 

Abstract: This paper deals with the application of Smith’s (1985) internal hydraulic theory to some special cases of Adriatic bora during ALPEX SOP. In particular, the case of 15 April 1982 is analysed for which the numerical simulation by Klemp and Durran (1987) is available. Their conclusion is that in this case neither inversion nor critical levels are important for the bora occurrence, but the low-level wave breaking due to weak flow in the upstream region. While this conclusion seems reasonable, from the present analysis it is argued that these results are valid only for the characteristics of the local bora in Senj and not for the flow across the higher mountains along the northern Adriatic coast.

This conclusion is based on a comparison with the results of the theory application to the stronger bora condition on the previous day and especially to the case of the strongest SOP bora on 6 March 1982 for which the hydraulic theory offered much better results.

 

Key words: Adriatic bora, hydraulic flow, wave breaking, severe winds.


STATISTICAL ANALYSIS OF SEVERE ADRIATIC BORA

 

Višnja Vučetić

 

Abstract: This paper presents the results of a statistical analysis of severe bora occurrence at two locations, Split and Dubrovnik, in mid-Adriatic. A situation with severe bora is defined as a period with wind, the direction of which is continuously between 360° and 90°, with a mean hourly speed ³17.0 ms-1 during at least one hour. In the 30-year period of observation (1958-1987) 116 situations with severe bora were registered in Split. This is four times less than in Senj, on the northern Adriatic. In Dubrovnik, severe bora is not so frequent. It is shown that severe bora with maximum gusts ³40.0 ms-1 appears along the entire Adriatic coast, but its duration and frequency decrease from north to south.

 

Key words: severe and strong bora, Split and Dubrovnik, statistical analysis of wind.


WIND PROFILE IN SURFACE LAYER OF THE ATMOSPHERE

 

Lidija Cvitan

 

Abstract: A wind model based on the Monin-Obukhov theory has been prepared for operational use over flat terrain in northern Croatia. The input parameters are air temperature at 2 and 35 m, wind speed at 35 m and a constant value for surface roughness length zo determined on the basis of terrain classification by Kondo and Yamayawa (1986). The iterative method was used to improve the accuracy of some similarity parameters. The wind profiles showed considerable dependence upon stability. Therefore special stability categories were introduced, based on the values of bulk Richardson number. Based on the similarity theory universal functions, the local functions were derived for each stability class. These functions were included into the model and only then a satisfactory speed profile simulation was achieved.

 

Key words: Monin-Obukhov theory, universal functions, iterative method, stability classification


CORRELATION BETWEEN METEOROLOGICAL FACTORS AND BLOOD COAGULATION PARAMETERS IN CEREBROVASCULAR PATIENTS

 

Nada Pleško1, Miroslava Klein-Pudar2, Ksenija Zaninović1

 

Abstract: This article presents the results of measurements of the various blood coagulation parameters and blood pressure of cerebrovascular patients in different weather conditions during the year. The air temperature involved was from 2-25°C and barometric pressure form 987-1004 hPa. Measurements have been performed through 6 three-day series, with a cold front passage over Zagreb in the central day of every series.

The results show that a significant correlation exists only between air temperature and both prothrombine and thrombine time. Prothrombine and thrombine times were longer with the higher air temperature. A highly significant correlation (r = -0.465, P<0.01) was obtained between the fibrinogen quantity and barometric pressure, showing that if the weather situation has a lower barometric pressure, the fibrinogen quantity is higher.

With regard to cold front passage, the mean prothrombine time is shortened from “n-1” to “n+1” day, while the mean thrombine time oscillates around the day with front passage and it is the shortest one day after front passage. The mean fibrinogen quantity is at its highest on the day before the front passage and at its lowest on the day of front passage.

 

Key words: cerebrovascular patients, cold front passage, air temperature, barometric pressure, blood coagulation parameters, blood pressure


CORRELATION OF PSYCHOSIS AND SUICIDE ATTEMPTS WITH METEOROLOGICAL FACTORS

 

Nada Pleško1, Vlasta Jasprica-Hrelec2, Ingrid Bošan-Kilibarda2, Berislav Banek2

 

Abstract: In this study the authors have tried to find out if there is any correlation between weather characteristics and acute psychoses and suicidal attempts (APSA). This research was based on SHMP data about the number of interventions in cases of APSA over a period of one year (VI/ 1988-V/1989) in Zagreb and on the meteorological data registered at the Observatory Zagreb-Maksimir in the same period as well as on weather charts prepared daily or collected in the Hydrometeorological Institute of Croatia in Zagreb. A possible correlation with weather types and passages of cold fronts and especially with meteorological elements as: cloudiness, barometric pressure, air temperature and relative humidity, was studied. Linear correlation coefficients were calculated for daily values and for moving three-, five- and seven-day periods.

The results revealed that the number of psychoses and suicidal attempts increased from month to month during the period considered, especially in females of the 15-20, 21-30 and 31-40 age groups. The acute psychosis and suicidal attempt number does not significantly differ with respect to weather type. The daily frequency of these pathological reactions increases in all seasons typically a day or two before and after a cold font passage, but differences are not significant. A significant correlation was obtained only with the duration of a high daily cloudiness amount. Suicidal attempts and psychoses increase significantly if the cloudy period lasts at least three days, but their significance is higher if the cloudy weather period is longer.

 

Key words: suicidal attempts, acute psychoses, weather types, passages of cold fronts, duration and cloudiness amount.


A STATISTICAL ANALYSIS OF INVERSIONS AND MIXING LAYER HIGHS IN ZAGREB AREA

 

Edita Lončar

 

Abstract: A statistical analysis of surface inversions, elevated inversions and mixing heights over Zagreb region has been performed by means of radio-sonde observations at Zagreb-Maksimir at 00 and 12 UTC (1977-82).

Annual frequency of all night inversions is 90%, and of all day inversions are about 65%. During all seasons the surface inversions most frequently reach the high of about 100 m. Height surface inversions are winter characteristics and having a persistency up to several days. Elevated inversion heights show a considerable seasonal and diurnal variability with their lower basis being much lower during the colder part of a year.

Mixing layer heights are greater during summer than during winter time.

 

Key words: ground (surface) inversions, elevated inversion, mixing height, statistical analysis, the area of Zagreb.


ANALYSIS OF PHENOLOGICAL CHARACTERISTICS IN THE REGION OF MOLVE

 

Višnja Vučetić

 

Abstract: The development stages of perennial trees (apricot, walnut, common spruce and locust-tree) al well as stages of honey bees were analysed in Đurđevac and Bjelovar in the periods before (1971-1980) and after (1981-1990) the marsh gas station Molve was putting in operation.

It has been shown that differences between the considered periods do exist. Colder winters in the second period caused late leafing and blooming of the common spruce, apricot and walnut trees. Warmer springs provoked earlier vegetation of the locust-trees and earlier stages of bees. However, the variations established were not of local character having been observed on both locations at the same time.

 

Key words: Phenological observations, the marsh gas station Molve.


CLIMATOLOGICAL OBSERVATION TIMES AT 07, 14 AND 21 LMT ARE RETAINED

 

Ivo Lukšić

 

Abstract: Yugoslavia has for a long time been considering whether it would be justified to change the climatological observation times at 07, 14 and 21 LMT to 06, 12 and 18 GMT. Finally, it has been decided to retain the present climatological observation times at 07, 14 and 21 LMT. The main reasons for this decision are: the special importance of data collected at these three climatological observation times, a better daily distribution of the climatological observation times at 07, 14 and 21 LMT, difficulties with recording instruments, satisfying climatological needs, homogeneity, and the fact that more frequent objections concerning the 21 LMT data quality are not justified.

 

Key words: climatological observation times.


SYSTEMATIC ERRORS IN DETERMINISTIC EXTENDED – RANGE WEATHER FORECASTS

 

Čedomir Branković

 

D.Sc.Thesis – Zagreb University, Faculty of Natural Science,

Zagreb, 1990, 111 pp.

 

 

Summary: The results from a set of 30-day integrations made with the ECMWF model over a period of two years are discussed. The impact of spectral horizontal resolution and of model changes on the model systematic errors and forecast skill in the extended range is studied.

The 30-day mean systematic errors in wind and geopotential heights reveal that, among the four horizontal resolutions (T21, T42, T63 and T106) considered, there is a distinct difference in behaviour between the T21 and other models, and a number of similarities between T42, T63 and T106 models. Due to unrealistic representation of eddy momentum flux, the T21 model significantly underestimates zonal flow in the southern hemisphere. On the other hand, too strong westerlies in the northern hemisphere at T42, T63 and T106 resolutions are the consequence of poor representation of the effects of small and subgrid-scale orography. In higher resolution models, these effects act, via better representation of planetary waves and sensible heat flux, as to improve thermal balance in the model stratosphere, ultimately leading to smaller errors than in T21 and T42 models.

Systematic errors in the tropics are large and the influence of horizontal resolution on tropical flow is generally weaker than in the extratropics. Errors in the tropics are related to inadequate simulation of diabatic forcing and are primarily associated with the parameterisation of deep convection and partially with model radiation. Apart form the rainfall, tropical errors are not much dependent on representation of orography in the model.

It is concluded that, because of its strong climate drift, the T21 model is the least suitable for extended-range and climate integrations. Both T106 and T63 models have smaller climate drift when compared with T21; however, they are probably too detailed for extended-range and climate integrations, especially having in the mind the computational burden for long integrations at these resolutions. Tropospheric systematic errors in the T42 model are quite comparable with those of higher resolution models, and in view of computational coast the T42 is a good compromise for climate runs.

Changes in model formulation, parameterisation of orographic gravity wave drag and an increase in vertical resolution, act as to reduce the model systematic errors. The effects of these changes are most pronounced beyond the medium range, thus confirming the importance of diagnosing systematic errors in the extended range for the model primarily designed for medium-range forecasting.

The skill scores are comparable at higher resolution models and are better than those at T21. Time averaging improves the model skill scores, the largest improvement being for T106 model at days 10-15. This is certainly encouraging, since it coincides with the limit of theoretical predictability. Skill scores are largely variable and depend on the development of the large scale atmospheric systems such as blocking. The model did not prove successful in predicting realistic transitions between various large scale flow types. Interannual variability of the mean flow also affects forecast skill, and therefore improvements in the model climatology do not necessarily reflect in improved skill scores. Whilst, in principle, spread can be used as a predictor for skill, it was found that the relationship between spread and skill is not always unique and reflects the influence of the annual cycle.


A TWO-LAYER SCALAR MODEL FOR POLLUTANT DISPERSION IN URBAN CONDITIONS

 

Marina Grčić

 

M.Sc. Thesis – Zagreb University, Faculty of Natural Sciences,

Zagreb, 1990, 116 pp.

 

Summary: A two-layer model (DSM) of urban air pollution is developed which describes temporal variations of pollutant concentrations.

Any model for the dispersion of pollutants over a city must take into consideration the effects of the city on the large-scale air flow and the complexity of urban pollutant sources. A qualitative description of the urban effect on climate and dispersion mechanisms is first given, and then a review is made of the capabilities of modern modelling techniques to simulate and elucidate these processes.

The DSM introduced the concept of two distinct urban layers in order to characterise the energy processes in an urban atmosphere. One, called the urban canopy layer (UCL), extends from the ground up to about mean roof level; the other, called the urban boundary layer (UBL), is a mesoscale internal boundary layer the characteristics of which are determined, at least partially, by the presence of the city beneath.

The DSM is formulated by using the concentration equation based on the principle of mass conservation. The equations are analytically solved for each layer with respect to the vertical turbulent pollutant flux between the layers.

Based on the flux-gradient theory an iterative procedure has been developed for the estimation of the turbulent pollutant flux which is expressed here as a function of atmospheric stability and difference of concentration between the layers.

The important input parameters of the DSM are: sensible heat flux, wind, air temperature, UBL height, emission rate, urban size and structure.

Where there is a lack of measured data several techniques are recommended for their evaluation.

The DSM has been used to simulate temporal variations of pollutant concentration in the UCL and UBL in a synthetic city during the observed meteorological conditions which included clear skies, low humidity and high pressure.

The simple, one-layer scalar model (often used as control model) has been applied to the same set of input parameters. A comparison of the results shows several advantages of the DSM.

The complexity of the relationship between input parameters, mathematical handing of physical processes and output values often makes it difficult to assess exactly what effect the change in a particular input should have had.

Sensitivity analyses were conducted on all inputs by varying a specific input over a range of typical urban values while all others were at chosen values. These analyses were undertaken because of uncertainty about the values to be assigned to some of the input parameters and to assess the model in terms of the physical reality of the results.

The sensitivity analyses showed the DSM’s general flexibility and applicability in handling complex emissions and atmospheric conditions in an urban area.