The increasing intensity and frequency of precipitation in recent decades, combined with the human interventions in watersheds, has drastically altered the natural regimes of water and sediment transport in watersheds over the whole contiguous United States. Sediment-transport related concerns include the sustainability of aquatic biology, the stability of the river morphology, and the security and vulnerability of various riverine structures. For the present context, the concerns are related to the acceleration of upland erosion (sediment production) and in-stream sediment-transport processes that eventually lead to sediment accumulation at culverts (structures that pass streams under roadways). This nuisance has become widespread in many transportation agencies in the United States, as it has a direct bearing on maintaining normal culvert operations during extreme flows when these waterway crossings are essential for the communities they serve. Despite the prevalence of culvert sedimentation, current specifications for culvert design do not typically consider aspects of sediment transport and deposition. The overall study objective is to systematically identify the likelihood of culvert sedimentation as a function of stream and culvert geometry, along with landscape characteristics (process drivers of culvert sedimentation) in the culvert drainage area. The ideal approach for predicting sedimentation is to track sediment sources dislocated from the watershed, their overland movement, and their delivery into the streams using physical-based modeling. However, there are considerable knowledge gaps in addressing the sedimentation at culverts as an end-to-end process, especially in connecting the upland with in-stream processes and simulating the sediment deposition at culverts in non-uniform, unsteady flows, while also taking into account the vegetation growth in culverts’ vicinity. It is, therefore, no surprise that existing research, textbooks, and guidelines do not typically provide adequate information on sediment control at culverts. This dissertation presents a generalizable data-driven framework that integrates various machine-learning and visual analytics techniques with GIS in a web-based geospatial platform to explore the complex environmental processes of culvert sedimentation. The framework offers systematic procedures for (1) classifying the culvert sedimentation degree using a time-series of aerial images; (2) identifying key process-drivers from a variety of environmental and culvert structural characteristics through feature selections and interactive visual interfaces; (3) supporting human interactions to perceive empirical relationships between drivers and the culvert sedimentation degree through multivariate Geovisualization and Self-Organizing Map (SOM); and (4) forecasting culvert sedimentation potential across Iowa using machine learning algorithms. Developed using modular design and atop national datasets, the framework is generalizable and extendable, and therefore can be applied to address similar river management issues, such as habitat deterioration and water pollution, at the Contiguous US scale. The platform developed through this Ph.D. study offers a web-based problem-solving environment for a) managing inventory and retrieving culvert structural information; b) integrating diverse culvert-related datasets (e.g., culvert inventory, hydrological and land use data, and observations on the degree of sedimentation in the vicinity of culverts) in a digital repository; c) supporting culvert field inspections and real-time data collection through mobile devices; and d) hosting the data-driven framework for exploring culvert sedimentation drivers and forecasting culvert sedimentation potential across Iowa. Insights provided through the data-driven framework can be applied to support decisions for culvert management and sedimentation mitigation, as well as to provide suggestions on parameter selections for the design of these structures.

<p>Erozija predstavlja vrlo složen fizički proces u kome pod dejstvom atmosferskih sila nastaju&nbsp;destruktivne promene na povr&scaron;inskom sloju zemlji&scaron;ta. Kada je osnovni agens pokretanja&nbsp;čestica zemlji&scaron;ta vetar, govori se o eroziji vetrom ili eolskoj eroziji. Eolska erozija je&nbsp;specifičan proces koji se odvija u izrazito složenim okolnostima uzajamnog delovanja&nbsp;brojnih prirodnih i antropogenih faktora uglavnom stohastičkog karaktera, zbog čega je&nbsp;njeno istraživanje kompleksan naučno-istraživački problem.</p><p>Osnovni cilj sprovedenih istraživanja je bio da se na izabranim lokalitetima Deliblatske&nbsp;pe&scaron;čare, najvećeg i najznačajnijeg područja takve vrste u Evropi, uspostavi praćenje stanja&nbsp;procesa eolske erozije, odnosno neposrednim, sistematskim merenjima u terenskim&nbsp;uslovima ustanove količine eolskog nanosa &ndash; pronos nanosa, odrede dominantni pravci&nbsp;njegovog kretanja, defini&scaron;e njegova unutargodi&scaron;nja raspodela i utvrdi efekat vegetacije na&nbsp;smanjenje intenziteta eolske erozije. U toku četvorogodi&scaron;njeg perioda su po prvi put na&nbsp;ovim prostorima, na istom lokalitetu, sprovedena uporedna istraživanja eolske erozije&nbsp;primenom mehaničkih hvatača nanosa i metoda zasnovanih na praćenju aktivnosti&nbsp;radionuklida ve&scaron;tačkog porekla 137Cs u zemlji&scaron;tu, čije količine služe da se posebnim&nbsp;teorijskim modelima pretvore u gubitke zemlji&scaron;ta.</p><p>Praćenje procesa eolske erozije je vr&scaron;eno od 2006. do 2009. godine na lokalitetu Cvjićev vis,&nbsp;koji je izabran kao karakterističan, jer je pored centralne pozicije na Deliblatskoj pe&scaron;čari,&nbsp;obrađivani povr&scaron;inski sloj zemlji&scaron;ta bio bez za&scaron;tite od vetra. Paralelno sa ovim merenjima,&nbsp;na lokalitetu Dragićev hat &ndash; rasadnik, praćenje procesa eolske erozije vr&scaron;eno je na dva&nbsp;merna mesta u periodu od maja 2006. godine do aprila 2007. godine u uslovima postojanja&nbsp;za&scaron;titne uloge vegetacionog pokrivača i/ili vegetacionog pojasa. Merenje intenziteta eolske&nbsp;erozije je realizovano statičnim hvatačima nanosa tipa deflametar (dimanzija ulaznog&nbsp;otvora 10 x 10 cm) orijentisanih prema određenim pravacima (N, NE, E, SE, S, SW, W, NW) da bi potpuno definisali procese eolske erozije u vektorskom smislu. Kvantifikacija eolskog nanosa statičnim hvatačima stalno usmerenim prema određenim pravcima duvanja vetrova, omogućila je da se posebno evidentiraju i razlikuju &bdquo;sumarna&rdquo; produkcija eolskog nanosa (aritmetički zbir zahvaćenih količina nanosa iz svih hvatača) i &bdquo;rezultujuća&rdquo; količina eolskog nanosa (vektorski zbir), jer te veličine određuju pored ukupno pokrenute količine eolskog nanosa (produkcija nanosa) i delove nanosa koji se transportuju van granica&nbsp;erozionog polja (gubitak zemlji&scaron;ta), generalni pravac i smer njegovog kretanja.</p><p>Merenja eolske erozije na lokalitetu Cvijićev vis za period 2006-2009. godine su ukazala na&nbsp;značajne procese eolske erozije koji su definisani srednjim godi&scaron;njim pronosom nanosa od&nbsp;4,48 kg m-1. Ustanovljeno je da su ukupno zahvaćene količine nanosa na 8 hvatača bile &Sigma; =&nbsp;25,94, 20,92, 52,98 i 43,47 kg m-1 god-1, a pronosi nanosa &Sigma;&#39; (&Sigma;/8) = 3,24, 2,61, 6,62 i 5,43 kg&nbsp;m-1 god-1 u 2006., 2007., 2008. i 2009. godini respektivno.</p><p>Pronosi nanosa za period od maja 2006. do aprila 2007. godine su pokazali da je na<br />lokalitetu Cvijićev vis koga karakteri&scaron;u neobraslost i otvorenost povr&scaron;ine zabeležen najveći&nbsp;intenzitet eolske erozije na godi&scaron;njem nivou koji je 4 puta veći u odnosu na lokalitet&nbsp;Dragićev hat &ndash; rasadnik I koga karakteri&scaron;u neobraslost i za&scaron;tićenost povr&scaron;ine i 30,2 puta veći&nbsp;u odnosu na Dragićev hat &ndash; rasadnik II koga karakteri&scaron;u obraslost i za&scaron;tićenost zemlji&scaron;ta. Na&nbsp;lokalitetu Dragićev hat &ndash; rasadnik I intenzitet eolske erozije na godi&scaron;njem nivou je 7,5 puta&nbsp;veći od onog na lokalitetu Dragićev hat &ndash; rasadnik II.</p><p>Gubici zemlji&scaron;ta izraženi preko rezultujućih mesečnih pronosa nanosa (vektorski zbir)<br />iznosili su 5,13, 2,04, 4,31 i 11,94 kg m-1 u 2006., 2007., 2008. i 2009. godini respektivno, a&nbsp;procentualni udeo godi&scaron;njih gubitaka zemlji&scaron;ta u odnosu na ukupnu produkciju nanosa&nbsp;(aritmetički zbir) se kretao od 8,1% do 27,5. Rezultujući pravac kretanja eolskog nanosa u&nbsp;toku perioda istraživanja bio je jugoistok - severozapad (SE-NW) pod uticajem dominantnog&nbsp;jugoistočng vetra &bdquo;Ko&scaron;ava&rdquo;.</p><p>Iako se najpouzdanije determinisanje eolske erozije i njenih efekata postiže na osnovu&nbsp;neposrednih sistematskih merenja u realnim terenskim uslovima, počev od kraja pro&scaron;log&nbsp;veka se sve vi&scaron;e primenuju i metode praćenja radionuklida iz radioaktivnih padavina,&nbsp;posebno 137Cs, u cilju procene gubitaka zemlji&scaron;ta i prostornog rasporeda eolskog nanosa.</p><p>Ukupan broj uzetih uzoraka za metodu praćenja količina 137Cs je iznosio 149, od kojih je bilo&nbsp;9 inicijalnih uzoraka (3 profila po 3 uzorka), 14 osnovnih uzoraka (2 profila po 7 uzoraka), 32&nbsp;ostala uzorka (8 profila po 4 uzorka), 36 referentnih uzoraka (9 profila po 4 uzorka) i 58&nbsp;uzoraka uzetih zrakasto po određenim pravcima (N, NE, E, SE, S, SW, W, NW).</p><p>Osnovni uzorak na neobrađenom zemlji&scaron;tu sa detektovanom količinom 137Cs od 10.603,57&nbsp;Bq m-2 predstavlja uporednu vrednost, tj. lokalni padavinski ulaz 137Cs za modele&nbsp;pretvaranja količina 137Cs u količine izgubljenog zemlji&scaron;ta. Ovaj uzorak odslikava sredinu na&nbsp;kojoj su sprovedena istraživanja i predstavlja uporednu vrednost koja može korektno da&nbsp;defini&scaron;e procese eolske erozije.</p><p>Za pretvaranje izmerenih količina 137Cs u količine izgubljenog zemlji&scaron;ta po modelima&nbsp;Walling-a kori&scaron;ćen je najjednostavniji proporcionalni model (PM) za obrađena zemlji&scaron;ta, a&nbsp;najprimenjeniji model profilne distribucije (PDM) za neobrađena zemlji&scaron;ta u okviru najnovije verzije PC-kompatibilnog softverskog paketa u Microsoft Excel Add-Ins varijanti. Modelom profilne distribucije (PDM) za sve uzorake uzete na eksperimentalnom području dobijeni su prosečni gubici zemlji&scaron;ta od 207,06 t ha-1 god-1 i 2,10 cm. Model Basher &amp; Webb je dao prosečne gubitke zemlji&scaron;ta od 212,18 t ha-1 god-1 i 2,09 cm. Gubici zemlji&scaron;ta dobijeni modelima pretvaranja količina 137Cs u količine izgubljenog zemlji&scaron;ta ukazuju na značajne procese eolske erozije definisane jakom i ekscesivnom eolskom erozijom.</p><p>Za 58 uzoraka zemlji&scaron;ta uzetih na tačkama raspoređenih zrakasto po određenim pravcima&nbsp;(N, NE, E, SE, S, SW, W, NW), pored koncentracija i količina 137Cs i 210Pbex, određene su&nbsp;prostorne koordinate X i Y i nadmorske visine uzetih uzoraka. Prostorne distribucije&nbsp;koncentracija 137Cs i 210Pbex izražene su izolinijama, a kreiranani su i 3D prikazi u procentima&nbsp;odstupanja 137Cs i 210Pbex od lokalnog padavinskog ulaza 137Cs i 210Pbex. U oba slučaja je&nbsp;primetan dominantan uticaj pravca jugoistok - severozapad (SE-NW), odnosno jugoistočnog&nbsp;vetra &bdquo;Ko&scaron;ava&rdquo;.</p><p>Komparativna analiza ovih metoda je ukazala na validnost u kvantifikaciji procesa eolske&nbsp;erozije i mogućnost njihove primene u budućnosti, a dobijeni rezultati produkcije eolskog&nbsp;nanosa i gubitaka zemlji&scaron;ta su dali doprinos oceni stanja degradacije zemlji&scaron;ta i ugroženosti Deliblatske pe&scaron;čare.&nbsp;</p><p>&nbsp;</p> / <p>Erosion is a very complex physical process which, under the impact of atmospheric forces,&nbsp;creates destructive changes on the soil surface layer. In case the primary agent of particle&nbsp;movement is wind, we talk about wind or aeolian erosion. Aeolian erosion is a specific&nbsp;process which occurs in extremely complex situations of mutual interaction of numerous&nbsp;natural and anthropogenic factors of mainly stochastic properties making its research a&nbsp;complex scientific-research problem.</p><p>The main goal of conducted research was to monitor the process of aeolian erosion at the&nbsp;chosen localities of Deliblato Sands, the largest and the most important area of the kind in&nbsp;Europe. In other words, the goal is to determine the quantities of aeolian sediment &ndash;&nbsp;sediment transport by direct systematic measurements in field conditions, determine&nbsp;dominant direction of sediment transport, define its annual distribution and determine the&nbsp;effect of vegetation on reducing the intensity of aeolian erosion. During a four-year period,&nbsp;for the first time in this area, i.e. at the same locality, a comparative research of aeolian&nbsp;erosion have been conducted using the mechanical sediment trap and activities based on&nbsp;137Cs radioisotope tracing technique for estimating soil losses using special theoretical&nbsp;models.</p><p>The monitoring of aeolian erosion processes was conducted during the period 2006 &ndash; 2009&nbsp;at Cvijićev vis which was chosen as a typical locality since it was, apart from the central&nbsp;position on Deliblato Sands, a cultivated surface without any wind protection. Alongside&nbsp;with these measurements, at Dragićev hat &ndash; nursery garden, the monitoring of aeolian&nbsp;erosion was conducted on two measurement points during the period May 2006 &ndash; April&nbsp;2007 in areas with the protective vegetative covers and/or vegetative belts. Aeolian erosion&nbsp;intensity measurement was performed by static sediment traps of the type &ldquo;deflametre&rdquo;&nbsp;(dimension of entry opening 10 x 10 cm) oriented on certain directions (N, NE, E, SE, S, SW,&nbsp;W, NW) in order to define the aeolian erosion processes in vector terms. The quantification&nbsp;of aeolian sediment using static traps constantly facing certain wind blowing directions enabled to log and differentiate &ldquo;summary&rdquo; yield of aeolian sediment (arithmetic sum of all sediment quantities from all traps) and &ldquo;resulting&rdquo; quantity of aeolian sediment (vector sum), since those quantities determine not only the entire amount of transported aeolian sediment (sediment yield) but also the sediment transported outside the areas of erosion field (soil loss), bur also the general direction of its transport.</p><p>Aeolian erosion measurement on Cvijićev vis for the period 2006 &ndash; 2009 indicated the<br />significant aeolian erosion processes which were defined by medium annual ediment<br />transport of 4.48 kg m-1. It has been determined that the total quantities of moved<br />sediment were 25.94, 20.92, 52.98 and 43.47 kg m-1 year-1, and sediment transport 3.24,&nbsp;2.61, 6.62 and 5.43 kg m-1 year-1 in 2006, 2007, 2008 and 2009 respectively.</p><p>Sediment transport for the period May 2006 &ndash; April 2007 showed that on Cvijićev vis which&nbsp;is characterized by bareness and openness there was the biggest aeolian erosion &ndash; four&nbsp;times bigger compared to Dragićev hat &ndash; nursery garden I characterized by bareness and&nbsp;protectiveness and 30.2 times bigger compared to Dragićev hat &ndash; nursery garden II&nbsp;characterized by overgrown condition and protectiveness of erosive field. At the locality&nbsp;Dragićev hat &ndash; nursery garden I the aeolian erosion was recorded 7.5 times bigger&nbsp;compared to the one recorded on Dragićev hat &ndash; nursery garden II.</p><p>Soil loses expressed through the resulting monthly sediment transport (vector sum)<br />equalled 5.13, 2.04, 4.31 and 11.94 kg m-1 in 2006, 2007, 2008 and 2009 respectively, and&nbsp;the percentage share of annual soil losses compared to total sediment yield (arithmetical&nbsp;sum) varied between 8.1% and 27.5%. The resulting aeolian sediment movement direction&nbsp;was SE-NW under the influence of the dominant southeast wind &ldquo;Koshava&rdquo;.</p><p>Even though the most reliable determination of aeolian erosion and its effects is based on&nbsp;direct systematic measurements in real time conditions in the field, starting from the end&nbsp;of the last century the methods of tracking radionuclide from radioactive precipitation,&nbsp;especially 137Cs, for the purposes of estimating the soil loss and spatial distribution of&nbsp;aeolian sediment, have been used increasingly.</p><p>The total number of samples taken for the method of monitoring the quantity of 137Cs was&nbsp;149, 9 of which were initial samples (3 profiles with 3 samples each), 14 main samples (2&nbsp;profiles with 7 samples each), 32 remaining samples (8 profiles with 4 samples each), 36&nbsp;reference samples (9 profiles with 4 samples each) and 58 samples taken radially on certain&nbsp;directions (N, NE, E, SE, S, SW, W, NW).</p><p>The main sample taken from the uncultivated land with the detected quantity of 137Cs of&nbsp;10,603.57 Bq m-2 represents the comparative value, i.e. local precipitation input of 137Cs for&nbsp;the models of 137Cs quantities conversion into the quantities of lost soil. This sample depicts&nbsp;the area where the research was conducted and represents the comparative value which&nbsp;can properly define the aeolian erosion processes.</p><p>For converting the measured quantities of 137Cs into the quantities of lost soil using Walling&nbsp;model the simplest proportional model (PM) for cultivated land was used and the most&nbsp;appropriate profile distribution model (PDM) for uncultivated soil with the newest version&nbsp;of PC compatible software package in Microsoft Excel Add-Ins. Using the profile distribution model (PDM) on all samples taken from the experimental area the quantities of average soil loss obtained were 207.06 t ha-1 year-1 and 2.10 cm. Basher &amp; Webb model gave the average soil loss of 212.18 t ha-1 year-1 and 2.09 cm. Soil loss calculated using the conversion of 137Cs quantities into the soil loss quantities indicate the significant aeolian processes defined by strong and excessive aeolian erosion.</p><p>For 58 soil samples taken from areas radially distributed on certain directions (N, NE, E, SE,&nbsp;S, SW, W, NW), apart from 137Cs and 210Pbex concentrations and quantities, spatial&nbsp;coordinates X and Y were determined as well as the altitude of taken samples. Spatial&nbsp;distribution of 137Cs and 210Pbex quantities are represented by isolines, and also 3D&nbsp;demonstrations were created showing the percentage of deviation of 137Cs and 210Pbex from&nbsp;the local precipitation input of 137Cs and 210Pbex. In both cases, the dominant direction SENW&nbsp;was noticeable, i.e. the southeast wind &ldquo;Koshava&rdquo;.</p><p>The comparative analysis of these methods indicated the validity in the quantification of&nbsp;aeolian erosion process and the possibility of its application in the future and the obtained&nbsp;results of aeolian sediment yield and soil loss contributed to determining the state of soil&nbsp;degradation and vulnerability of Deliblato Sands.</p>