Vrednotenje vpliva prestrezanja padavin na kinetično energijo dežnih kapljic : magistrsko delo

Radulović, Lana (avtor), Šraj, Mojca (mentor), Zabret, Katarina (komentor)

Med procesom prestrezanja padavin vegetacija posredno vpliva na različne naravne procese, med drugim tudi na erozijo tal. Kot kazalnik erozijskega potenciala padavin se pogosto uporablja kinetična energija (KE) dežnih kapljic. Sprememba mikrostrukture dežnih kapljic po stiku z vegetacijo pomembno vpliva na spremembo njihove kinetične energije. Cilj te naloge je ovrednotiti vpliv prestrezanja padavin na kinetično energijo dežnih kapljic pod krošnjami dveh različnih drevesnih vrst, v različnih fenoloških fazah in v različnih meteoroloških pogojih. V obravnavanem dvoletnem obdobju smo s tremi optičnimi disdrometri pridobili 1-minutne podatke o padavinah na prostem ter o prepuščenih padavinah pod krošnjami breze in bora. Skupaj je bilo v obravnavanem obdobju zaznanih 171 padavinskih dogodkov, ki smo jih razdelili na suho in mokro podobdobje ter na dve fenološki fazi, in sicer na obdobje vegetacije in obdobje mirovanja. Za vsak padavinski dogodek smo izračunali velikosti kapljic (D50) in njihovo KE. Dodatno smo analizirali tudi podatke o meteoroloških razmerah med dogodki. Za oceno odnosa med vplivnimi spremenljivkami in KE dežnih kapljic smo uporabili dve metodi strojnega učenja, metodo ojačanih regresijskih dreves (BRT) in metodo naključnega gozda (RF). Z obema metodama smo analizirali dve ločeni skupini podatkov, in sicer vpliv značilnosti padavin ter vpliv značilnosti prepuščenih padavin na KE dežnih kapljic pod krošnjami dreves. Rezultati kažejo, da na KE dežnih kapljic močno vpliva prisotnost listov na krošnji, saj je breza veliko bolj vplivala na znižanje KE v obdobju vegetacije. Glede na celotno obravnavano obdobje je breza v povprečju znižala KE dežnih kapljic za 32 % v obdobju vegetacije, v obdobju mirovanja pa za 18 %. Pomembno vlogo ima tudi vrsta vegetacije, saj bor prestreže večjo količino padavin kot breza (za 37.5 % večjo glede na celotno obravnavano obdobje) in s tem pomembneje vpliva tudi na KE dežnih kapljic. Obe metodi strojnega učenja sta pokazali, da je količina padavin oziroma prepuščenih padavin najvplivnejša spremenljivka, ne glede na obdobje ali tip vegetacije. Rezultati kažejo tudi, da ima trajanje padavin večji vpliv na KE dežnih kapljic v obdobju, ko je krošnja olistana ter da imata D50 in maksimalna hitrost vetra večji vpliv v primeru, ko obravnavamo vpliv prepuščenih padavin na KE dežnih kapljic pod krošnjami dreves.

• Jezik: Slovenski jezik
• Ključne besede: magistrska dela, gradbeništvo, prestrezanje padavin, prepuščene padavine, kinetična energija, vegetacije, strojno učenje
• Vrsta gradiva: Magistrsko delo/naloga
• Tipologija: 2.09 - Magistrsko delo
• Organizacija: FGG - Fakulteta za gradbeništvo in geodezijo
• Kraj izida: Ljubljana
• Založnik: [L. Radulović]
• Leto izida: 2025
• Št. strani: 1 spletni vir (1 datoteka PDF (XI, 64 str.))
• PID: 20.500.12556/RUL-170209
• UDK: 551.578.1:533.73(043.2)
• COBISS.SI-ID: 241204995
• Datum objave v RUL: 02.07.2025

Sekundarni jezik

• Jezik: Angleški jezik
• Naslov: Evaluation of the effect of rainfall interception on raindrop kinetic energy : master thesis

Izvleček

During the process of interception, vegetation indirectly influences various natural processes, including soil erosion. Kinetic energy (KE) of raindrops is often used as an indicator of the erosive potential of rainfall. The change in the microstructure of raindrops after contact with vegetation has a significant influence on the change in their kinetic energy The aim of this thesis is to evaluate the influence of precipitation interception on the kinetic energy of raindrops under the canopy of two different tree species, in different phenological phases and under different meteorological conditions. Over the two-year period, 1-minute data on rainfall in an open location and on throughfall under the birch and pine canopies were obtained with three optical disdrometers. A total of 171 precipitation events were recorded, divided into dry and wet sub-periods, and classified into two phenological phases, namely the leafed and the leafless. For each rainfall event, we calculated the size (D50) and KE of the raindrops. In addition, we analysed data on meteorological conditions during the events. To assess the relationship between the influencing variables and the KE of the raindrops, we used two machine learning methods, namely the boosted regression trees (BRT) method and the random forest (RF) method. The two methods were applied to two sets of data, namely the influence of rainfall characteristics and the influence of throughfall characteristics on the KE of drops under the tree canopy. The results show that the KE of raindrops is strongly influenced by the presence of leaves in the canopy, with birch having a much greater effect on the decrease of KE during the leafed season. On average, birch reduced raindrop KE by 32% on average during the leafed season and by 18% during the leafless season over the whole period considered. The type of vegetation also plays an important role, as pine intercepts a higher amount of rainfall than birch (by 37.5% more over the entire period considered) and thus has a more significant effect on the KE of raindrops. Both machine learning methods showed that the amount of precipitation is the most influential variable, regardless of the period or vegetation type. The results also show that throughfall duration has a greater influence on the KE during the period when the canopy is leafed, and that D50 and maximum wind speed have a greater influence when considering the influence of throughfall on KE of raindrops under the tree canopy.

• Ključne besede: master thesis, civil engineering, rainfall interception, throughfall, kinetic energy, vegetation, machine learning