prikaz prve stranice dokumenta Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill)
  Download
PDF 1.59 MB
master's thesis
Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill)
2023. urn:nbn:hr:151:414978
Spišić, Marija
Josip Juraj Strossmayer University of Osijek
Faculty of Agrobiotechnical Sciences Osijek
Department of Plant Production and Biotechnology

Cite this document

Spišić, M. (2023). Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill) (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:151:414978

Spišić, Marija. "Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill)." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:151:414978

Spišić, Marija. "Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill)." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:151:414978

Spišić, M. (2023). 'Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill)', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek, accessed 13 February 2025, https://urn.nsk.hr/urn:nbn:hr:151:414978

Spišić M. Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill) [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek; 2023 [cited 2025 February 13] Available at: https://urn.nsk.hr/urn:nbn:hr:151:414978

M. Spišić, "Usporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill)", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Agrobiotechnical Sciences Osijek, Osijek, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:151:414978

Please login to the repository to save this object to your list.
Metadata
TitleUsporedba različitih metoda praćenja vodnog stresa kod soje (Glycine max (L.) Merrill)
Title (english)Comparison of various methods for monitoring drought stress in soybean (Glycine max (L.) Merrill)
AuthorMarija Spišić
MentorMonika Marković (mentor)
Committee memberIrena Rapčan (predsjednik povjerenstva)
Committee memberMaja Matoša Kočar (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Agrobiotechnical Sciences Osijek
(Department of Plant Production and Biotechnology)
Osijek
Defense date and country2023-10-20, Croatia
Scientific / art field,
discipline and subdiscipline		BIOTECHNICAL SCIENCES
Agronomy
Crop Production
Abstract
Istraživanje je provedeno na Poljoprivrednom institutu Osijek. Pokus je postavljen u vegetacijskim posudama u plasteniku s jednom sortom soje (OS Nevena). Pokus je posijan 25. travnja 2022. godine. Biljke su podijeljene u dva tretmana navodnjavanja – jedan u optimalmnim uvjetima vlažnosti tla (n1, 100 % retencijskog kapaciteta za vodu, Rkv), a druge su bile izložene stresu izazvanom nedostatkom vode (n2, 50 % Rkv) u fazi cvatnje (f1) i nalijevanju zrna (f2). Obroci navodnjavanja su prilagođeni stanju vlažnosti tla, odnosno tretmanu navodnjavanja te su bili u rasponu od 2 do 8 L tijekom mjeseca svibnja, od 0 do 10 L tijekom mjeseca lipnja te od 0 do 8 L tijekom mjeseca srpnja. Norma navodnjavanja je tijekom mjeseca svibnja bila 76 L (n1f1+f2) te 68 L (n2f1 i n2f2). Tijekom mjeseca lipnja je norma navodnjavanja bila 154 L (n1f1+f2 i n2f2) i 126 L (n2f1) te 128 L(n1af1+f2), 120 L (n2f1) i 98 L (n2f2) tijekom mjeseca srpnja. Senzori za mjerenje vlažnosti tla su zabilježili promjene vlažnosti tla, odnosno dinamiku vlažnosti tla na svim tretmanima navodnjavanja. Najbrža reakcija senzora je zamijećena kod TDR300 i AT senzora koji mjere volumni sadržaj vode u tlu (%vol.). Tenziometar i Watermark su pokazali ujednačenije rezultate mjerenja u uvjetima povoljne vlažnosti, odnosno na n1 (kontrolnom) tretmanu. Oscilacije temperature lista u oba tretmana su pratile oscilacije temperature okoline. Također, zabilježen je negativan utjecaj suše na sadržaj klorofila u listu. Osim toga, stres izazvan sušom nije previše utjecao na maksimalni kvantni prinos fotosustava II. Indeks fotosintetske aktivnosti (PIABS) je pokazao niže vrijednosti u tretmanu zasušivanja (n2) u odnosu na kontrolu (n1) te smanjenje vrijednosti PIABS do kojih je došlo trećeg dana zasušivanja u nalijevanju. Sveukupno, modeli kNN i neuronska mreža su postigli najbolje performanse u ovom istraživanju, dok su modeli SVM i logistička regresija ostvarili nešto niže vrijednosti. Zaključuje se da je kNN model najuspješniji u klasifikaciji stresa i da je najpouzdaniji u predviđanju pozitivne klase (stres). Model kNN postigao je visoku vrijednost AUC-a od 0,976. Također, model je postigao visoku vrijednost CA (0,921), što ukazuje na to da je model vrlo precizan u klasifikaciji obje klase.
Abstract (english)
The research was conducted at the Agricultural Institute Osijek. The experiment was set up in vegetation pots in a greenhouse with one soybean variety (OS Nevena). The sowing date was April 25, 2022. The plants were divided into two treatments - one in optimal soil moisture conditions (n1, 80% Water holding capacity, WHC), and the other were exposed to stress caused by a lack of water (50% WHC), in the flowering (f1) and grain filling (f2) phase. Irrigation rations ranged from 2l at the beginning of the research to 8l at the end of May. On average, the soil water content ranged from 49.00 cbar (n2f1) to 60.68 cbar (n2f2). Sensors for measuring soil moisture registered changes, i.e. the dynamics of soil moisture in all treatments. The fastest sensor response was observed with the TDR300 and AT sensors that measure the volume content of water in the soil (% vol.). Tensiometer and Watermark showed more uniform measurement results in conditions of favorable humidity, i.e. in the n1 treatment. Leaf temperature oscillations in both treatments followed ambient temperature oscillations. Also, a negative impact of drought on the content of chlorophyll in the leaf was recorded. In addition, drought stress does not have too much influence on the maximum quantum yield of photosystem II. The index of photosynthetic activity (PIABS) shows lower values in the drying treatment (n2) compared to the control (n1) and a decrease in PIABS values that occurred on the third day of drying in pouring. Overall, the kNN and neural network models achieved the best performance in this study, while the SVM and logistic regression models achieved slightly lower values. We conclude that the kNN model is the most successful in stress classification and is the most reliable in predicting the positive class (stress). The kNN model achieved a high AUC value of 0.976. Also, the model achieved a high CA value (0.921), which indicates that the model is very accurate in the classification of both classes.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:151:414978
Study programmeTitle: Plant Production; specializations in: Plant Production, Plant Nutrition and Soil Science, Plant Selection and Seed Production, Plant Protection
Course: Plant Production
Study programme type: university
Study level: graduate
Academic / professional title: magistar/ magistra inženjer/inženjerka bilinogojstva (magistar/ magistra inženjer/inženjerka bilinogojstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Public notediplomski rad
Created on2023-11-07 08:25:21