Analiza kratkorajnih jakih količina oborine predstavlja važan interes u inženjerskoj hidrološkoj praksi, osobito prilikom planiranja i projektiranja objekata koji moraju podnijeti ekstremne količine oborine. Dosadašnja istraživanja ekstremnih oborina u Hrvatskoj ukazala su na prostorne različitosti njihovih trendova. Stoga su u ovom radu ispitane promjene maksimalnih godišnjih kratkotrajnih količina oborine za trajanja od 10 min do 2 h u razdoblju od 1955. do 2010. godine na dvije meteorološke postaje koje imaju različite oborinske režime: Split (maritimni) i Varaždin (kontinentalni). Analiza je provedena procjenom trenda iz razdoblja 1955.- 2010. te usporedbom procjena očekivanih kratkotrajnih maksimuma primjenom opće razdiobe ekstrema (GEV razdioba) iz dva kraća razdoblja: 1955.-1980. i 1981.-2010. Rezultati trenda na obje postaje ukazuju na slabo izražene, statistički neznačajne, vremenske promjene kratkotrajnih količina oborine od sredine 20. stoljeća. S druge strane, usporedba dvaju razdoblja ukazuje na slabo izraženu prisutnost pojačanja kratkotrajne oborine u novijem razdoblju na obje postaje. U radu je dodatno naglašena i važnost korištenja intervala povjerenja kojima se utvrđuje nesigurnost procjena očekivanih maksimuma za primjenu u praksi. Osim toga, zaključeno je da se uzroci sve učestalijih problema koji se javljaju s oborinskom odvodnjom pri pojavama intenzivnih oborina moraju tražiti izvan domene mogućih vremenskih promjena značajki samih kratkotrajnih jakih oborina..; Analysis of short-term heavy precipitation presents an important aspect of hydrologic engineering, particularly in relation to planning and design of facilities which are exposed to extreme quantities of precipitation. Previous investigations of extreme precipitation in Croatia revealed spatial differences of their trends. This paper, therefore, contains testing of changes in maximum annual short-term precipitation quantities in the duration from 10 minutes to 2 hours in the period from 1955 to 2010 at two meteorological stations with different precipitation regimes: Split (maritime) and Varaždin (continental). The analysis was conducted by assessment of trends in the period from 1955 – 2010, and a subsequent comparison of assessments od expected short-term maximums by application of Generalized Extreme Value distribution (GEV distribution) from two shorter periods: 1955 -1980 and 1981 - 2010. The trend results at both stations indicate poorly pronounced, statistically insignificant temporal changes in shortterm precipitation quantities from the middle of the 20th century. On the other hand, the comparison of the two periods indicates a poorly pronounced presence of increased short-term precipitation in a more recent period at both stations. The paper further emphasizes the importance of utilization of confidence intervals to determine uncertainty of assessments of expected maximums for practical implementation. Additionally, a conclusion was drawn that the causes of increasingly frequent problems related to precipitation drainage during intense precipitation events should be looked for outside of the area of potential temporal changes of short-term heavy precipitation characteristics.