Dissertations / Theses on the topic 'Rainfall simulator'

Thesis (M.S.)--University of Missouri-Columbia, 2007.
The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file (viewed on October 25, 2007) Vita. Includes bibliographical references.

The thicketization of the semi-arid region of the United States has resulted in a dramatic change allowing invasive woody species to dominate the landscape with an unknown impact to the water budget. This landscape transformation has created a need to study the hydrology of the region and in particular the effects of increased brush on the water cycle. To study the effects of invasive brush on the water budget, a portable abovecanopy rainfall simulator was developed for plot scale hydrologic research. The rainfall simulator was tested at various field locations, including within the Edwards Plateau, to replicate natural rainfall events on typical hillslope-scale plots. The rainfall simulator was used to quantify aspects of the water budget for a 7 m by 14 m research plot on the Edwards Plateau in Texas. Three rainfall simulation dates were selected for detailed hydrologic analysis. Overall, throughfall accounted for 74% of the water applied to the plot, while 26% of applied water was in the form of stemflow. Lateral subsurface flow represented 33% of the water measured leaving the research plot. A notable result of rainfall simulations was extensive lateral subsurface flow and no surface runoff. The rainfall simulator has proven to be a cost-effective and efficient research tool for replicating natural rainfall in arid and semi-arid environments.

This study considered runoff of pesticides from cotton fields using a rainfall simulator. The Australian cotton industry is based on clay soils on low sloping land and uses a hill-furrow surface geometry. These conditions are unlike those in many studies of pesticide dissipation, sorption and runoff and there has been little previous research into hydrology, erosion and pesticide runoff in the industry. Of particular interest was to characterise runoff of different pesticides, clarity the underlying factors controlling pesticide runoff, and investigate management practices to reduce runoff of pesticides with contrasting chemical properties, for the conditions found in the cotton industry. Runoff behaviour of different pesticides has often been studied independently. Separating the inherent behaviour of the pesticide and the conditions of the study, and comparing pesticides, is difficult. Runoff concentrations depend on the hydrology and erosion of each site, and particularly on the timing of runoff events after spraying. These factors can be controlled and/or measured using a rainfall simulator. Furthermore, multi-residue pesticide analysis now allows study of a number of pesticides simultaneously, so that their behaviour can be compared directly for the same site conditions and management options. Review of the literature indicated that a large variation in pesticide runoff is related to application rate, formulation and placement, and dissipation, and that only a shallow soil surface layer contributes pesticides to runoff. In this study, the analysis was simplified by only considering placement on the soil surface (i.e. not on plant foliage) of liquid/emulsified formulations and soilincorporated sprays. A simple conceptual framework was used to compare and integrate data from simulated rainfall studies of pesticides used in the cotton industry and contrast them with data in the literature. By comparing pesticide runoff to the concentration in the soil surface at the start of rain, one set of factors, those that occur before the rainfall event - application and dissipation, were separated from those that occur during the rainfall event - leaching and runoff extraction. To complete the picture of how pesticides get from the spray nozzle to the edge of a field in runoff, four main areas were considered - dissipation, runoff extraction, sediment-water partitioning and management. Dissipation data were collected at four sites across the cotton growing areas and runoff data at three of these sites, on soils with medium to high clay content. Some 14 pesticides were studied, including the insecticide endosulfan (a- and (3-isomers and the breakdown product endosulfan sulfate) at all sites. Dissipation studies concentrated on the 3-6 weeks after application, when concentrations are highest, and on the soil surface layer that contributes pesticides to runoff. For practical reasons, soil concentrations were measured in the 0-25mm soil depth. While the major emphasis was on dissipation of endosulfan from soil and crop residues, several organophosphate (OP) insecticides (chlorpyrifos, dimethoate and profenofos) and a range of herbicides (fluometuron, metolachlor, prometryn, diuron, pendimethalin, pyrithiobac sodium) were also studied. Dissipation in the 0-25mm soil depth followed first-order exponential decay, with one phase, for most pesticides. However, large initial losses occurred for several of the insecticides studied: 35-55% for endosulfan, but only when applied at higher temperatures, most likely due to volatilisation. Initial losses of 50-75% occurred for the OP’s, dimethoate (inconsistently), chlorpyrifos and profenofos. Dissipation half-lives in the 0-25mm soil depth (after initial losses) were 6-20 days for endosulfan (total of a-, (3- and sulfate), 8-13 days for organophosphates, and 13-32 days for herbicides, a-endosulfan consistently dissipated more rapidly than p-endosulfan, but the two isomers were affected differently by site and/or climatic factors. Dissipation of endosulfan was similar for ULV and EC formulations, for bare and covered soil and for band and blanket sprays, but was somewhat slower after two applications. Only small amounts of endosulfan sulfate formed in dry soils, while more formed in temporarily wet soil, contributing about half the total endosulfan remaining after 30 days. Dissipation in surface 0-25mm soil was more rapid than in 0-50mm soil, but this varied from no effect for the rapidly dissipated OP’s to 1.6 times faster for endosulfan, and varied according to Koc for the herbicides. Herbicides with lower Koc dissipated faster in the surface layer than those with higher Koc, due to greater downward movement. Downward movement decreased the apparent halflife in the 0-25mm soil and increased the apparent half-life in 0-50mm soil. Half-lives in 0-25mm soil were considerably lower than published ‘selected’ values. Dissipation of endosulfan was consistent with studies in other warmer climates. The shallow depth of soil studied (which enhanced downward movement) and application on the surface contributed to this more rapid dissipation. The results are consistent with observations that “runoff available residues” dissipate more rapidly than generally expected for bulk soil. Endosulfan dissipated rapidly (e.g. 75-90%) from crop residue cover (wheat or cotton trash) within the first day after spraying, apparently a result of volatilisation. Half-lives for endosulfan on crop residues after the initial loss were similar to those in bare and covered soils. The data indicate that a benefit of retaining crop residues on the soil surface, in addition to reducing runoff and sediment losses, is that it intercepts and dissipates the endosulfan more rapidly than when sprayed onto soil. Runoff extraction was investigated, in a simple empirical analysis, by comparing concentration in soil (mg/kg) before rain and event average concentration in runoff (pg/L), using data from three rainfall simulator studies in cotton fields, for 14 pesticides, and from the literature. The ratio of runoff to soil concentrations, or the linear regression slope fitted through the origin, was termed the runoff extraction ratio (ER0). The pesticides varied widely in solubility (0.003-700,OOOmg/L) and ranged from strongly (DDE, KD~ 15,000) to weakly sorbed (fluometuron, dimethoate, pyrithiobac sodium, KD <30). Runoff extraction behaviour from bare soil was remarkably consistent for pesticides of widely different properties. Total concentrations in runoff of each pesticide were closely related to concentrations in the soil (0-25mm) before rain, generally with a similar relationship for all pesticides and sites, over four orders of magnitude range in concentrations. As a first approximation, concentration in runoff (pg/L) = 28 times concentration in soil (mg/kg), (or Ero = 28). Runoff extraction was also somewhat similar for dissolved N and P, and organic N. Ero values were not related to partition coefficients (KP) measure in runoff. However, runoff extraction did decrease with time after spraying and was lower for aged DDE and trifluralin at one site. This is considered to relate to lower concentrations in the surface few mm of soil (c.f. 0-25mm soil) over time. ERO values were similar for the slopes studied (0.2-4%), for long and short plots, and for banded and blanket spray plots. Runoff extraction was reduced where cover reduces sediment concentration. Runoff extraction was significantly lower for a weakly sorbed pesticide (dimethoate) in only one instance and not for a range of other weakly sorbed pesticides at the other sites. Concentrations in the water and sediment phases in runoff, and in sediment (mg/kg), were also linearly related to soil concentrations for pesticides of similar KP, but extraction in two phases varied according to normal partitioning (Eqn 5-5). The sediment concentration in runoff (10-60 g/L from bare plots) had a secondary effect on ER0, and only affected ERO when sediment concentration was low (i.e. with cover). This is because higher sediment concentrations were associated with lower concentrations in the sediment (mg/kg), due to greater desorption and decreasing physical enrichment. Less physical enrichment (due to size-selective sediment sorting) occurred than observed on coarser textured soil (e.g. enrichment ratio up to 8), with enrichment ratios mostly less than 1.0 (due to desorption) and no greater than 2.0. For all pesticides, the concentration in sediment (mg/kg) was within a factor of about two of the soil concentration adjusted for desorption using the normal partitioning equation. Organic carbon and clay were also only slightly enriched in sediment, despite considerable deposition in the furrows. This is because the soils eroded as aggregates (due to low sand and high clay content), and because coarser sediment had greater concentrations of sorbed pesticides than finer sediment, the opposite of what is normally expected (e.g. where coarser sediment is sand). The notable similarity of runoff extraction ratio for all pesticides in the rainfall simulator studies was probably because (a) the main factor that limit runoff of weakly sorbed chemicals, i.e. leaching from the runoff mixing zone, was ineffective because of low infiltration and ponding of infiltrated water in the shallow tilled layer in the bottom of furrows, (b) sediment concentrations were high enough to ensure transport of strongly sorbed pesticides, and (c) all pesticides had some transport in both the water and sediment phases, diluting the response to sediment load. The concentration of pesticide extracted from soil into runoff appears to be determined by the soil concentration, with, in the absence of significant leaching and with sufficient sediment transport, little differentiation between pesticides of different partition properties. This is partly because, on any plot, the same mass of soil and the same volume of water are involved in mixing, independent of the chemical being considered, and because factors that increase extraction of solutes also tend to increase detachment of sediment. Analysis of published runoff data for a range of pesticides in US croplands indicated similar average runoff extraction to the rainfall simulator studies in Australian cotton fields. However, runoff extraction was higher for much more erosive conditions (e.g. cultivated 10-15% slopes) and lower for low erosion conditions (furrow irrigation on low sloping fields in California). Runoff extraction was similar for this latter case (i.e. ERO~30) once adjusted to a higher sediment concentration. Analysis of the rainfall simulator and published data presents a conceptual framework where the major drivers of pesticide runoff were separated between (a) application rate and dissipation, described by soil concentration at the start of rain, which accounts for five orders of magnitude differences in runoff concentrations, and (b) runoff extraction during the rainfall event, which varied over a limited range. The first of these factors causes most of the difference in runoff between pesticides. Partition coefficients in runoff (KP) were not affected by cover and wheel traffic treatments even though these treatments had large effects on pesticide runoff concentrations. KP values increased with time after spraying, rapidly in the first few days and more slowly over the next few weeks, for all pesticides. KP values were greater than soil sorption KD values, increasingly so for pesticides of lower sorption. Thus pesticides normally considered weakly sorbed were much more sorbed in sediment than expected, particularly at longer times. Conversely, moderately/strongly sorbed pesticides, such as endosulfan, were less sorbed than expected in the first day or so. Partitioning appeared to be influenced by both time of contact with soil and time of mixing (during rain). The results are conceptually consistent with a two-compartment, bi-phasic (fast-slow) sorption model, with the soil in the runoff-mixing layer under rainfall being a continuous dilution system. The ‘slow’ phase, due to diffusion into less accessible soil domains, leads to increasing partition coefficients with greater time of contact. The short time of mixing means that the water phase is mainly interacting with the ‘fast’ or most accessible fraction, while the ‘slow’ fraction remains in the sediment phase. Percentages in the water phase in runoff, for 14 pesticides, roughly followed a published relationship with solubility, and an empirical relationship with soil sorption Koc values, but only for erosive conditions. These relationships do not reflect the full range in responses that occur due to the likely range of concentrations and organic carbon content of sediment, or the increase in KP with time. Because of lower KP values soon after spraying, less soluble pesticides had 20-45% in water. Conversely, a few days/weeks after spraying, more soluble pesticides had only 60-80% in water. Thus all pesticides tended to have a ‘foot in each camp’ and some potential for management using erosion control practices. It is an over simplification to expect ‘percent in water’ to be a characteristic of a pesticide. So long as sediment concentration and KP can be estimated, the percentage in the water phase can be calculated quite simply from first principles (Eqn 5-5) and behaviour for relevant field conditions can be assessed. This equation was used to show that reported values of percentage in the water phase for endosulfan that appeared to conflict (20-95%) and the values from the rainfall simulator plots (15- 45%) are explained by differences in sediment concentration and organic carbon in the studies. A wide range of percentages in water (10-95%) will occur for pesticides with KP of 5-500 (or solubilities -1-100), such as endosulfan, for the range of sediment concentrations and organic carbon that might occur in the environment. Improved practices are needed to minimise soil erosion, and related agrochemical transport, from cotton fields during rain. The most influential practice used in other agricultural industries, that is, retaining crop residues as surface cover, is rarely practiced in the Australian cotton industry. Therefore two options available to cotton growers, namely retention of surface cover and controlling wheel traffic, were evaluated using simulated rain on a well-aggregated black Vertosol. Increasing cover (0-60%) resulted in decreasing runoff, soil loss and sediment concentration. Runoff and soil loss were reduced by an order of magnitude with about 50% cover and by a small amount with notraffic. Cover and no traffic combined gave least runoff and soil loss. Pesticide transport in runoff was also reduced strongly by retaining on-ground cover and somewhat reduced by avoiding prior wheel traffic. With 45-60% cover, concentrations were reduced 5-fold for a-, |3- and total endosulfan; halved for endosulfan sulfate, trifluralin and DDE, and unchanged for prometryn. Cover had more effect on endosulfan because cover intercepted and dissipated the sprayed endosulfan, reducing concentrations in surface soil. Cover greatly reduced total pesticide losses (g/ha) because cover reduced runoff and soil loss considerably. With 45-60% cover, total losses were reduced by 90-98%. No-traffic gave 40% lower losses, and enhanced the effect of cover, but did not prevent large pesticide losses from bare plots. Cover provided more control of more soil-sorbed pesticides (endosulfan, trifluralin and DDE). Control of the less sorbed prometryn was largely due to cover reducing runoff. An examination of the practical requirements for maintaining effective cover in cotton farming systems indicated that most of the perceived conflicts with insect, weed and irrigation management could be overcome, although further study is needed. Many of these results have only been possible because of the use of the rainfall simulator, multiresidue pesticide analysis and the availability of sufficient resources. Such opportunities are rare in field research. By allowing an intensive regime of runoff sampling at controlled times after pesticide applications, the study has yielded data with more significance, enabling the conclusions made above regarding the relative behaviour of individual pesticides and their extraction from soil in runoff. The author acknowledges the contributions made by others to this study, but all of the experimental work and the data reported in this thesis were under his control.

Made available in DSpace on 2015-03-26T12:31:10Z (GMT). No. of bitstreams: 1 texto completo.pdf: 3461017 bytes, checksum: c343053cbc2e3541876491e2e17050a4 (MD5) Previous issue date: 2009-08-10
Conselho Nacional de Desenvolvimento Científico e Tecnológico
Rainfall is considered the main climatic factor related to the degradation of agricultural lands, and among their characteristics the intensity stands out as the main conditioning factor of the erosive process. The present work had the objective of evaluating the influence of different precipitation profiles in the soil and water losses. Four precipitation profiles were considered: exponential negative, forwarded double exponential, delayed double exponential and a profile with constant intensity, being the total applied depth of 55 mm and duration of 30 min. The simulator was installed in an experimental area with Inceptsol. The treatments consisted in the simulation of the four precipitation profiles in conditions of bare soil, applied three times in 24 h-intervals, characterizing three initial conditions of soil moisture: low, medium and high. Thus the treatments were defined based on the combination of the precipitation profiles and the conditions of initial soil moisture, using six replications. It was used a completely randomized design, and the water and soil losses data were submitted to the analysis of variance (ANOVA) and the Tukey's test (5%). During the first application, which consisted in the simulation of rains in soils with low initial moisture there was not superficial runoff, consequently there was not soil loss. In the second application, where the soils were with intermediate initial moisture, differences among the treatments were not obtained due to the high variability observed. In the third application, statistically significant differences among the treatments were found for soils with conditions of high moisture. The largest losses of water and soil were obtained with the negative exponential profile, followed by the delayed, forwarded and constant profiles. Therefore, it was possible to conclude that there was influence of the precipitation profiles on the soil and water losses, however only for the treatments with soils of high moisture.
A precipitação é considerada o principal fator climático relacionado à degradação de terras agrícolas e, entre suas características, a intensidade se destaca como o principal fator condicionador do processo erosivo. O presente trabalho teve por objetivo avaliar a influência de diferentes perfis de precipitação nas perdas de solo e água. Foram definidos quatro perfis de precipitação: exponencial negativo, duplo exponencial adiantado, duplo exponencial atrasado e um perfil com intensidade constante, que foram aplicados por meio de um simulador de chuvas, sendo uma lâmina total de 55 mm e duração de 30 min comum a todos os perfis. O simulador foi instalado em uma área experimental cujo solo foi classificado como Cambissolo Háplico. Os tratamentos consistiram na simulação dos quatro perfis de precipitação em condições de solo descoberto, aplicados por três vezes em intervalos de 24 h, caracterizando três condições iniciais de umidade do solo: baixa, intermediária e alta. Assim, os tratamentos foram definidos pela combinação dos perfis de precipitação e das condições de umidade inicial do solo, sendo realizadas para cada um deles seis repetições. O delineamento experimental utilizado foi o inteiramente casualizado (DIC) e os totais de perdas de água e solo submetidos à análise de variância (ANOVA) e ao teste de Tukey (5%). Durante a primeira aplicação, que consistiu na simulação de chuvas em solos com baixa umidade inicial, não houve escoamento superficial e, consequentemente, não ocorreram perdas de solo. Na segunda aplicação, condição em que os solos se encontravam com umidade inicial intermediária, não foram obtidas diferenças estatisticamente significativas entre os tratamentos devido à alta variabilidade observada. Na terceira aplicação foram encontradas diferenças estatisticamente significativas entre os tratamentos, para solos com condições de umidade próxima à saturação. As maiores perdas de água e de solo foram obtidas para o perfil exponencial negativo, seguido pelos perfis duplo exponencial atrasado, adiantado e o constante. Assim, foi possível concluir que houve influência dos perfis de precipitação nas perdas de solo e água, porém apenas para os tratamentos com umidade do solo próxima a saturação.

Networks of small, often mobile, polarimetric radars are gaining popularity in the hydrometeorology community due to their rainfall observing capabilities and relative low purchase cost. In recent years, a number of installations have become operational around the globe. The problem of signal attenuation by intervening rainfall has been recognized as the major source of error in rainfall estimation by short-wavelength (C-, X, K-band) radars. The simultaneous observation of precipitation by multiple radars creates new prospects for better and more robust attenuation correction algorithms and, consequently, yields more accurate rainfall estimation. The University of Iowa hydrometeorology group's acquisition of a network of four mobile, polarimetric, X-band radars has resulted in the need for a thoughtful evaluation of the instrument. In this work, we use computer simulations and the data collected by The University of Iowa Polarimetric Radar Network to study the performance of attenuation correction methods in single-radar and network-based arrangements. To support the computer simulations, we developed a comprehensive polarimetric radar network simulator, which replicates the essential aspects of the radar network rainfall observing process. The simulations are based on a series of physics- and stochastic-based simulated rainfall events occurring over the area of interest. The characteristics of the simulated radars are those of The University of Iowa Polarimetric Radar Network. We assess the correction methods by analyzing the errors in reflectivity and rainfall rate over the area of interest covered by the network's radars. To enable the implementation of the attenuation correction methods to the data collected by The University of Iowa Polarimetric Radar Network, we first developed a set of utilities to assist with efficient data collection and analysis. Next, we conducted a series of calibration tests to evaluate the relative calibration and channel balance of the 2 network's radars. Finally, in an attempt to verify the results obtained via computer simulations, we applied the set of attenuation correction algorithms to the data collected by The University of Iowa Polarimetric Radar Network.

The increase of animal agriculture coupled with excess manure production, and the reduced availability of land has led to the over application of animal manure on agricultural fields. The excessive application of manure is responsible for nutrient and bacterial pollution of downstream waterbodies. Manure application based on the crop phosphorus (P) requirements has been recommended as a viable method to reduce nutrient pollution. A plot scale study was conducted to measure the loss of nutrients and bacterial transport in runoff from cropland treated with poultry litter, dairy manure and inorganic fertilizer according to the P requirements of the crop. Three simulated rainfall events were conducted 1, 2 and 35 days after planting of corn. Highest P and N concentrations were observed in the runoff from plots treated with poultry litter, followed by dairy manure and inorganic fertilizer. The poultry litter treated plots exhibited highest concentrations of bioavailable P in the runoff, compared to all other treatments. The P from poultry litter treated plots was also mostly in the soluble form, which underscores the need to control the runoff from cropland in order to decrease the P losses from the poultry litter treated fields. The edge of the field nutrient concentrations observed in this study were high enough to cause severe to moderate eutrophication problems in downstream waterbodies unless they are diluted. In general, nutrient concentrations were lower during the second simulated event, compared with those from the first event. A significant reduction in the nutrient concentrations of runoff was observed from the second to the third simulated event for all the treatments. This reduction was attributed to the loss of nutrients by natural rainfall-runoff events during the time period between the second and the third simulated rainfall event, plant uptake of nutrients, sorption and leaching processes. The indicator bacteria analyzed in the present study were fecal Coliform (FC), Escherichia Coli (E.Coli) and Enterococcus (ENT). The bacterial concentrations reported in the runoff for the first and second simulated events were 104 to 105 times higher than the federal and state limits for primary contact recreation waters. No significant effect of treatments was observed on the bacterial concentrations in runoff. The highest concentrations were observed for FC, followed by ENT and EC in the runoff. The ratio of bacteria removed in runoff to the bacteria applied also followed the above trend. The concentrations of bacteria generally increased from the first to second simulated event; unlike the nutrients. However, the bacterial concentrations dropped significantly from second to the third simulated rainfall event to the levels lower than those designated for primary contact recreation water limits. This reduction was attributed to the washing away of bacteria by the heavy rainfall-runoff events in the period between second and third simulated rainfall events and the die-off of bacteria. The results reported from this study suggest that the manure application based on crop P requirements can also be a significant source of nutrient pollution and should be coupled with other best management practices (BMPs) also to reduce nutrient pollution. The results also suggest that the manure treated cropland can be a source for significant indicator bacterial pollution and appropriate BMPs are required to mitigate their effect.
Master of Science

A canavicultura é de extrema importância para o Brasil e para o mundo. O aumento da produção de cana-de-açúcar reflete em maior uso de herbicidas, como o diuron e a hexazinona. A adoção do sistema de cana crua, em que a palha permanece sobre o solo após colheita mecanizada, implica em alteração na dinâmica ambiental destes herbicidas. Portanto, o objetivo deste trabalho foi avaliar as perdas de diuron e hexazinona por escoamento superficial em sistemas de cana crua. O experimento constou de sistema fatorial 3x2x2 (12 tratamentos) com delineamento em blocos casualizados e 4 repetições, totalizando 48 parcelas. Os fatores foram: 3 níveis de palha (0, 50% e 100% da dose = 14 t ha-1); 2 níveis de umidade de solo (10 e 18% v/v) e 2 momentos de chuva (0 e 3 dias após aplicação dos produtos). Sobre as parcelas de 1 m2, foi utilizado um simulador de chuvas ajustado para evento com intensidade de 80 mm h-1 durante uma hora e meia, resultando num volume total precipitado de 120 mm. Um produto comercial contendo diuron e hexazinona foi aplicado na dose de 3 kg ha-1 e calda de 700 L ha-1, conforme recomendações do fabricante. As quantidades de água e sedimentos escoadas foram registradas e as concentrações dos herbicidas analisadas em solução por UPLC. Os resultados foram submetidos à ANOVA e, em caso de interação, as médias foram comparadas pelo teste de Tukey (p<0,05). A quantidade de diuron e hexazinona nos sedimentos foi estimada fazendo uso dos dados de sorção da literatura. A presença de palha sobre o solo reduziu as perdas de água, sedimentos e diuron, mas não teve efeitos sobre as perdas de hexazinona. Em outras palavras, a cobertura morta não reduz as perdas de moléculas muito solúveis, como a hexazinona. As maiores perdas de diuron e hexazinona foram observadas em solução, mesmo no tratamento controle (sem palha), uma vez que a palha reduz a massa de sedimentos desprendida. No entanto, não houve diferença nas perdas entre os dois níveis de palha (50 e 100%), sugerindo que 7 t ha-1 são suficientes para atenuar as perdas de água, sedimentos e diuron. Maiores teores de água no solo (18 versus 10% v/v) implicaram em maiores perdas dos herbicidas por escoamento superficial. Já o momento da chuva não afetou as perdas dos herbicidas, indicando que 3 dias não foram suficientes para a dissipação ou maior sorção dessas moléculas.
Sugarcane is a major crop in Brazil and of great importance to the world. Higher yields implicate in higher use of pesticides, such as diuron and hexazinone. The adoption of green cane system, in which the straw is kept in the soil surface after mechanical harvesting, has changed the environmental behavior of theses herbicides. Therefore, the goal of this research was to evaluate runoff losses of diuron and hexazinone in green cane systems. The 3x2x2 (12 treatments) factorial experiment was performed in a randomized block with 4 replicates. The factors were i) 3 levels of sugarcane straw (0, 50% and 100%, based on a dose of 14 t ha-1); ii) 2 levels of initial soil moisture (10 and 18% VWC), and iii) 2 rainfall periods (0 and 3 dafter herbicides application). A rainfall simulator was adjusted to simulate an 80 mm h-1 rainfall event for one and a half hour (120 mm) over plots of 1 m2. A commercial product containing diuron and hexazinone was used at rate of 3 kg ha-1 dissolved in 700 L ha-1, according to label recommendations. The amounts of water and sediments were registered and herbicides concentrations analyzed by UPLC. Herbicides attached to the sediments were estimated according to sorption data from the literature. The results were evaluated by ANOVA and means compared by Tukey test (p<0.05). Sugarcane straw decreased water, sediments, and diuron losses by runoff, but did not affect hexazinone losses. In other words, crop residues cannot prevent losses of highly soluble molecules, such as hexazinone. Greater herbicides losses were observed in the aqueous phase, even for the control treatment (without straw), since straw reduces the amounts of detached sediments. However, no difference was observed between the two levels of straw (50 and 100%), meaning that 7 t ha-1 is sufficient for mitigating water, sediments, and diuron losses by runoff. Higher soil moisture (18 versus 10%) resulted in higher herbicides runoff. Yet, rainfall period did not affect herbicide losses, indicating that 3 days were not long enough for enhancing these herbicides dissipation or sorption.

Submitted by Jorge Silva (jorgelmsilva@ufrrj.br) on 2017-05-05T18:33:54Z No. of bitstreams: 1 2010 - Frederico Santos Machado.pdf: 2164167 bytes, checksum: 28047d32a367720ee18ef9cf4c58be45 (MD5)
Made available in DSpace on 2017-05-05T18:33:54Z (GMT). No. of bitstreams: 1 2010 - Frederico Santos Machado.pdf: 2164167 bytes, checksum: 28047d32a367720ee18ef9cf4c58be45 (MD5) Previous issue date: 2010-05-27
The hydric erosion is one of main environmental impact in Central Amazonia region, causing serious economic consequences for implantation and recovery of operational locations of PETROBRAS S/A oil and natural gas exploration and production sites. This study was carried through in Petroliferous Province of Urucu, municipality of Coari (AM). The objective was to quantify soil loss (SL) and runoff (SD) in different soil classes and under vegetal coverings using a portable rain simulator. Sixteen batteries of tests were done, eight batteries for each soil class (Fluventic Dystrustepts e Kanhaplic Haplustults). For each battery three repetitions had been made in the following covers: forest, forest without litter and grass. The results of Tukey test at 0.05 level showed that soil classes were not significantly different between themselves. On the other hand, the organic carbon percentage and the soil bulk density had greatest importance for the soil loss. These attributes, plus fine sand and clay, influenced most in the runoff. It was also possible to observe that runoff was the attribute of highest correlation with soil loss. The conversion of forest areas into grass increased, at least, four times the water runoff, changing it from 14.2 mm to 57.7 mm in the grass area, and, at least, two times loss of soil mass, achieving the value of 35.68 t.ha-1.year-1 in the grass area. Although with these values, grassy areas only differed significantly from the forest areas in the runoff. The runoff values tend to increase twofold when litter is removed from the forest soils, while the soil losses increased 40%. The attributes that were important for the soil classes were the same for cover types, however with differences in the attribute clay content for soil loss, and coarse sand for runoff. For the studied treatments, the attribute that showed highest correlation with soil loss was runoff.
A eros?o h?drica ? um dos principais impactos ambientais na regi?o da Amaz?nia Central, causando s?rias conseq??ncias, inclusive econ?micas, para a implanta??o e recupera??o de loca??es operacionais de explora??o e produ??o de petr?leo e g?s natural da PETROBRAS S/A. Este estudo foi realizado na Prov?ncia Petrol?fera de Urucu, munic?pio de Coari (AM) e teve como objetivo quantificar a perda de solo (PS) e o escoamento superficial (ES) em diferentes grupos de solo e coberturas vegetais, utilizando um simulador de chuva port?til. Foram efetuadas 16 baterias de testes, sendo 4 baterias para cada grupo de solo (ARGISSOLO AMARELO Al?tico, ARGISSOLO VERMELHO-AMARELO Al?tico, CAMBISSOLO H?PLICO Al?tico e CAMBISSOLO H?PLICO Tb Distr?fico). Em cada bateria foram realizados tr?s repeti??es nas seguintes coberturas: floresta, floresta sem serrapilheira e gram?nea. As classes de solo n?o diferiram significativamente entre si ao n?vel de 5%, segundo o teste de Tukey. Para a PS, a porcentagem de carbono org?nico e a densidade do solo t?m grande import?ncia para os grupos de solo. Esses atributos, mais a areia fina e argila influenciaram o ES. Tamb?m foi poss?vel observar que o escoamento foi o atributo com maior coeficiente de correla??o com a perda de solo. A convers?o das ?reas de floresta em gram?nea aumentou em no m?nimo quatro vezes a l?mina de ?gua escoada, passando de 14,2 mm para 57,7 mm na gram?nea e, no m?nimo, duas vezes a massa de solo perdida, chegando a 35,8 t.ha-1.ano-1 na gram?nea. Apesar desses valores, as ?reas cobertas por gram?neas apenas diferiram significativamente das ?reas de florestas em rela??o ao ES. Quando se retira a serrapilheira em solos florestais, os valores de escoamento dobram, enquanto que as perdas aumentam em 40%. Os atributos que tiveram import?ncia para as classes de solo foram os mesmos para os tipos de coberturas, por?m com diferen?as no atributo teor de argila para a perda de solo e no teor de areia grossa para o escoamento. Para os tratamentos estudados, o escoamento continua sendo o atributo com maior coeficiente de correla??o com a perda.

Accelerated soil erosion is one of the major threats to agricultural production in Ethiopia and the Harerge region is not exceptional. It is estimated that about 1.5 billion tones of soil is being eroded every year in Ethiopia. In the extreme cases, especially for the highlands, the rate of soil loss is estimated to reach up to 300 t ha-1yr-1 with an average of about 70 t ha -1yr-1 which is beyond any tolerable level. The government have made different attempts to avert the situation since 1975 through initiation of a massive program of soil conservation and rehabilitation of severely degraded lands. Despite considerable efforts, the achievements were far bellow expectations. This study was aimed at assessing the effect of some soil properties, rainfall intensity and slope gradients on surface sealing, soil erodibility, runoff and soil loss from selected sites in the Harerge region, eastern Ethiopia, using simulated rainfall. Soil loss was also estimated for the sites using Soil Loss Estimation Model for Southern Africa (SLEMSA) and the Universal soil Loss Equation (USLE). Moreover, the effectiveness of various rates and patterns of wheat residue mulching in controlling soil loss was also evaluated for one of the study sites, (i.e. Regosol of Alemaya University), under both rainfall simulation and field natural rainfall conditions. For most of the erosion parameters, the interaction among soil texture, slope gradient and rainfall intensity was significant. In general however, high rainfall intensity induced high runoff, sediment yield and splash. The effect of slope gradients on most of the erosion parameters was not significant as the slope length was too small to bring about a concentrated flow. The effect of soils dominated by any one of the three soil separates on the erosion parameters was largely dependent on rainfall intensity and slope gradient. The soils form the 15 different sites in Harerge showed different degrees of vulnerability to surface sealing, runoff and sediment yield. These differences were associated with various soil properties. Correlation of soil properties to the erosion parameters revealed that aggregate stability was the main factor that determined the susceptibility of soils to sealing, runoff and soil loss. This was in turn affected by organic carbon content, percent clay and exchangeable sodium percentage (ESP). Soils with relatively high ESP such as those at Babile (13.85) and Gelemso (7.18) were among the lowest in their aggregate stability (percent water stable aggregates of 0.25 –2.0mm diameter); and have highest runoff and sediment yield as compared to other soils in the study. Similarly, most of those soils with relatively low ESP, high organic carbon content (OC%) and high water stable aggregates such as Hamaressa, AU (Alemaya University) vertisol and AU regosol were among the least susceptible to sealing and interrill erosion. Nevertheless, some exceptions include soils like those of Hirna where high runoff was recorded whilst having relatively high OC%, low ESP and high water stable aggregates. Both the SLEMSA and USLE models were able to identify the erosion hazards for the study sites. Despite the differences in the procedures of the two models, significant correlation (r = 0.87) was observed between the values estimated by the two methods. Both models estimated higher soil loss for Gelemso, Babile, Karamara and Hamaressa. Soil loss was lower for Diredawa, AU-vertisol and AU-Alluvial all of which occur on a relatively low slope gradients. The high soil loss for Babile and Gelemso conforms with the relative soil erodibility values obtained under rainfall simulation suggesting that soil erodibility, among others, is the main factor contributing to high soil loss for these soils. The difference in the estimated soil losses for the different sites was a function of the interaction of the various factors involved. Though the laboratory soil erodibility values were low to medium for Hamaressa and Karamara, the estimated soil loss was higher owing to the field topographic situations such as high slope gradient. SLEMSA and USLE showed different degrees of sensitivities to their input variables for the conditions of the study sites. SLEMSA was highly sensitive to changes in rainfall kinetic energy (E) and soil erodibility (F) and less sensitive to the cover and slope length factors. The sensitivity of SLEMSA to changes in the cover factor was higher for areas having initially smaller percentage rainfall interception values. On the other hand, USLE was highly sensitive to slope gradient and less so to slope length as compared to the other input factors. The study on the various rates and application patterns of wheat residue on runoff and soil loss both in the laboratory rainfall simulation and under field natural rainfall conditions revealed that surface application of crop residue is more effective in reducing soil loss and runoff than incorporating the same amount of the residue into the soil. Likewise, for a particular residue application method, runoff and soil loss decreased with increasing application rate of the mulch. However, the difference was not significant between 4 Mg ha-1 and 8 Mg ha-1 wheat straw rates suggesting that the former can effectively control soil loss and can be used in areas where there is limitation of crop residues provided that other conditions are similar to that of the study site (AU Regosols). The effectiveness of lower rates of straw (i.e. less than 4 Mg ha-1 ) should also be studied. It should however be noted that the effectiveness of mulching in controlling soils loss and runoff could be different under various slope gradients, rainfall characteristics and cover types that were not covered in this study. Integrated soil and water conservation research is required to develop a comprehensive database for modelling various soil erosion parameters. Further research is therefore required on the effect of soil properties (with special emphasis to aggregate stability, clay mineralogy, exchangeable cations, soil texture and organic matter), types and rates of crop residues, cropping and tillage systems, mechanical and biological soil conservation measures on soil erosion and its conservation for a better estimation of the actual soil loss in the study sites. Copyright 2004, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. Please cite as follows: Bobe, BW 2004, Evaluation of soil erosion in the Harerge region of Ethiopia using soil loss models, rainfall simulation and field trials, PhD thesis, University of Pretoria, Pretoria, viewed yymmdd < http://upetd.up.ac.za/thesis/available/etd-08022004-141533 / >
Thesis (PhD (Soil Science))--University of Pretoria, 2004.
Plant Production and Soil Science
unrestricted

This study measured sediment, nutrient and pesticide edge-of-field and deep percolation losses from conventional tillage (CT), strip tillage (ST), and no-till (NT) treatments in a burley tobacco production system. The field experiment results show that the CT treatment yielded more total runoff, 93.6 mm, than ST and NT. Compared to the CT treatment, ST reduced the total mass loss of NO₃⁻ by 37 percent, NH₃ by 54 percent, PO₄³⁻ by 12 percent, TP by 65 percent and TSS by 64 percent. The NT treatment reduced the total mass of NO₃⁻ by 49 percent, NH₃ by 46 percent, PO₄³⁻ by 17 percent, TP by 73 percent and TSS by 77 percent. Field experiment results showed that, respectively, the ST and NT treatments yielded 77 and 82 percent less chlorpyrifos, an insecticide, mass loss in runoff compared to CT. With respect to flumetralin loss, a growth inhibitor, the NT treatment yielded 30 percent less and ST yielded 6.8 percent more flumetralin mass, compared to CT. The percolated water results show that compared to the CT treatment, ST resulted in 33 percent less NH₃, 7.5 percent less TKN, and 39 percent less PO₄³⁻ total mass. The NT treatment yielded 41 percent less NH₃ and 20 percent less TKN total mass loss compared to the CT treatment. The NT treatment had the most NO₃⁻, PO₄³⁻, and TP mass loss below the root zone, however NT also had the most percolated water by 53 mm (838 L). The ST treatment yielded 77 percent less, and NT yielded 82 percent less chlorpyrifos mass loss compared to the CT treatment. Compared to CT, the NT treatment resulted in 30 percent less and ST yielded 7 percent more flumetralin mass in percolated water. The GLEAMS model was used to simulate runoff, nutrient, sediment and pesticide losses from the same three tillage practices evaluated in the field experiment. The model results showed that for runoff volume, TSS, nitrogen, chlorpyrifos and flumetralin in runoff, the conventional tillage practice generated greater losses than the conservation tillage practices. Compared to the field experiment results, GLEAMS under predicted nitrogen and phosphorus in percolated water. The conservation tillage practices simulated in GLEAMS were effective in minimizing the loss of agricultural pollutants.
Master of Science

Submitted by Jorge Silva (jorgelmsilva@ufrrj.br) on 2017-04-25T20:29:14Z No. of bitstreams: 1 2012 - Eliete Nazar? Eduardo.pdf: 947644 bytes, checksum: d85c318ce9c161ce157d42e89ed6c4af (MD5)
Made available in DSpace on 2017-04-25T20:29:14Z (GMT). No. of bitstreams: 1 2012 - Eliete Nazar? Eduardo.pdf: 947644 bytes, checksum: d85c318ce9c161ce157d42e89ed6c4af (MD5) Previous issue date: 2012-02-27
FAPERJ
The purpose of this study was to determine the soil erodibility (K) and soil cover and management factor (C) under natural and simulated rainfall in an Udult soil in Serop?dica, Rio de Janeiro State, Brazil. In the study with natural rainfall, the soil losses (SL) were evaluated from measurements in Wischmeier standard-plots in the period of 2006 to 2011, with their respective erosivity index (EI30). The soil erodibility values were calculated from the ratio of soil loss and the annual rainfall erosivity (Ka) and total period (Kt), considering the angular coefficient generated from the linear regression analysis between those parameters (Kci and Kct). For the determination of C factor under natural rainfall, it was evaluated the treatments maize in contour lines (MN) and maize in slope lines (MMA) during three years of cultivation. In the study with simulated rainfall, soil losses were evaluated during the period of September to December 2011, under the same treatments (MN and MMA). The soils erodibility under natural rainfall for 5 years of study is 0.0090 Mg ha h ha-1 MJ-1 mm-1. For the simulated rain, this value is 0.00021 Mg ha h ha-1 MJ-1 mm-1; under natural rainfall, the average values of C factor calculated are 0.0070 and 0.0132 Mg ha Mg-1 ha-1, respectively, for maize grown in contour and slope line, and 0.0932 and 0.1298 Mg ha Mg-1 ha-1 under simulated rainfall. In both studies, the C factor showed higher values in the beginning of the study period, decreasing with the evolution of culture. The higher C factor values obtained in the study with simulated rainfall are associated with higher erosivity index resulting from the characteristics adopted in the use of simulator rainfall.
O objetivo desse trabalho foi determinar a erodibilidade do solo (K) e o fator manejo e cobertura do solo (C), sob chuva natural e simulada, em um ARGISSOLO VERMELHO-AMARELO em Serop?dica (RJ), informa??es necess?rias para o planejamento do uso da terra e conserva??o do solo e da ?gua. No estudo com chuva natural foram avaliadas as perdas de solo (PS) ocorridas em parcelas padr?o de Wischmeier, no per?odo de 2006 a 2011, com seus respectivos ?ndices de erosividades (EI30). As erodibilidades do solo foram calculadas pelo quociente entre a perda de solo e a erosividade das chuvas anuais (Ka) e total no per?odo (Kt), considerando o coeficiente angular gerado na an?lise de regress?o linear entre esses mesmos par?metros (Kci e Kct). Para a determina??o do fator C, sob chuva natural foram avaliados os tratamentos de milho em n?vel (MN) e de milho morro a baixo (MMA) em tr?s anos de cultivo. No estudo com chuva simulada, as perdas de solo foram avaliadas no per?odo de setembro a dezembro de 2011, sob os mesmos tratamentos (MN e MMA). A erodibilidade do solo estudado sob chuva natural, para 5 anos de estudo ? de 0,0090 Mg ha h ha-1 MJ-1 mm-1. Para a chuva simulada, esse valor ? de 0,00021 Mg ha h ha-1 MJ-1 mm-1; sob chuva natural, os valores m?dios do fator C calculados s?o de 0,0070 e 0,0132 Mg ha Mg-1 ha-1, respectivamente, para a cultura do milho cultivado em n?vel e morro abaixo, e de 0,0932 e 0,1298 Mg ha Mg-1 ha-1 sob chuva simulada. Em ambos os estudos, o fator C apresentou maiores valores no in?cio do per?odo avaliado, decrescendo com a evolu??o da cultura. Os elevados valores do fator C obtidos no estudo com chuva simulada est?o associados ao elevado ?ndice de erosividade resultante das caracter?sticas adotadas no uso do simulador de chuva.

This dissertation is focused to modeling of the rainfall in laboratory conditions by the rainfall simulator invented in VUT Brno. Results of the measuring are compared with measured values of the original rainfall simulator located in VVU VSH VUT in Brno by the Kninicky village in 1976-1980. Dissertation verifies rightness of substituting the natural rain with the rainfall simulator on bare soil in the original location, kinetic energy of the rain drops is substituted with the energy gained from the weight of the drops - the flow. Generally the dissertation determines the possibilities of the rainfall simulator's current location.

Two experimental rainfall simulation plots in the Edwards Aquifer region of Texas were established to measure the effects of brush clearing on surface and subsurface water movement pathways. Multi-stage rainfall simulations were carried out at a site with Juniperus ashei (ashe juniper) cover both before and after brush removal, with three replications of a particular rainfall event for each vegetation condition. Similar simulations were carried out on a plot with a longstanding grass cover. Both plots included trenches at their downhill ends for observation of shallow lateral subsurface flow. Canopy interception was found to represent a major water loss, with interception of 32.7 mm for an average 166 mm, 5.25 hr rainfall event. Brush clearing had little impact on surface runoff, with no overland flow occurring at the juniper plot for either vegetation condition, while 31.9 percent of applied rainfall moved as overland flow at the grass plot. This difference was attributed to differences in the structure and permeability of the epikarst. Brush removal caused significant (90 percent confidence level) reduction in shallow lateral subsurface flow into the trench after brush removal, with 56.7 percent of water reaching the surface entering the trench for the pre-cut condition and only 43.4 percent for the post-cut condition. However, subsurface water movement through other pathways increased from 31.0 to 54.1 percent after brush removal. This additional water, due to removal of canopy interception, could either move off-site through conduit and fracture flow or remain on site as storage in conduits, unconsolidated caliche/marl layers, or in soil pockets. Two tracer tests with fluorescent dyes were also conducted using simulated rainfall to assess discrete flow paths discharging into the trench at the downhill end of the juniper plot. Analysis of samples from sixteen outlet locations revealed that not all areas of the plot were connected hydraulically to the trench. Additionally, subsurface flow paths were found to have a high degree of interconnection, linking conduit flow outlets with multiple inlet locations on the plot surface. Conduits showed strong connection with an area surrounding juniper vegetation, with rapid water flow (up to 2.4 m/h) from this area.

In this study, large scale rainfall simulation was used to evaluate runoff generation from canopy and intercanopy areas within an ashe juniper woodland of the Edwards Plateau. One 3 x 12 m site was established beneath the canopy of mature ashe juniper trees and two sites were established in intercanopy areas. At the base of each plot a trench was constructed for capturing and monitoring shallow subsurface flow. Rainfall simulations on the juniper site produced little surface runoff even though rainfall intensity exceeded 145mm/hour on some occasions. A total of 82.6% of the water applied to the juniper dominated site was accounted for as shallow subsurface flow. The dynamic nature of shallow subsurface flow indicate this process is driven chiefly by macropore flow. On the intercanopy site, 12.67% of the water left the site as surface runoff and ≤3% left as shallow subsurface flow. Large root channels and conduits, which were not present on the intercanopy site, within the soil may promote shallow subsurface flow beneath the juniper canopy. This study is the first to document and suggest shallow subsurface flow occurs on Texas rangelands. The results of this experiment indicate shallow subsurface flow is an important mode of runoff generation on the Edwards Plateau.

Simulated annealing (Kirkpatrick et al, 1983) is used to estimate the parameters of a mathematical model that predicts the water yield from a catchment. The calibration problem involves finding the global minimum of a multivariate function that has many extraneous local minima, a situation in which conventional optimisation methods are ineffective. The objective function which quantifies discrepancies between the computed and observed streamflows must be carefully selected to satisfy the least square assumptions. Several published simulated annealing algorithms have been implemented, tested and evaluated using standard test functions. Appropriate cooling schedules are found for each algorithm and test function investigated. The number of function evaluations required to find the minimum is compared to published results for the test functions using either simulated annealing and other global optimisation methods. A new simulated annealing algorithm based on the Hooke and Jeeves (1961) pattern search method is developed and compared with existing algorithms from the literature.

High rates of runoff and soil loss occur in temperate areas in late winter and early spring. A low-intensity rainfall simulator and a snowmelt simulator were developed to study soil erosion by snowmelt and/or spring rainfall on a sandy loam soil of southern Quebec. Erosive agent, land slope, thaw depth of partly frozen soil, and intensity and duration of the event were studied in the laboratory using the simulators. All these factors had significant effects on soil erosion. Sediment concentration, soil loss rate, and total soil loss increased with increasing land slope, thaw depth, and intensity, duration, and impact energy of the erosive agent. Duration showed a significant interaction with each of the other factors. interaction between thaw depth and intensity of the event was significant under both snowmelt and rainfall simulation, whereas interaction between land slope and intensity was significant under rainfall simulation, but not under snowmelt simulation. Erosive agent also interacted with each of the other factors. All these interactions increased soil loss. The presence of a frozen sublayer interacted with the other factors and greatly increased runoff, sediment concentration and total soil loss. Rainfall caused more soil loss than did snowmelt under the same conditions.

Rainfall is a natural process, which has a high degree of variability in both space and time. Information on the spatial and temporal variability of rainfall plays an important role in the process of surface runoff generation. Hence it is important for a variety of applications in hydrology and water resources management. The spatial variability of rainfall can be substantial even for very small catchments and an important factor in the reliability of rainfall-runoff simulations. Catchments in urban areas usually are small, and the management problems often require the numerical simulation of catchment processes and hence the need to consider the spatial and temporal variability of rainfall. A need exists, therefore, to analyse the sensitivity of rainfall-runoff behaviour of catchment modelling systems (CMS) to imperfect knowledge of rainfall input, in order to judge whether or not they are reliable and robust, especially if they are to be used for operational purposes. Development of a methodology for identification of storm events according to the degree of heterogeneity in space and time and thence development of a detailed spatial and temporal rainfall model within a hydroinformatic environment utilising real-time data has been the focus of this project. The improvement in runoff prediction accuracy and hence the importance of the rainfall input model in runoff prediction is then demonstrated through the application of a CMS for differing variability of real storm events to catchments with differing orders of scale. The study identified both spatial and temporal semi-variograms, which were produced by plotting the semi-variance of gauge records in space and time against distance and time respectively. These semi-variograms were utilised in introducing estimators to measure the degree of heterogeneity of each individual storm events in their space and time scale. Also, the proposed estimators use ground based gauge records of the real storm events and do not rely on delicate meteorological interpretations. As the results of the investigation on the developed semi-variogram approach, real storm events were categorised as being High Spatial-High Temporal (HS-HT); High Spatial-Low Temporal; (HS-LT); Low Spatial-High Temporal (LS-HT); and Low Spatial-Low Temporal variability.A comparatively detailed rainfall distribution model in space and time was developed within the Geographical Information Systems (GIS). The enhanced rainfall representation in both space and time scale is made feasible in the study by the aid of the powerful spatial analytic capability of GIS. The basis of this rainfall model is an extension of the rainfall model developed by Luk and Ball (1998) through a temporal discretisation of the storm event. From this model, improved estimates of the spatially distributed with smaller time steps hyetographs suited for especially the urban catchments could be obtained. The importance of the detailed space-time rainfall model in improving the robustness of runoff prediction of CMS was investigated by comparing error parameters for predictions from CMS using alternate rainfall models, for various degrees of spatiotemporal heterogeneity events. Also it is appropriate to investigate whether the degree of this improvement to be dependent on the variability of the storm event which is assessed by the adopted semi-variogram approach. From the investigations made, it was found that the spline surface rainfall model, which considered the spatial and temporal variability of the rainfall in greater detail than the Thiessen rainfall model resulted in predicted hydrographs that more closely duplicated the recorded hydrograph for the same parameter set. The degree of this improvement in the predicted hydrograph was found to be dependent on the spatial and temporal variability of the storm event as measured by the proposed semi-variogram approach for assessing this feature of a storm event. The analysis is based on forty real events recorded from the Centennial Park Catchment (1.3km2) and the Upper Parramatta River Catchment (110km2) in Sydney, Australia. These two case study catchments were selected to ensure that catchment scale effects were incorporated in the conclusions developed during the study.

This thesis presents the development and application of a downscaling framework for multi site simulation of daily rainfall. The rainfall simulation is achieved in two stages. First, rainfall occurrences at multiple sites are downscaled, which is followed by the generation of daily rainfall amounts at each site identified as wet. A continuous weather state based nonparametric downscaling model conditional on atmospheric predictors and a previous day average rainfall state is developed for simulation of multi site rainfall occurrences. A nonparametric kernel density approach is used for simulation of rainfall amounts at individual sites conditional on atmospheric variables and the previous day rainfall amount. The proposed model maintains spatial correlation of rainfall occurrences by simulating concurrently at all stations and of amounts by using random innovations that are spatially correlated yet serially independent. Temporal dependence is reproduced in the occurrence series by conditioning on previous day average wetness fraction and assuming the weather states to be Markovian, and in the amount series by conditioning on the previous day rainfall amount. The seasonal transition is maintained by simulating rainfall on a day-to-day basis using a moving window formulation. The developed downscaling framework is calibrated using the relevant atmospheric variables and rainfall records of 30 stations around Sydney, Australia. Results indicate a better representation of the spatio-temporal structure of the observed rainfall as compared to existing alternatives. Subsequently, the framework is applied to predict plausible changes in rainfall in warmer conditions using the same set of atmospheric variables for future climate obtained as a General Circulation Model simulation. While the case studies presented are restricted to a specific region, the downscaling model is designed to be useful in any generic catchment modelling and management activity and/or for investigating possible changes that might be experienced by hydrological, agricultural and ecological systems in future climates.

The effect of drought and simulated rainfall on wheat (Triticum aestivum L.) seed and grain quality during development and maturation, including delayed harvest, were investigated with pot-grown plants harvested serially in a ventilated poly tunnel. Grain quality was assessed by seed mean dry weight (MOW), moisture content, HFN (Hagberg falling number), 50S (sodium dodecyl sulphate sedimentation), nitrogen content, sulphur content a nd seed size; seed quality by ability to germinate and subsequent air-dry seed storage survival. Developmental duration was the dominant factor influencing quality: HFN increased progressively and substantially from early seed filling to beyond harvest maturity in a sigmoidal pattern; 50S in a linear trend in 2012, but plateaued in 2013 from the end of seed filling; seed longevity in a curvi-linear trend (quantified by a polynomial) till harvest maturity; ability to germinate was maximal (100% normal germination) from 15-17 days before to 30-32 days after mass maturity (control). Stopping irrigation before seed filling ended reduced MOW and grain size, but increased HFN, 50S, germinability and seed longevity. Drought resulted in more rapid increase in HFN (14 days after anthesis > 21 DAA > 28 DAA > Control) and earlier attainment of maximum seed longevity. Ear wetting once close to harvest maturity reduced HFN considerably; thereafter it increased, though less than the control. Ear wetting reduced but root wetting increased HFN early in development, whereas both treatments at harvest maturity reduced seed and grain quality. Ear wetting reduced subsequent seed longevity immediately, but a period of re-drying in planta improved longevity - surpassing the control. Ear wetting (25-50mm) once at harvest maturity resulted in greater subsequent seed longevity than wetting for several days or >50mm. Hence direct and indirect and immediate and delayed effects of simulated rainfall were detected; drought improved and rainfall reduced HFN, but seed quality was increased by both.

This research presents an integration of a stochastic space-time rainfall model and distributed hydrologic simulation with GIS. The integrated simulation system consists of three subsystems: a stochastic space-time rainfall model, a geographical information system (GIS), and a distributed physically-based hydrologic model. The developed stochastic space-time rainfall model is capable of estimating the storm movement and simulating a random rainfall field over a study area, based on the measurement from three raingauges. An optimization-based lag-k correlation method was developed to estimate the storm movement, and a stochastic model was developed to simulate the rainfall field. A GIS tool, ARC/INFO, was integrated into this simulation system. GIS has been applied to automatically extract the spatially distributed parameters for hydrologic modeling. Digital elevation modeling techniques were used to process a high resolution digital map. A distributed physically-based hydrologic model, operated in HEC-1, simulated the stochastic, distributed, interrelated hydrological processes. The Green-Ampt equation is used for modeling the infiltration process, kinematic wave approximation for infiltration-excess overland flow, and the diffusion wave model for the unsteady channel flow. Two small nested experimental watersheds in southern Arizona were chosen as the study area where three raingauges are located. Using five recorded storm events, a series of simulations were performed under a variety of conditions. The simulation results show the model performs very well, by comparing the simulated runoff peak flow and runoff depth with the measured ones, and evaluated by the model efficiency. Both model structure and model parameter uncertainties were investigated in the sensitivity analysis. The statistical tests for the simulation results show that it is important to model stochastic rainfall with storm movement, which caused a significant change in runoff peak flow and runoff depth from that where the input is only one gage data. The sensitivity of runoff to roughness factor N and hydraulic conductivity Ks were intensively investigated. The research demonstrated this integrated system presents an improved simulation environment for the distributed hydrology.

Landslides are a natural hazard that cause impactful effects not only on the earth’s natural environment (i.e. morphology, flora, and fauna) but also on people and infrastructures. Direct and indirect damage to property and human settlements, as well as many casualties all over the world are unfortunately consequences of landslide events. Hence, the prediction and modeling of such events is of high interest to scholars aiming at understanding both the triggering mechanism and the post-failure dynamics, either by empirical or physically-based approaches. The present thesis addresses this key problem by identifying innovative approaches to one of the most key aspects of landslide hazard assessments, namely the spatio-temporal prediction of landslide occurrence. In particular, the main goals of the work can be summarized as follows: i) the improvement of the spatial prediction of landslides through physically-based models, with a special focus on the real advantages and disadvantages of 1D vs. 3D slope stability analysis at the catchment scale; ii) the analysis of the post-failure dynamics as well as the assessment of the influence of the geotechnical and rheological parameters influencing the front celerity and the impact force of the sliding mass; iii) the improvement of performances of empirical rainfall thresholds through the integration of the reanalysis soil moisture information within a hydro-meteorological framework.

More than a billion dollars are spent annually on stream restoration in the United States (Bernhardt et al., 2005), but the science remains immature. A promising technique for estimating a single or range of design discharges is the generalization of a parsimonious conceptual continuous simulation model. In this study the Probability Distributed Model (PDM), was generalized for the Maryland and Virginia Piedmont. Two hundred and sixty years of daily average flow data from fifteen watersheds were used to calibrate PDM. Because the application of the study is to stream restoration, the model was calibrated to discharges greater than two times baseflow and less than flows with a return period of ten years. The hydrologic calibration parameters were related to watershed characteristics through regression analysis, and these equations were used to calculate regional model parameters based on watershed characteristics for a single "ungaged" independent evaluation watershed in the region. Simulated flow was compared to observed flow; the model simulated discharges of lower return periods moderately well (e.g., within 13% of observed for a flow with a five year return period). These results indicate this technique may be useful for stream restoration and wetland design.
Master of Science

This study investigates the February 17th, 2006 massive rockslide that occured in the island of Leyte, Philippines following heavy rainfall and four minor earthquakes. The rockslide is considered one of the largest and most catastrophic slides in the last few decades as it completely inundated the village of Guinsaugon, taking the lives of approximately 1,400 of the 1,800 residents of the village.

The distinct element simulation of the rockslide is performed using 3DEC (Three-Dimensional Distinct Element Code) in order to investigage the underlying triggering mechanism of the slide as well as the behavior of the debris flow. The 3DEC models were established based on field observations from the U.S. Reconnaissance team and material and joint properties based on in-situ and laboratory test results. The possible triggering mechanisms considered in the distinct element analyses were the rainfall-induced hydraulic pressurization of the fault forming part of the main scarp, as well as the seismic acceleration due to the minor earthquakes that occured the morning of the slide.

The results of the analyses and simulations indicate that the rainfall-induced hydraulic pressurization of the fault was potentially the main triger for the initiation of the slide. The minor earthquakes, which occured before and around the time of the slide initiation, appeared to have very little effect on the triggering mechanism and the debris flow are comparable to witness accounts and field observations. The results presented in this study are expected to provide better understanding of rockslides such as the one that occured in the Philippines on February 17, 2006. With further improvements in computational capabilities in the future, distinct element simulations can have the potential to reliably predict the initiation and behavior of slides, and help mitigate their impact.
Master of Science

The effect of natural and simulated rainfall during seed development and maturation on wheat (Triticum aestivum L.) seed quality development, particularly changes in subsequent air-dry longevity (stored hermetically at 40°C with 15% moisture content), was investigated in poly-tunnel (2010) and field experiments (2011 and 2012). Avoiding rain during seed development and maturation accelerated the rate and reduced the duration of seed filling and thus reduced final seed weight, while increase in rainfall slowed down the rate and increased the duration of seed filling and increased seed weight in wet and cold weather (2011), but dry and warm weather (2012) had the opposite effect. Maximum germination of seeds dried after harvest was attained at 39-49 days after anthesis (DAA) and maintained throughout until 70 DAA. Seeds not dried attained ability to germinate slightly later with subsequent fluctuations, probably associated with dormancy. Potential longevity increased during this period, reaching a maximum at 53-56 DAA, coinciding with harvest maturity, and then declining in 2012 but not in 2011. Wetting ears reduced longevity at all stages of seed development when seeds were harvested soon after the event, and more so after two wettings, but considerable recovery in subsequent longevity occurred when seeds were harvested after 1 or 7 days re-drying after wetting. Reduction in rainfall significantly reduced crude protein and the incidence of blackpoint in both years. Hagberg Falling Number (HFN) decreased in 2011, but increased in 2012, perhaps due to wetter and cooler conditions in 2012. Rainfall affected seed quality development but the extent of damage varied amongst developmental stages, being greater at later developmental stages although considerable ability for recovery with re-drying was noted. Rainfall had a negative effect on most of grain quality parameters except crude protein concentration.

Computer modeling is the powerful tool for simulating nature’s behavior; however, still more efforts are need for reaching perfect simulation with computer models (especially in the hydrological field of study). In this Master’s thesis, the accuracy of the HEC-HMS computer model for long term rainfall-runoff simulation was evaluated across Sweden. Five different catchments from north to south of Sweden were selected and then simulation have done for 34 years of available data. Simulation was conducted using daily, monthly and yearly time scale resolutions. Results from the north to the south of Sweden were completely different. Simulated runoff and observed runoff in northern catchments followed the same pattern over different time scales but in the southern part of Sweden the results had different patterns in space and time. The best results with HEC-HMS were found in the northern catchments with steep main river slopes. In the southern catchments the model could not predict runoff in any realistic manner at any time and space scale. In total the HEC-HMS model cannot simulate the rainfall runoff for long periods of simulation across Sweden. This is especially true in southern parts of the country dominate with low elevation catchments. However, with regards to its ability for event-based simulation HEC-HMS could be a suitable tool to simulate flood event discharges that are needed for road or other hydraulic structures designs. But, this would require significant amounts of calibration and model development.

Simulating the processes of rainfall and runoff are at the core of hydrologic modeling. Geomorphologic features, rainfall variability, soil types, and water table depths strongly influence hydrological process in Florida ecosystems. Topographic characteristics of the terrain define the stream paths and landscape. Alteration of these characteristics as a result of urban and/or agricultural developments, for example, can highly influence wetlands and river basin response. There are two predominant landforms in Florida: wetlands, where Variable Saturated Areas form near streams causing saturation excess runoff, and uplands where runoff is mainly generated by infiltration excess. The objective of this work is to analyze the impacts of geomorphologic and hydrologic characteristics on runoff mechanisms in humid environments such as Florida. In general, most research at the hillslope scale use hypothetical values of rainfall, sometimes non-realistic values, and single slope forms to explain the geomorphic and hydrologic process on Variable Saturated Areas. In this thesis, the complexity of hillslope processes on actual Florida topography is assessed by coupling a Digital Elevation Model with a two-dimensional variable saturated-unsaturated flow model called HYDRUS-2D. Actual rainfall records and soil parameters from the Characterization Data for Selected Florida Soils, Soil Survey were used to evaluate hydrologic impacts. A commercial software package, River Tools was used to display and extract topographic information from the Digital Elevation Models. Results show that when inflitration excess runoff is dominant, infiltration and runoff are very sensitive to time resolution, especially for convective storms. When saturation excess occurs, runoff is not affected by rainfall intensity. However, saturated hydraulic conductivity, depth to the water table, slope and curvature highly influence the extent of Variable Saturated Areas. Results indicate runoff in shallow water table environments is produced mainly by subsurface storm runoff, running below the surface, except in hillslopes with concave curvature and mild slopes. Additionally, concave hillslopes generate more saturation excess runoff than straight and convex hillslopes.

Runoff from agricultural fields amended with animal manure or fertilizer is a source of phosphorus (P) pollution to surface waters, which can have harmful effects such as eutrophication. The objectives of this study were to evaluate the impact of soil P status and the P composition of manure sources on P in runoff, characterize the effects of manure sources on mass loss of dissolved reactive P (DRP), total dissolved P (TDP), algal available P (AAP) and total P (TP) in runoff, and enhance the PLAT database with respect to soluble P attenuating factor (SPAF) and non-soluble P attenuating factor (NSPAF) values. Soil boxes set at 5% slopes received 7.5 cm hr-1 of simulated rainfall. Study soils included a Kenansville loamy sand (loamy siliceous subactive thermic Arenic Hapludults, a Coastal Plain soil) and a Davidson silt loam (kaolinitic thermic Rhodic Kandiudults, a Piedmont soil). Soil test P concentrations ranged from 16 to 283 mg P kg-1. Sources of P included broiler litter (BRL), breeder manure (BRD), breeder manure treated with three rates of alum (Al2(SO4)3) BRD0-0 kg m-2, BRDL-3.9 kg m-2, and BRDH-7.8 kg m-2 and DAP along with an unamended control. All manure sources were applied at 66 kg P ha-1. Water extractable P (WEP) represented an average of 10 ?b 6% total P in manure. Runoff samples were taken over a 30-min period. Piedmont soil contained greater amounts of clay, Al and Fe concentrations, and higher P sorption capacities that produced significantly lower DRP, TDP, AAP, and TP losses than the Coastal Plain soil. Runoff P loss did not differ for low and high STP soils of same taxonomy with the exception of AAP mass losses for Coastal Plain soil samples. Water extractable P in manures accounted for all DRP lost in runoff with DRP correlating strongly with WEP concentration (0.9961). A weak relationship between DRP in runoff and WEP applied to soil boxes was observed (R2=0.6547) and increased when a possible outlying manure treatment, BRL, was omitted from regression data (0.9927). Overall, manures containing the highest WEP concentrations supplied the largest losses of DRP in runoff. Manure treated with 3.9 and 7.8 kg m-2 of Al2(SO4)3 (alum) decreased DRP in runoff by 29%. Values calculated for PLAT SPAF and NSPAF coefficients were higher for Coastal Plain soil than Piedmont soil and overall higher than default values in PLAT. Management based on these results should help minimize harmful effects of P in runoff.

Thesis (Ph. D. in the Field of Hydrology)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, February 2013.
"February 2013." Cataloged from PDF version of thesis.
Includes bibliographical references (p. 290-307).
The Maritime Continent experiences strong moist convection, which produces significant rainfall and drives large fluxes of heat and moisture to the upper troposphere. Despite the importance of these processes to global circulations, current predictions of climate change over this region are still highly uncertain, largely due to inadequate representation of the diurnally-varying processes related to convection. In this work, a coupled numerical model of the land-atmosphere system (RegCM3-IBIS) is used to investigate how more physically-realistic representations of these processes can be incorporated into large-scale climate models. In particular, this work improves simulations of convective-radiative feedbacks and the role of cumulus clouds in mediating the diurnal cycle of rainfall. Three key contributions are made to the development of RegCM3-IBIS. Two pieces of work relate directly to the formation and dissipation of convective clouds: a new representation of convective cloud cover, and a new parameterization of convective rainfall production. These formulations only contain parameters that can be directly quantified from observational data, are independent of model user choices such as domain size or resolution, and explicitly account for subgrid variability in cloud water content and nonlinearities in rainfall production. The third key piece of work introduces a new method for representation of cloud formation within the boundary layer. A comprehensive evaluation of the improved model was undertaken using a range of satellite-derived and ground-based datasets, including a new dataset from Singapore's Changi airport that documents diurnal variation of the local boundary layer height. The performance of RegCM3-IBIS with the new formulations is greatly improved across all evaluation metrics, including cloud cover, cloud liquid water, radiative fluxes and rainfall, indicating consistent improvement in physical realism throughout the simulation. This work demonstrates that: (1) moist convection strongly influences the near surface environment by mediating the incoming solar radiation and net radiation at the surface; (2) dissipation of convective cloud via rainfall plays an equally important role in the convective-radiative feedback as the formation of that cloud; and (3) over parts of the Maritime Continent, rainfall is a product of diurnally-varying convective processes that operate at small spatial scales, on the order of 1 km.
by Rebecca L. Gianotti.
Ph.D.in the Field of Hydrology

Measurements of runoff quantity and quality from three cattle stocking treatments applied to pastureland in southwestern Virginia indicate the need for further research to determine treatment effects. Three cattle stocking treatments (1) Continuous, 2) Rotational, and 3) Mob) were applied to three pastures at the Virginia Tech Prices Fork Research Farm. Rainfall simulations were performed over replicated plots in each treatment to induce runoff for collection of runoff quantity and quality data during the 2012 grazing season. Additionally, rainfall simulations were performed prior to applying the grazing treatments to establish initial conditions. Monitored runoff quantity and quality response variables included runoff depth, mean nutrient concentrations, and nutrient mass loss. Response variables were compared among the three pastures for initial conditions and among treatments for post-treatment conditions. Additionally, the trends in response variables within the 2012 season were compared among treatments. Plot and rainfall conditions that were expected to influence responses were also collected and analyzed in relation to response variables. Analyses of the response variables suggested that the variability within treatments likely muted any treatment effect on the response variables. Therefore, we concluded that further research is needed to determine treatment effects on runoff quantity and quality.
Master of Science

In 1910, Paris experienced its biggest flood in the 20th century. In 2010, for the anniversary of this event – supposed to happen every 100 years ! – the flood prediction model that is now used on the Seine basin was tested on its simulation,… and failed to reproduce the observed flood volume. This paper will try to explain, and correct, such disappointing results. Many hypotheses have been tested and based on their results, it has been decided to develop a frost module in order to assess the influence of this phenomenon – that is not taken into account by the lumped hydrological model that is used – on the flood formation. A soil temperature model using air temperature as input data was also designed because soil temperature data were not available in 1910. The addition of the frost module did not, however, bring many improvements to the 1910 flood simulation because frost is a too rare phenomenon on the Seine basin for the module to be correctly calibrated. However, new perspectives are presented to continue the research on this phenomenon.

Multi-Polynomial Higher Order Neural Network Group Models (MPHONNG) program developed by the author will be studied in this thesis. The thesis also investigates the use of MPHONNG for financial data and rainfall data simulation and prediction. The MPHONNG is combined with characteristics of Polynomial function, Trigonometric polynomial function and Sigmoid polynomial function. The models are constructed with three layers Multi-Polynomial Higher Order Neural Network and the weights of the models are derived directly from the coefficents of the Polynomial form, Trignometric polynomial form and Sigmoid polynomial form. To the best of the authors knowledge, it is the first attempt to use MPHONNG for financial data and rainfall data simulation and prediction. Results proved satisfactory, and confirmed that MPHONNG is capable of handling high frequency, high order nonlinear and discontinuous data.
Master of Science (Hons)

Historic methods for time series predictions on ungauged sites in the UK have tended to focus on the regionalisation and regression of model parameters against catchment characteristics. Owing to wide variations in catchment characteristics and the (often) poor identification of model parameters, this has resulted in highly uncertain predictions on the ungauged site. However, only very few studies have sought to assess uncertainties in the predicted hydrograph. Methods from the UK Flood Estimation Handbook, that are normally applied for an event design hydrograph, are adopted to choose a pooling group of hydrologically similar gauged catchments to an ungauged application site on the River Tyne. Model simulations are derived for each pooling group catchment with a BETA rainfall-runoff model structure conditioned for the catchment. The BETA rainfall-runoff model simulations are developed using a Monte Carlo approach. For the estimation of uncertainty a modification of the GLUE methodology is applied. Gauging station errors are used to develop limits of acceptability for selecting behavioural model simulations and the final uncertainty limits are obtained with a set of performance thresholds. Prediction limits are derived from a set of calibration and validation simulations for each catchment. Methods are investigated for the carry over of data from the pooled group of models to the ungauged site to develop a weighted model set prediction with pooled prediction limits. Further development of this methodology may offer some interesting approaches for cross-validation of models and further improvements in uncertainty estimation in hydrological regionalisation.

Multi-Polynomial Higher Order Neural Network Group Models (MPHONNG) program developed by the author will be studied in this thesis. The thesis also investigates the use of MPHONNG for financial data and rainfall data simulation and prediction. The MPHONNG is combined with characteristics of Polynomial function, Trigonometric polynomial function and Sigmoid polynomial function. The models are constructed with three layers Multi-Polynomial Higher Order Neural Network and the weights of the models are derived directly from the coefficents of the Polynomial form, Trignometric polynomial form and Sigmoid polynomial form. To the best of the authors knowledge, it is the first attempt to use MPHONNG for financial data and rainfall data simulation and prediction. Results proved satisfactory, and confirmed that MPHONNG is capable of handling high frequency, high order nonlinear and discontinuous data.

Thesis (M.Sc. (Hons.))--University of Western Sydney, 2001.
"A thesis presented to the Department of Computing and Information Systems, University of Western Sydney, Macarthur in partial fulfilment of the requirements for the degree of Master of Science (Honours)" Bibliography : leaves 129-139.

In the last 50 years, flooding has figured as the most frequent and widespread natural disaster globally. Extreme precipitation events stemming from climate change could alter the hydro-geological regime resulting in increased flood risk. Near real-time precipitation monitoring at local scale is essential for flood risk mitigation in urban and suburban areas, due to their high vulnerability. Presently, most of the rainfall data is obtained from ground‐based measurements or remote sensing that provide limited information in terms of temporal or spatial resolution. Other problems may be due to the high costs. Furthermore, rain gauges are unevenly spread and usually placed away from urban centers. In this context, a big potential is represented by the use of innovative techniques to develop low-cost monitoring systems. Despite the diversity of purposes, methods and epistemological fields, the literature on the visual effects of the rain supports the idea of camera-based rain sensors but tends to be device-specific. The present thesis aims to investigate the use of easily available photographing devices as rain detectors-gauges to develop a dense network of low-cost rainfall sensors to support the traditional methods with an expeditious solution embeddable into smart devices. As opposed to existing works, the study focuses on maximizing the number of image sources (like smartphones, general-purpose surveillance cameras, dashboard cameras, webcams, digital cameras, etc.). This encompasses cases where it is not possible to adjust the camera parameters or obtain shots in timelines or videos. Using a Deep Learning approach, the rainfall characterization can be achieved through the analysis of the perceptual aspects that determine whether and how a photograph represents a rainy condition. The first scenario of interest for the supervised learning was a binary classification; the binary output (presence or absence of rain) allows the detection of the presence of precipitation: the cameras act as rain detectors. Similarly, the second scenario of interest was a multi-class classification; the multi-class output described a range of quasi-instantaneous rainfall intensity: the cameras act as rain estimators. Using Transfer Learning with Convolutional Neural Networks, the developed models were compiled, trained, validated, and tested. The preparation of the classifiers included the preparation of a suitable dataset encompassing unconstrained verisimilar settings: open data, several data owned by National Research Institute for Earth Science and Disaster Prevention - NIED (dashboard cameras in Japan coupled with high precision multi-parameter radar data), and experimental activities conducted in the NIED Large Scale Rainfall Simulator. The outcomes were applied to a real-world scenario, with the experimentation through a pre-existent surveillance camera using 5G connectivity provided by Telecom Italia S.p.A. in the city of Matera (Italy). Analysis unfolded on several levels providing an overview of generic issues relating to the urban flood risk paradigm and specific territorial questions inherent with the case study. These include the context aspects, the important role of rainfall from driving the millennial urban evolution to determining present criticality, and components of a Web prototype for flood risk communication at local scale. The results and the model deployment raise the possibility that low‐cost technologies and local capacities can help to retrieve rainfall information for flood early warning systems based on the identification of a significant meteorological state. The binary model reached accuracy and F1 score values of 85.28% and 0.86 for the test, and 83.35% and 0.82 for the deployment. The multi-class model reached test average accuracy and macro-averaged F1 score values of 77.71% and 0.73 for the 6-way classifier, and 78.05% and 0.81 for the 5-class. The best performances were obtained in heavy rainfall and no-rain conditions, whereas the mispredictions are related to less severe precipitation. The proposed method has limited operational requirements, can be easily and quickly implemented in real use cases, exploiting pre-existent devices with a parsimonious use of economic and computational resources. The classification can be performed on single photographs taken in disparate conditions by commonly used acquisition devices, i.e. by static or moving cameras without adjusted parameters. This approach is especially useful in urban areas where measurement methods such as rain gauges encounter installation difficulties or operational limitations or in contexts where there is no availability of remote sensing data. The system does not suit scenes that are also misleading for human visual perception. The approximations inherent in the output are acknowledged. Additional data may be gathered to address gaps that are apparent and improve the accuracy of the precipitation intensity prediction. Future research might explore the integration with further experiments and crowdsourced data, to promote communication, participation, and dialogue among stakeholders and to increase public awareness, emergency response, and civic engagement through the smart community idea.
Negli ultimi 50 anni, le alluvioni si sono confermate come il disastro naturale più frequente e diffuso a livello globale. Tra gli impatti degli eventi meteorologici estremi, conseguenti ai cambiamenti climatici, rientrano le alterazioni del regime idrogeologico con conseguente incremento del rischio alluvionale. Il monitoraggio delle precipitazioni in tempo quasi reale su scala locale è essenziale per la mitigazione del rischio di alluvione in ambito urbano e periurbano, aree connotate da un'elevata vulnerabilità. Attualmente, la maggior parte dei dati sulle precipitazioni è ottenuta da misurazioni a terra o telerilevamento che forniscono informazioni limitate in termini di risoluzione temporale o spaziale. Ulteriori problemi possono derivare dagli elevati costi. Inoltre i pluviometri sono distribuiti in modo non uniforme e spesso posizionati piuttosto lontano dai centri urbani, comportando criticità e discontinuità nel monitoraggio. In questo contesto, un grande potenziale è rappresentato dall'utilizzo di tecniche innovative per sviluppare sistemi inediti di monitoraggio a basso costo. Nonostante la diversità di scopi, metodi e campi epistemologici, la letteratura sugli effetti visivi della pioggia supporta l'idea di sensori di pioggia basati su telecamera, ma tende ad essere specifica per dispositivo scelto. La presente tesi punta a indagare l'uso di dispositivi fotografici facilmente reperibili come rilevatori-misuratori di pioggia, per sviluppare una fitta rete di sensori a basso costo a supporto dei metodi tradizionali con una soluzione rapida incorporabile in dispositivi intelligenti. A differenza dei lavori esistenti, lo studio si concentra sulla massimizzazione del numero di fonti di immagini (smartphone, telecamere di sorveglianza generiche, telecamere da cruscotto, webcam, telecamere digitali, ecc.). Ciò comprende casi in cui non sia possibile regolare i parametri fotografici o ottenere scatti in timeline o video. Utilizzando un approccio di Deep Learning, la caratterizzazione delle precipitazioni può essere ottenuta attraverso l'analisi degli aspetti percettivi che determinano se e come una fotografia rappresenti una condizione di pioggia. Il primo scenario di interesse per l'apprendimento supervisionato è una classificazione binaria; l'output binario (presenza o assenza di pioggia) consente la rilevazione della presenza di precipitazione: gli apparecchi fotografici fungono da rivelatori di pioggia. Analogamente, il secondo scenario di interesse è una classificazione multi-classe; l'output multi-classe descrive un intervallo di intensità delle precipitazioni quasi istantanee: le fotocamere fungono da misuratori di pioggia. Utilizzando tecniche di Transfer Learning con reti neurali convoluzionali, i modelli sviluppati sono stati compilati, addestrati, convalidati e testati. La preparazione dei classificatori ha incluso la preparazione di un set di dati adeguato con impostazioni verosimili e non vincolate: dati aperti, diversi dati di proprietà del National Research Institute for Earth Science and Disaster Prevention - NIED (telecamere dashboard in Giappone accoppiate con dati radar multiparametrici ad alta precisione) e attività sperimentali condotte nel simulatore di pioggia su larga scala del NIED. I risultati sono stati applicati a uno scenario reale, con la sperimentazione attraverso una telecamera di sorveglianza preesistente che utilizza la connettività 5G fornita da Telecom Italia S.p.A. nella città di Matera (Italia). L'analisi si è svolta su più livelli, fornendo una panoramica sulle questioni relative al paradigma del rischio di alluvione in ambito urbano e questioni territoriali specifiche inerenti al caso di studio. Queste ultime includono diversi aspetti del contesto, l'importante ruolo delle piogge dal guidare l'evoluzione millenaria della morfologia urbana alla determinazione delle criticità attuali, oltre ad alcune componenti di un prototipo Web per la comunicazione del rischio alluvionale su scala locale. I risultati ottenuti e l'implementazione del modello corroborano la possibilità che le tecnologie a basso costo e le capacità locali possano aiutare a caratterizzare la forzante pluviometrica a supporto dei sistemi di allerta precoce basati sull'identificazione di uno stato meteorologico significativo. Il modello binario ha raggiunto un'accuratezza e un F1-score di 85,28% e 0,86 per il set di test e di 83,35% e 0,82 per l'implementazione nel caso di studio. Il modello multi-classe ha raggiunto un'accuratezza media e F1-score medio (macro-average) di 77,71% e 0,73 per il classificatore a 6 vie e 78,05% e 0,81 per quello a 5 classi. Le prestazioni migliori sono state ottenute nelle classi relative a forti precipitazioni e assenza di pioggia, mentre le previsioni errate sono legate a precipitazioni meno estreme. Il metodo proposto richiede requisiti operativi limitati, può essere implementato facilmente e rapidamente in casi d'uso reali, sfruttando dispositivi preesistenti con un uso parsimonioso di risorse economiche e computazionali. La classificazione può essere eseguita su singole fotografie scattate in condizioni disparate da dispositivi di acquisizione di uso comune, ovvero da telecamere statiche o in movimento senza regolazione dei parametri. Questo approccio potrebbe essere particolarmente utile nelle aree urbane in cui i metodi di misurazione come i pluviometri incontrano difficoltà di installazione o limitazioni operative o in contesti in cui non sono disponibili dati di telerilevamento o radar. Il sistema non si adatta a scene che sono fuorvianti anche per la percezione visiva umana. I limiti attuali risiedono nelle approssimazioni intrinseche negli output. Per colmare le lacune evidenti e migliorare l'accuratezza della previsione dell'intensità di precipitazione, sarebbe possibile un'ulteriore raccolta di dati. Sviluppi futuri potrebbero riguardare l'integrazione con ulteriori esperimenti in campo e dati da crowdsourcing, per promuovere comunicazione, partecipazione e dialogo aumentando la resilienza attraverso consapevolezza pubblica e impegno civico in una concezione di comunità smart.