O presente estudo objetivou a análise, em lavouras sob semeadura direta, da distribuição vertical e horizontal da resistência à penetração medida por valores georreferênciados do Índice de Cone (IC) e a influência do teor de água do solo, permitindo assim a interpolação para análise de relações causa e efeito. Fatores importantes como o teor de argila, silte, densidade, espaço poroso e teor de matéria orgânica do solo também foram monitorados. Utilizou-se um penetrômetro hidráulico eletrônico, receptores de GPS, amostradores de solo e programas computacionais para o tratamento e obtenção dos resultados em três áreas de estudo sob semeadura direta. Os resultados foram apresentados na forma de mapas em diferentes profundidades, os quais permitem a visualização da espacialidade dos atributos, e pela regressão entre os dados interpolados através da função que apresentou o coeficiente de determinação (R2) mais elevado. Constatou-se a ocorrência de valores maiores para o IC nas profundidades compreendidas de 0,1 a 0,2 m em todas as áreas estudadas. Observou-se também valores menores de IC em locais com maior teor de matéria orgânica, a qual coincidiu com maiores teores de argila e silte em uma das áreas para a qual estes três atributos foram monitorados. A visualização das relações entre fatores nos mapas e a regressão adequada entre os valores interpolados dos atributos destes nem sempre foram possíveis, mas quando houve esta possibilidade, a inter-relação entre o IC e o teor de água, assim como todas as relações espaciais com os demais atributos do solo estudados foram visíveis nos mapas e constatáveis pelo modelo ajustado. O IC foi influenciado principalmente pelo teor de água, densidade e espaço poroso do solo. No entanto seu valor também variou em decorrência de outras inter-relações existentes, as quais estão relacionadas aos teores de argila, silte e matéria orgânica. / The objective of this study is to measure the vertical and horizontal distribution of cone index (CI) values in no- till systems and the influence of soil water content. Data were interpolated allowing for cause and effect analysis. Other important factors as clay and silte content, bulk density, porous space and organic matter content had been monitored. A hydraulically penetrometer was used together with GPS receiver, soil auger and computer programs to collect and analize the data in tree different areas. Results allowed for the generation of maps of the attributes on different depts, and the regression among the interpolated data, choosing the models with the higher determination coefficient (R2). It was evidenced, in all areas, high CI values at depth betwen 0,1 to 0,2 m. Lower values of CI occurred, in one of the areas, in the same place of higher organic matter, in the same located of higher clay and silt content. The visualization of relationship between two factors in the maps and the regression among the interpolated values extracted from those maps was not always possible, but when that possibility took place, the relationship between the CI and water content, as the other spatial relations, was visible in the maps and achievable by the modeling of the data extracted from them. The CI was first influenced by the water content, bulk density and porous space. Nevertheless it also varied as function of clay, silte and organic matter content.

Das várias proposições para a determinação da umidade do solo, a técnica da reflectometria no domínio do tempo (TDR) vem despertando bastante interesse, pois apresenta características desejáveis, como a mensuração em tempo real e a possibilidade de leituras automatizadas. Por outro lado, além do custo elevado dos equipamentos, este método ainda apresenta necessidade de calibração. O desempenho no campo também carece ser mais bem estudado. Com o objetivo de avaliar a performance de um equipamento de TDR com sondas segmentadas em condições de campo, utilizando como método de referência a tensiometria, foram conduzidos experimentos em dois períodos de anos consecutivos, com a coleta de valores da umidade do solo e da respectiva leitura do TDR, em 48 baterias compostas de uma haste de TDR de cinco segmentos e um tensiômetro para cada segmento, dispostas em uma transeção em um Latossolo Vermelho-amarelo no campus da ESALQ/USP em Piracicaba/SP. Embora trabalhosa, a tensiometria mostrou-se suficiente como método de referência para a umidade no campo, pois sua metodologia é bem estabelecida e seu funcionamento não influencia nem é influenciado pelo TDR. As umidades observadas, obtidas a partir do potencial matricial e das curvas de retenção, mostraram padrão de variação semelhante entre os dois períodos, ao contrário das leituras do TDR, que exibiram médias e amplitudes pouco correspondentes entre os pontos num período e noutro. A correlação entre a umidade e a leitura do TDR foi em geral bastante reduzida e quanto mais restrito o conjunto analisado, como por haste ou segmentos, melhores resultados foram obtidos. As curvas de calibração mostraram grande variabilidade entre segmentos, tanto para o coeficiente de determinação como para o padrão da resposta, expresso pelo coeficiente angular das retas. Ajustes efetuados com os dados agrupados por haste ou por camada exibiram aumento da dispersão e forte diminuição do coeficiente de determinação. As retas de calibração também exibiram padrões diferenciados entre os períodos e ajustes com os dados dos dois conjuntos mostraram-se em geral ruins. Não foi possível identificar nenhum padrão para as variações da calibração ao longo da transeção, para algumas camadas houve indícios de dependência espacial, mas nenhuma relação definida com os atributos físicos determinados foi encontrada. Os resultados deste trabalho vêm complementar outros estudos reportados na literatura, mostrando que a aplicabilidade deste modelo de TDR segmentado é praticamente inviável, pois as variações da resposta do TDR são muito grandes entre as várias hastes ou mesmo entre segmentos de uma mesma haste, ou ainda entre os dois períodos. O trabalho também mostra a importância de estudos no campo para a validação de novos métodos ou equipamentos de determinação da umidade do solo. / Among many methods to assess soil water content, time domain reflectometry (TDR) is of recent interest due to its many desirable features as real-time method allowing automation. On the other hand, the equipment is expensive and needs calibration. More studies about its field performance are also required. With the objective of evaluating a TDR device with multilevel probes at field conditions, using tensiometry as reference method for soil water content, two experiments were conducted in consecutives years, by collecting values of soil water content and the respective TDR response in 48 batteries, that were made up of a TDR probe with 5 segments and a tensiometer for each segment, installed in a Red-yellow Oxisol (Hapludox) at the ESALQ/USP campus in Piracicaba/SP - Brasil. Although laborious, tensiometry showed to be a good reference to assess soil water content, because its methodology is well established and its functioning does not influence nor is influenced by the TDR. Soil water content obtained by matric potential and retention curves showed similar patterns in both periods, contrasting with TDR response that did not exhibit correspondence for average or range of measures between periods. The correlation between soil water content and TDR response was in general very small, and smaller data sets, per probes or per segment, showed better correlations. The calibration curves showed much variability among segments, both for the determination indexes and the response pattern. Data fitting per probe or level showed increased dispersion and strong reduction for determination indexes. Calibration also showed different patterns between periods and fitting with data from both periods were in general bad. It was not possible to identify tendencies for variations of calibration along the transect, nor correlation with offer soil attributes. These results complement other studies reported in the literature, showing that this type of probe is hard to be used, due the high variations among probes or even among segments of one probe, besides of variations in time. The importance of field studies for validation of new methods or devices for soil water content measurements are also presented.

Horizontes coesos de solos da região nordeste do Brasil, faixa de domínio dos Tabuleiros Costeiros, são uma séria limitação para a produtividade das culturas devido ao incremento acentuado da resistência do solo à penetração, sob condições de secamento do solo. Solos similares foram reconhecidos na Austrália e identificados como hardsetting. O objetivo desta pesquisa foi demonstrar que o caráter coeso de um Latossolo Amarelo localizado na Bahia é equivalente ao atributo hardsetting utilizado na descrição do solos australianos. A tese foi dividida em três capítulos. O primeiro capítulo apresenta uma revisão de literatura sobre os conceitos que norteiam os designativos coeso e hardsetting. O segundo capítulo descreve uma pesquisa de campo que demonstra que solos com caráter coeso e hardsetting apresentam o mesmo comportamento, no que diz respeito às feições morfológicas e resistência à penetração. O terceiro capítulo apresenta um estudo de laboratório que demonstrou, por meio da análise de parâmetros físicos e micromorfológicos, que o adensamento natural das partículas foi o principal fator que contribuiu com a ocorrência do caráter coeso do solo. / "Cohesive" soil horizons from the northeastern region of Brazil, domain of Low Coastal Tablelands, represents a serious limitation for crop production due to the huge increase in soil resistance to penetration as the soil dries. Similar soils have been reported in Australia and identified as hardsetting. The objective of this research was to demonstrate that a Yellow Cohesive Latosol located in Bahia is equivalent to the hardsetting attribute used in the description of Australian soils. The thesis was divide in three chapters. The first chapter presents a literature review on the "cohesive" and hardsetting concepts. The second chapter describes a field research showing that a "cohesive" and a hardsetting soil have the same behavior regarding their morphological features and resistance to penetration. The third chapter presents a laboratory study where is demonstrated that the natural densification of soil particles has a major contribution on the "cohesive" behavior.

Os solos coesos ocorrem por quase toda a faixa litorânea brasileira ocupada pelos Tabuleiros Costeiros, e caracterizam-se por apresentar um incremento acentuado na sua resistência, a qual causa restrições no seu preparo e no crescimento das plantas. Isso ocorre devido ao mesmo apresentar-se duro a extremamente duro, quando seco e friável quando úmido. Dentre os Estados da região nordeste que possuem uma extensa área litorânea ocupada pelos Tabuleiros está o Ceará, com mais de 10.000 km2 ocupados por solos que podem ser enquadrados dentro do contexto de solos coesos. Com isso levantou-se a hipótese de que os solos coesos do Estado do Ceará apresentam as mesmas restrições físicas e químicas encontradas em outros solos pertencentes aos Tabuleiros Costeiros. O objetivo geral desta tese foi identificar a ocorrência de solos coesos no Estado do Ceará, bem como, avaliar o seu comportamento através do uso de atributos morfológicos, físicos, micromorfológicos e análise de imagens. Para alcançar o objetivo geral foram estabelecidos três objetivos específicos. Cada objetivo específico contemplou um trabalho e é apresentado como um capítulo desta tese. No primeiro procurou-se identificar e caracterizar a ocorrência de solos coesos no Ceará, usando atributos morfológicos, químicos e físicos. No segundo, foi feita uma análise do comportamento de um Argissolo Acinzentado coeso através do uso de atributos físicos, especificamente: resistência tênsil (RT), resistência do solo à penetração (RP), densidade do solo (Ds) e distribuição da porosidade total (PT). No terceiro e último trabalho realizou-se um detalhamento microscópico dos solos estudados, dando ênfase à distribuição, tamanho e orientação dos poros nas áreas adensadas e abertas (menos adensadas) do horizonte coeso, através do uso da análise de imagens. Os resultados do primeiro trabalho mostraram que os horizontes coesos encontrados no Ceará, apresentam-se semelhantes aos horizontes coesos encontrados em outros Estados brasileiros. Os resultados referentes ao segundo trabalho evidenciaram que os valores de RT, RP e Ds, mostraram-se mais elevados no solo coeso do Ceará, do que em outras áreas de Tabuleiros Costeiros estudadas no Brasil. No terceiro trabalho foi observado que o adensamento natural existente nos horizontes coesos do Ceará, além de influenciar a quantidade de poros, também influencia o seu formato e a sua orientação, e que as áreas adensadas, medidas por análise de imagem, foram os principais fatores que influenciaram o espaço poroso do solo, sendo responsável pelo aumento na resistência do solo. / The hardsetting soils may be found for almost the whole Brazilian coastal area of the Coastal Tablelands. They are characterized by presenting an accentuated increment in their mechanical resistance, which causes restrictions in the soil preparation and in the plants growth. These happen due to hardsetting soils become hard to extremely hard when dry, and crumbly when humid. Ceara is one of the northeast States that possess an extensive coastal area of Coastal Tablelands, with more than 10.000 km2 of hardsetting soils. Taking this into account, it was hypothesized that hardsetting soils of Ceara State had the same physical and chemical restrictions identified in hardsetting soils in other of Coastal Tablelands. The general objective of this thesis was to identify the occurrence of hardsetting soils in the Ceara State as well as to evaluate their behavior by using the morphological, physical, and micromorphological attributes, and image analysis. To accomplish the general objective three specific objectives were established. Each specific objective led to the accomplishment of one research, and is presented as a chapter of this thesis. In the first research was identified and characterized the occurrence of hardsetting soils in Ceara by using morphologic, chemical and physical attributes. In the second research, it was studied the behavior of a hardsetting Gray Argisol by using soil physical attributes, specifically: tensile strength (RT), soil resistance to penetration (RP), soil bulk density (Ds), and pore size distribution (PT). In the last research, the soil micromorphological features were studied, giving emphasis to the distribution, size and orientation of the pores in the more denser and less denser areas of the hardsetting horizon, by using image analysis. The results of the first research showed that the hardsetting horizons found in Ceara are similar to the hardsetting horizons found in other Brazilian States. The results regarding the second research indicated that the RT, RP and Ds values were higher in the hardsetting soil of Ceara than those measured in other hardsetting soils of the Coastal Tablelands. In the third research was observed that the existent natural compaction in the hardsetting horizons of Ceara influences the amount of pores as well as their shape and orientation, and that the denser areas, measured by image analysis, were the main factor that influenced the soil porosity, being responsible for the increase in the soil mechanical resistance.

Sendo a condutividade hidráulica do solo um parâmetro que traduz a facilidade com que a água se movimenta ao longo do perfil de solo, sua determinação, principalmente no campo, torna-se imprescindível, visto que o movimento da água no solo está diretamente relacionado à produção das culturas agrícolas. Com isso, a aplicação de um método confiável que quantifique a condutividade hidráulica versus umidade, a função ( ) q K , certamente contribuirá muito para uma correta avaliação dos resultados, principalmente tendo em conta a grande variabilidade dessa função devido à variabilidade natural das características do solo. Neste trabalho, testou-se o método do perfil instantâneo com o objetivo de avaliar a função ( ) q K com e sem a suposição de gradiente unitário e também o seu comportamento com relação aos os horizontes pedológicos do solo. A determinação da função foi feita usando tensiômetros com manômetros de mercúrio instalados às profundidades de 0,1; 0,2; 0,3; 0,4; 0,5; 0,6; 0,7; 0,8; 0,9; 1,0; 1,1 e 1,2 m e curvas de retenção determinadas para as mesmas profundidades, num Latossolo Vermelho Distrófico argissólico A moderado textura média, localizado no campo experimental da Escola Superior de Agricultura "Luiz de Queiroz", Universidade de São Paulo. A análise dos resultados foi feita pelo teste F de análise de variância, a fim de testar o paralelismo, a igualdade dos interceptos e a coincidência entre as retas lnK versus q, considerando e não considerando a suposição do gradiente unitário. Além disso, foram feitos também testes de similaridade da função ( ) q K de cada profundidade do solo pelo índice de concordância de Willmott (1981). Com base na análise dos resultados pode-se dizer que, com os procedimentos utilizados de delimitação da área para inundação e redistribuição da água, podem ser empregados modelos simples que consideram unitário o gradiente de potencial total e que, pela comparação feita entre as profundidades, a separação do perfil do solo em horizontes pedológicos parece seguir a separação em camadas com funções ( ) q K semelhantes, apenas a partir da profundidade 0,6 m. / The soil hydraulic conductivity is a parameter that quantifies the ability of the soil to conduct water along the soil profile. Hence its field determination is indispensable, since soil water movement is directly related to the field crop productions. Thus, the use of an adequate method that assesses the hydraulic conductivity as a function of soil-water content, the ( ) q K function, certainly will very contribute to a proper evaluation of results, mainly if it is considered the great variability of this function as a result of the natural variability of the soil characteristics. In this work, the instantaneous profile method was tested with the objective to evaluate a) the ( ) q K function taking and not taking into consideration the assumption of unit gradient and b) the behaviour of the ( ) q K function with respect to the pedological horizons of the soil. The determination of the function was made using mercury manometer tensiometers installed at the depths of 0.1; 0.2; 0.3; 0.4; 0.5; 0.6; 0.7; 0.8; 0.9; 1.0; 1.1 and 1.2 m and soil water retention curves determined for the same depths, in an Oxisol, located at experimental fields of Escola Superior de Agricultura Luiz de Queiroz, University de São Paulo, in Piracicaba (SP), Brasil (22 o 42’ 43,3’’ W and 47 o 37’ 10,4’’ S). The result analysis was made using an F-test in order to verify the paralelism, the intercept equality and the coincidence of the lnK versus q straight lines, taking and not taking account of the assumption of unit hydraulic gradient. Basides, similarity tests of the ( ) q K function were also made by the Willmott (1981) index. From the analysis of results, it could be concluded that, with the procedures used to delimit the experimental plots in this experiment to water the soil profile and to follow the water redistribuition, the unit gradient approches seem to be adequate to assess the field ( ) q K function and also that the pedological horizons seem to follow the layers of similar ( ) q K function, for depths greater than 0.6 m.

This thesis details the development and calibration of a model created by coupling a land surface simulation model named CLASS with a hydrologic model named WATFLOOD. The resulting model, known as WatCLASS, is able to serve as a lower boundary for an atmospheric model. In addition, WatCLASS can act independently of an atmospheric model to simulate fluxes of energy and moisture from the land surface including streamflow. These flux outputs are generated based on conservation equations for both heat and moisture ensuring result continuity. WatCLASS has been tested over both the data rich BOREAS domains at fine scales and the large but data poor domain of the Mackenzie River at coarse scale. The results, while encouraging, point to errors in the model physics related primarily to soil moisture transport in partially frozen soils and permafrost. Now that a fully coupled model has been developed, there is a need for continued research by refining model processes and test WatCLASS's robustness using new datasets that are beginning to emerge. Hydrologic models provide a mechanism for the improvement of atmospheric simulation though two important mechanisms. First, atmospheric inputs to the land surface, such as rainfall and temperature, are transformed by vegetation and soil systems into outputs of energy and mass. One of these mass outputs, which have been routinely measured with a high degree of accuracy, is streamflow. Through the use of hydrologic simulations, inputs from atmospheric models may be transformed to streamflow to assess reliability of precipitation and temperature. In this situation, hydrologic models act in an analogous way to a large rain gauge whose surface area is that of a watershed. WatCLASS has been shown to be able to fulfill this task by simulating streamflow from atmospheric forcing data over multi-year simulation periods and the large domains necessary to allow integration with limited area atmospheric models. A second, more important, role exists for hydrologic models within atmospheric simulations. The earth's surface acts as a boundary condition for the atmosphere. Besides the output of streamflow, which is not often considered in atmospheric modeling, the earth's surface also outputs fluxes of energy in the form of evaporation, known as latent heat and near surface heating, known as sensible heat. By simulating streamflow and hence soil moisture over the land surface, hydrologic models, when properly enabled with both energy and water balance capabilities, can influence the apportioning of the relative quantities of latent and sensible heat flux that are required by atmospheric models. WatCLASS has shown that by improving streamflow simulations, evaporation amounts are reduced by approximately 70% (1271mm to 740mm) during a three year simulation period in the BOREAS northern old black spruce site (NSA-OBS) as compared to the use of CLASS alone. To create a model that can act both as a lower boundary for the atmosphere and a hydrologic model, two choices are available. This model can be constructed from scratch with all the caveats and problems associated with proving a new model and having it accepted by the atmospheric community. An alternate mechanism, more likely to be successfully implemented, was chosen for the development of WatCLASS. Here, two proven and well tested models, WATFLOOD and CLASS, were coupled in a phased integration strategy that allowed development to proceed on model components independently. The ultimate goal of this implementation strategy, a fully coupled atmospheric - land surface - hydrologic model, was developed for MC2-CLASS-WATFLOOD. Initial testing of this model, over the Saguenay region of Quebec, has yet to show that adding WATFLOOD to CLASS produces significant impacts on atmospheric simulation. It is suspected, that this is due to the short term nature of the weather simulation that is dominated by initial conditions imposed on the atmospheric model during the data assimilation cycle. To model the hydrologic system, using the domain of an atmospheric model, requires that methods be developed to characterize land surface forms that influence hydrologic response. Methods, such as GRU (Grouped Response Unit) developed for WATFLOOD, need to be extended to taken advantage of alternate data forms, such as soil and topography, in a way that allows parameters to be selected <I>a priori</I>. Use of GIS (Geographical Information System) and large data bases to assist in development of these relationships has been started here. Some success in creating DEMs, (Digital Elevation Model) which are able to reproduce watershed areas, was achieved. These methods build on existing software implementations to include lake boundaries information as a topographic data source. Other data needs of hydrologic models will build on relationships between land cover, soil, and topography to assist in establishing grouping of these variables required to determine hydrologic similarity. This final aspect of the research is currently in its infancy but provides a platform from which to explore for future initiatives. Original contributions of this thesis are centered on the addition of a lateral flow generation mechanism within a land surface scheme. This addition has shown a positive impact on flux returns to the atmosphere when compared to measured values and also provide increased realism to the model since measured streamflow is reproduced. These contributions have been encapsulated into a computer model known as WatCLASS, which together with the implementation plan, as presented, should lead to future atmospheric simulation improvements.

This thesis details the development and calibration of a model created by coupling a land surface simulation model named CLASS with a hydrologic model named WATFLOOD. The resulting model, known as WatCLASS, is able to serve as a lower boundary for an atmospheric model. In addition, WatCLASS can act independently of an atmospheric model to simulate fluxes of energy and moisture from the land surface including streamflow. These flux outputs are generated based on conservation equations for both heat and moisture ensuring result continuity. WatCLASS has been tested over both the data rich BOREAS domains at fine scales and the large but data poor domain of the Mackenzie River at coarse scale. The results, while encouraging, point to errors in the model physics related primarily to soil moisture transport in partially frozen soils and permafrost. Now that a fully coupled model has been developed, there is a need for continued research by refining model processes and test WatCLASS's robustness using new datasets that are beginning to emerge. Hydrologic models provide a mechanism for the improvement of atmospheric simulation though two important mechanisms. First, atmospheric inputs to the land surface, such as rainfall and temperature, are transformed by vegetation and soil systems into outputs of energy and mass. One of these mass outputs, which have been routinely measured with a high degree of accuracy, is streamflow. Through the use of hydrologic simulations, inputs from atmospheric models may be transformed to streamflow to assess reliability of precipitation and temperature. In this situation, hydrologic models act in an analogous way to a large rain gauge whose surface area is that of a watershed. WatCLASS has been shown to be able to fulfill this task by simulating streamflow from atmospheric forcing data over multi-year simulation periods and the large domains necessary to allow integration with limited area atmospheric models. A second, more important, role exists for hydrologic models within atmospheric simulations. The earth's surface acts as a boundary condition for the atmosphere. Besides the output of streamflow, which is not often considered in atmospheric modeling, the earth's surface also outputs fluxes of energy in the form of evaporation, known as latent heat and near surface heating, known as sensible heat. By simulating streamflow and hence soil moisture over the land surface, hydrologic models, when properly enabled with both energy and water balance capabilities, can influence the apportioning of the relative quantities of latent and sensible heat flux that are required by atmospheric models. WatCLASS has shown that by improving streamflow simulations, evaporation amounts are reduced by approximately 70% (1271mm to 740mm) during a three year simulation period in the BOREAS northern old black spruce site (NSA-OBS) as compared to the use of CLASS alone. To create a model that can act both as a lower boundary for the atmosphere and a hydrologic model, two choices are available. This model can be constructed from scratch with all the caveats and problems associated with proving a new model and having it accepted by the atmospheric community. An alternate mechanism, more likely to be successfully implemented, was chosen for the development of WatCLASS. Here, two proven and well tested models, WATFLOOD and CLASS, were coupled in a phased integration strategy that allowed development to proceed on model components independently. The ultimate goal of this implementation strategy, a fully coupled atmospheric - land surface - hydrologic model, was developed for MC2-CLASS-WATFLOOD. Initial testing of this model, over the Saguenay region of Quebec, has yet to show that adding WATFLOOD to CLASS produces significant impacts on atmospheric simulation. It is suspected, that this is due to the short term nature of the weather simulation that is dominated by initial conditions imposed on the atmospheric model during the data assimilation cycle. To model the hydrologic system, using the domain of an atmospheric model, requires that methods be developed to characterize land surface forms that influence hydrologic response. Methods, such as GRU (Grouped Response Unit) developed for WATFLOOD, need to be extended to taken advantage of alternate data forms, such as soil and topography, in a way that allows parameters to be selected <I>a priori</I>. Use of GIS (Geographical Information System) and large data bases to assist in development of these relationships has been started here. Some success in creating DEMs, (Digital Elevation Model) which are able to reproduce watershed areas, was achieved. These methods build on existing software implementations to include lake boundaries information as a topographic data source. Other data needs of hydrologic models will build on relationships between land cover, soil, and topography to assist in establishing grouping of these variables required to determine hydrologic similarity. This final aspect of the research is currently in its infancy but provides a platform from which to explore for future initiatives. Original contributions of this thesis are centered on the addition of a lateral flow generation mechanism within a land surface scheme. This addition has shown a positive impact on flux returns to the atmosphere when compared to measured values and also provide increased realism to the model since measured streamflow is reproduced. These contributions have been encapsulated into a computer model known as WatCLASS, which together with the implementation plan, as presented, should lead to future atmospheric simulation improvements.

published_or_final_version / abstract / toc / Geography and Geology / Master / Master of Philosophy

A two year study was initiated in 1999 at the Macdonald Campus Farm, on a 2.4 ha site consisting mainly of St-Amable sand to shallow loamy sand (Typic Endoaquent; Humic Gleysol). The site was planted to alfalfa ( Medicago sativa) prior to the establishment of the experimental design and has been planted to corn (Zea mays L.) since 1991. The experiment was set up as a randomized complete block design and consisted of three tillage levels (NT, no-till; RT, reduced tillage; and CT, conventional tillage) and two residue levels (-R, no residues; +R, with residues). This study was performed in order to ascertain the effect of these management systems on soil organic carbon (SOC), particulate organic matter carbon (POM-C), microbial biomass carbon (MBC) levels and soil physical properties. Soil physical properties measured included bulk density, saturated hydraulic conductivity (Ksat), dry and wet aggregate stability, total porosity, macroporosity at -1 kPa and -6 kPa of matric potential, and moisture content. Crop yield was also measured. Results indicated that lower rates of mineralization of POM-C under NT+R in the surface 0--10 cm led to significantly higher SOC at the same depth. There were no significant differences between treatments at the 10--20 cm depth. The MBC was not a good indicator of the differences in soil organic matter levels between treatments. No-till treatments had higher bulk densities, reduced total porosity and macroporosity, increased moisture content, and increased constant-head measured Ksat. Differences in Ksat as measured with the Guelph permeameter, were not significant between treatments, most likely due to increased earthworm activity in NT. There were no significant differences in crop yield in 1999, but a significantly wetter year in 2000 caused lower grain yields in NT+R compared to NT-R and CT+R.

Long-term environmental quality is closely linked to maintaining soil quality. Therefore, assessment of the effect of land use on soil chemical, physical and biological properties provides fundamental information about soil sustainability. The role of vegetation on soil structural and aggregate stability has received much attention over the last few decades. The loss of soil due to erosion and reduction in organic matter due to change in vegetation is usually caused by invasion of grassland by trees. The response of soils to land use depends on the inherent soil properties and environmental conditions thus the effect of land use on soil stability is site specific. The objectives of this study were to assess and compare soil aggregate stability under trees and in open grassland from open-savannah woodland and nearby deeply eroded dongas (gullies). Within the dongas, tree encroachment was expected to improve soil structural stability by increasing the organic matter content. Results that were obtained from this study are aimed at improving management of soil for smallholder and pastoral farming systems. The study site was located at Potshini, KwaZulu-Natal Province, South Africa, about 18 km south-east of Bergville. The vegetation of the area is classified as grassland biome. Acacia sieberiana var. Woodii has encroached into the valley especially onto the eroded areas. Within the study area itself the dominant parent materials are sandstone and dolerite, with colluvium covering the lower slopes. The dominant soil was classified as Hutton form. Effects of different vegetation types (grass and trees) on the soil structural stability was evaluated for their effects on soil organic carbon (SOC), mean weight diameter, bulk density, root density, clay mineralogy and some other chemical properties. For this study the site was divided into two areas i.e., the donga (D) and the grassland (G). These were then each subdivided into two parts namely, donga under a tree (DUT) with a corresponding open area (DOA), and grassland under a tree (GUT) with its corresponding open area (GOA). Three bulk samples were collected from each sampling area from 0 to 20 cm (topsoil only) using a spade. Samples for bulk density were collected from 0 - 10 cm. For soil aggregate stability determinations, samples were dried and sieved to collect soil aggregates between 2.8 and 5 mm. Some of the bulk sample was analysed for SOC, pH, exchangeable bases, nitrogen and clay mineralogy. For root biomass, soil samples were collected from 0 - 10 cm using a stainless steel cylinder of 1766 cm3 volume. The type of vegetation affected the soil physical and chemical properties of the soil in the investigated horizon (A horizon). Although the results were not significantly (p > 0.05) different, the open area showed a positive effect on soil structural stability where higher soil aggregate stability and root density were observed, as opposed to under trees in both donga and grassland. Due to the non-significant differences observed in both donga and grassland, the effect of trees and grass on the aggregate stability was considered as the positive effect. The bulk density showed an unusual trend. Bulk density was higher in the open areas where high aggregate stability was observed. From these results it was concluded that because trees have larger roots than grass these are more effective in loosening the soil particles but less effective in stabilizing the soil aggregates. In other words, the high amount of fine roots increased the soil stability while large roots improved the loosening of the soil particles. The amount of vegetation considerably affected all the physical and chemical soil properties investigated. The SOC and root density was considerably higher under grassland compared to the donga. The clay mineralogy differed between the donga and grassland. The donga soils had a higher kaolinite content and amount of interstratified smectite/mica compared to grassland. The presence of interstratified smectite/mica in the donga could cause lower aggregate stability due to shrinking-swelling cycles during wetting-drying conditions. The effects of mineralogy have been overridden by organic carbon in the grassland. From this study it was revealed that soil aggregate stability is the product of interactions between soil clay minerals, and organic fractions which are influenced by soil environment, land use and soil properties. This means that all the factors involved in aggregate formation and their stabilization are interdependent. It is suggested that factors that were driving the soil aggregate stability in the donga were different to the ones in grassland. It is suggested that the aggregate stability was driven by inorganic factors i.e., clay, clay mineralogy, bases and CEC in the donga while organic carbon and plants roots were dominating in the grassland / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.