Thesis (MScAgric)--University of Stellenbosch, 2003. / ENGLISH ABSTRACT: Near infrared reflectance spectroscopy (NIRS) was evaluated as a tool to predict the chemical composition of ostrich meat, mutton and feedstuffs. Seventy-three calibrations were developed. NIRS analyses were conducted on an InfraAlyzer 500 spectrophotometer between 1100 and 2500 nm wavelengths. Near infrared reflectance calibrations were developed for (i) the proximate chemical composition of ostrich meat; (ii) the proximate chemical and mineral composition of mutton; (iii) the chemical composition and digestibility of lupins, full-fat canola and the determination of alkaloids in lupins; (iv) the chemical composition and digestibility of winter grains and maize; (v) the chemical composition and digestibility of cereal hay, cereal straw, wheat stubble and alfalfa-grass/hay mixtures. The chemical composition of different types of winter grain produced in the Western Cape area of South Africa, as well as the chemical composition and digestibility of winter grain hay and straw produced in a Mediterranean rainfall area were also determined. Near infrared reflectance spectrometry proved to be successful for the prediction of crude protein (CP) and fat in both animal species, as well as for ash and dry matter (DM) in freeze-dried mutton. Accurate calibrations were also developed for certain minerals (K, P, Na, Mg, Fe and Zn). Calibrations proved to be accurate for all the relevant chemical constituents in lupins, as well as the DM, CP and fat calibrations in full-fat canola. The alkaloid calibration showed potential although more samples should be included for a more accurate calibration. Winter grain calibrations were accurate for fat, acid detergent fibre (ADF), in vitro organic matter digestibility (IVOMD), lysine and methionine. All chemical composition calibrations of maize, except for ash, were accurate for future prediction. Chemical composition calibrations of wheat stubble and alfalfa-grass hay mixtures resulted in prediction with similar or slightly higher accuracy than calibrations reported in the literature. The calibrations developed for cereal hay and straw were not suitable for prediction purposes. Possible reasons were discussed to explain these inaccurate calibrations. Variation shown between different types of cereal grain (2-row barley, 6-row barley, oats, wheat and triticale) accentuated the need for the analysis of different batches of grain produced under different conditions for use in animal feed. The chemical composition and digestibility of winter grain straw and hay were also obtained. These values provide a database for calculation of inclusion levels of these feedstuffs in animal diets. These investigations showed the NIRS to be a successful and rapid tool for the prediction of the chemical composition of ostrich and lamb meat and locally produced feedstuffs. / AFRIKAANSE OPSOMMING: Naby infrarooi refleksie spektroskopie (NIRS) is geëvalueer as 'n metode om die chemiese samestelling van volstruis- en skaapvleis, asook voedingsgewasse te voorspel. Drie en sewentig kalibrasies is ontwikkel. NIRS ontledings is gedoen met ‘n InfraAlyzer 500 spektrofotometer tussen die golflengtes 1100 en 2500 nm. Naby infrarooi refleksie kalibrasies is ontwikkel vir (i) die basiese chemiese samestelling van volstruisvleis; (ii) die basiese chemiese en minerale samestelling van lamsvleis; (iii) die chemiese samestelling en verteerbaarheid van lupiene en volvet kanola en die bepaling van alkaloïede in lupiene; (iv) die chemiese samestelling en verteerbaarheid van wintergrane en mielies; (v) die chemiese samestelling en verteerbaarheid van graanhooie, graanstrooie, koringstoppels and lusern-gras/hooi mengsels. Die chemiese samestelling van verskillende tipes wintergrane wat in die Wes-Kaap omgewing van Suid-Afrika geproduseer word, sowel as die chemiese samestelling en verteerbaarheid van wintergraanhooi en –strooi wat geproduseer word in ‘n Mediterreënse reënval omgewing, is ook bepaal. Naby infrarooi refleksie spektrofotometrie was suksesvol in die voorspelling van ru-proteïen (RP) en vet in beide dierspesies, sowel as vir die bepaling van as en droë materiaal (DM) in gevriesdroogde lamsvleis. Akkurate kalibrasies is ook ontwikkel vir sekere minerale (K, P, Na, Mg, Fe en Zn). Kalibrasies ontwikkel vir die chemiese samestellings komponente in lupiene is reg deur die bank akkuraat, sowel as die DM, RP en vet kalibrasies vir volvet kanola. Die alkaloïed kalibrasie het goeie potensiaal getoon, alhoewel meer alkaloïedmonsters benodig word vir ‘n akkurate kalibrasie. Wintergraan kalibrasies was akkuraat vir vet, suurbestande vesels (SBV), in vitro organiese materiaal verteerbaarheid (IVOMV), lisien en metionien. Al die chemiese komponent kalibrasies vir mielies, behalwe die een vir as, was akkuraat vir toekomstige bepalings. Chemiese komponent kalibrasies vir koring stoppels en luserngras/ hooi mengsels het akkuraatheid getoon wat gelyk, of effe beter, is as kalibrasies wat in die literatuur opgeteken is. Kalibrasies ontwikkel vir graanhooie en graanstrooie was nie geskik vir voorspellingsdoeleindes nie. Moontlike redes vir die onakkurate kalibrasies is volledig bespreek. Variasie gevind tussen verskillende tipes wintergrane (2-ry gars, 6-ry gars, hawer, koring en korog) het die behoefte beklemtoon vir ontledings van grane geproduseer onder verskillende omgewingstoestande vir gebruik in dierevoeding. Die chemiese samestelling en verteerbaarheid van wintergraanstrooi en –hooi is ook bepaal. Hierdie waardes verskaf ‘n databasis vir berekening by insluiting van hierdie voergewasse in diererantsoene. Hierdie ondersoek het bewys dat NIRS ‘n suksesvolle en vinnige metode is vir die voorspelling van die chemiese samestelling van volstruis- en lamsvleis en plaaslik geproduseerde voergewasse.

A cadeia produtiva da cana-de-açúcar vem sofrendo problemas de diversas naturezas, sendo a mais comum a estiagem, agravada pelas mudanças climáticas que reduzem a disponibilidade de água no solo, afetando diretamente a produtividade da cultura. Uma grande proporção da cultura da cana-de-açúcar não é irrigada, sendo sujeita a alterações entre estações úmidas e secas em condições tropicais e subtropicais, mas quando é irrigada, tem-se observado um incremento significativo na produtividade da cultura. As necessidades hídricas da cultura devem ser atendidas, tanto, na quantidade requerida, quanto no momento oportuno. Para isto, devem ser quantificados parâmetros relacionados com o seu estado hídrico. No entanto, os métodos empregados convencionalmente são demorados, custosos e invasivos. Como alternativa que ajuda a reduzir tempo e custos, o sensoriamento remoto hiperespectral vem sendo utilizado para estimar o estado hídrico em diferentes escalas, uma vez que permite a captura de grande quantidade de informação rapidamente. Para o presente trabalho, o comportamento espectral da vegetação de 400 a 2500 nm, foi utilizado na quantificação de alguns parâmetros que estabelecem o seu estado hídrico. As avaliações tanto em casa de vegetação quanto em laboratório foram feitas em folhas de cana-de-açúcar submetidas a déficit hídrico programado. Para os dados de laboratório foram obtidos R2 > 0,8 na região do visível e R2 < 0,55, na região do infravermelho próximo para CRA (conteúdo relativo de água). Para EEA (espessura equivalente da água) foi obtido um R2 < 0,6 na região do infravermelho próximo. / The sugarcane agribusiness has been suffering several kinds of problems. The most common is the drought caused by the weather changes, which reduce the water availability in the soil, affecting directly the crop yield. A large proportion of the sugarcane crop is not irrigated undergoing changes between wet and dry seasons in tropical and subtropical conditions, but when it is irrigated, it has been possible to observe an increase in the crop yield. The crop water requirements must be provided, both at the required amount and at the right time. To do this, parameters related to its moisture status have to be quantized. However, conventional methods are slow, invasive and expensive. As an alternative to reduce time and costs, the hyperspectral remote sensing has been being used to estimate the water status at different scales, because it allows capturing big amounts of information quickly. In the present study, the spectral behavior of vegetation between 400 and 2500 nm was used to quantify some parameters that establish its water status. The evaluations were conducted both in the greenhouse and the laboratory on sugarcane leaves under programmed water deficit. The laboratory data obtained were R2> 0.8 in the visible region and R2 <0.55 in the near infrared region for the RWC (relative water content). For the EWT (equivalent water thickness) was obtained a R2 <0.6 in the near infrared region.

A cadeia produtiva da cana-de-açúcar vem sofrendo problemas de diversas naturezas, sendo a mais comum a estiagem, agravada pelas mudanças climáticas que reduzem a disponibilidade de água no solo, afetando diretamente a produtividade da cultura. Uma grande proporção da cultura da cana-de-açúcar não é irrigada, sendo sujeita a alterações entre estações úmidas e secas em condições tropicais e subtropicais, mas quando é irrigada, tem-se observado um incremento significativo na produtividade da cultura. As necessidades hídricas da cultura devem ser atendidas, tanto, na quantidade requerida, quanto no momento oportuno. Para isto, devem ser quantificados parâmetros relacionados com o seu estado hídrico. No entanto, os métodos empregados convencionalmente são demorados, custosos e invasivos. Como alternativa que ajuda a reduzir tempo e custos, o sensoriamento remoto hiperespectral vem sendo utilizado para estimar o estado hídrico em diferentes escalas, uma vez que permite a captura de grande quantidade de informação rapidamente. Para o presente trabalho, o comportamento espectral da vegetação de 400 a 2500 nm, foi utilizado na quantificação de alguns parâmetros que estabelecem o seu estado hídrico. As avaliações tanto em casa de vegetação quanto em laboratório foram feitas em folhas de cana-de-açúcar submetidas a déficit hídrico programado. Para os dados de laboratório foram obtidos R2 > 0,8 na região do visível e R2 < 0,55, na região do infravermelho próximo para CRA (conteúdo relativo de água). Para EEA (espessura equivalente da água) foi obtido um R2 < 0,6 na região do infravermelho próximo. / The sugarcane agribusiness has been suffering several kinds of problems. The most common is the drought caused by the weather changes, which reduce the water availability in the soil, affecting directly the crop yield. A large proportion of the sugarcane crop is not irrigated undergoing changes between wet and dry seasons in tropical and subtropical conditions, but when it is irrigated, it has been possible to observe an increase in the crop yield. The crop water requirements must be provided, both at the required amount and at the right time. To do this, parameters related to its moisture status have to be quantized. However, conventional methods are slow, invasive and expensive. As an alternative to reduce time and costs, the hyperspectral remote sensing has been being used to estimate the water status at different scales, because it allows capturing big amounts of information quickly. In the present study, the spectral behavior of vegetation between 400 and 2500 nm was used to quantify some parameters that establish its water status. The evaluations were conducted both in the greenhouse and the laboratory on sugarcane leaves under programmed water deficit. The laboratory data obtained were R2> 0.8 in the visible region and R2 <0.55 in the near infrared region for the RWC (relative water content). For the EWT (equivalent water thickness) was obtained a R2 <0.6 in the near infrared region.

Near infrared reflectance spectroscopy (NIRS) and geostatistical techniques were used to predict diet quality of sub-Saharan African (SSA) livestock, and to create cokriged estimated diet quality maps for cattle across a landscape. Rations of native vegetation were stall-fed to cattle (Bos indicus), sheep (Ovis aries), and goats (Capra hircus) to generate diet-fecal pair data. Trials were conducted in Ethiopia, Kenya, Uganda, Tanzania, and Ghana. Historical data from Ethiopia, Nigeria, and Niger were included. Diet samples were analyzed for crude protein (CP%), and digestible organic matter (DOM%), while feces were scanned for NIR spectra. NIRS equations were developed from data using modified partial least square (MPLS) regression. Coefficients of determination (R2) of CP for cattle, sheep, and goats were 0.92, 0.95, and 0.97, with corresponding standard errors of calibration (SEC) being 0.90, 0.79, and 0.80, respectively. Standard errors of cross validation (SECV) for CP were 1.12%, 1.08%, and 1.03% for cattle, sheep, and goats, respectively. R2 and SEC values for DOM were 0.88, 0.94, 0.94 and 2.82%, 1.68%, and 2.65%, for cattle, sheep, and goats, respectively. Corresponding SECV values for DOM were 3.26%, 2.07%, and 3.30%, respectively. The statistics reported were within the acceptable limits for NIRS calibrations. The results indicate that dietary CP and DOM of free-ranging SSA livestock can be predicted with the same precision as that of conventional wet chemistry methods. The cattle equation was used to predict cattle fecal samples collected, from February to August 2000, from selected households located within the northern Ghana savanna. The predicted CP% and DOM% were used with Normalized Differential Vegetation Index (NDVI) data, and cokriging technique to create diet quality maps for March and July 2000 for the northern Ghana savanna. Cross validation results indicated a moderate capability of cokriging to estimate predicted CP% for March (r2 = 0.687, SEp = 1.736) and July (r2 = 0.513, SEp = 1.558). Cokriged-estimated DOM value for July was above average (r2 = 0.584, SEp = 3.611), while March DOM% estimation was rather poor (r2 = 0.132, SEp = 3.891). The techniques of cokriging and creation of diet quality maps were moderately successful in this study.

Quantification of blood fraction (fblood), blood oxygenation (S<img src="http://onlinelibrary.wiley.com/store/10.1002/jbio.201200055/asset/equation/tex2gif-inf-2.gif?v=1&amp;t=h70man4a&amp;s=4a6d004ec608a2a6ec8e8597f73bdb6be30286e8" />), melanin, lipofuscin and oxidised and reduced Cytochrome aa 3 and c was done from diffuse reflectance spectra captured in cortex, white matter, globus pallidus internus (GPi) and subthalamus during stereotactic implantations of 29 deep brain stimulation (DBS) electrodes with the aim of investigating whether the chromophores can give physiological information about the targets for DBS. Double-sided Mann-Whitney U -tests showed more lipofuscin in GPi compared to white matter and subthalamus (p &lt; 0.05). Compared to the other structures, fbloodwas significantly higher in cortex (p &lt; 0.05) and S<img src="http://onlinelibrary.wiley.com/store/10.1002/jbio.201200055/asset/equation/tex2gif-inf-4.gif?v=1&amp;t=h70man4c&amp;s=855c70105e88a292de25618487573dfc7d30e08a" /> lower in GPi (p &lt; 0.05). Median values and range for fblood were 1.0 [0.2–6.0]% in the cortex, 0.3 [0.1–8.2]% in white matter, 0.2 [0.1–0.8]% in the GPi and 0.2 [0.1–11.7]% in the subthalamus. Corresponding values for S<img src="http://onlinelibrary.wiley.com/store/10.1002/jbio.201200055/asset/equation/tex2gif-inf-6.gif?v=1&amp;t=h70man4e&amp;s=151ec25bee7270bcfc2292e70d6f4aea18348dbc" /> was 20 [0–81]% in the cortex, 29 [0–78]% in white matter, 0 [0–0]% in the GPi and 0 [0–92]% in the subthalamus. In conclusion, the measurements indicate very low oxygenation and blood volume for DBS patients, especially in the GPi. It would be of great interest to investigate whether this is due to the disease, the normal situation or an artefact of doing invasive measurements.

Near infrared reflectance spectroscopy (NIRS) and geostatistical techniques were used to predict diet quality of sub-Saharan African (SSA) livestock, and to create cokriged estimated diet quality maps for cattle across a landscape. Rations of native vegetation were stall-fed to cattle (Bos indicus), sheep (Ovis aries), and goats (Capra hircus) to generate diet-fecal pair data. Trials were conducted in Ethiopia, Kenya, Uganda, Tanzania, and Ghana. Historical data from Ethiopia, Nigeria, and Niger were included. Diet samples were analyzed for crude protein (CP%), and digestible organic matter (DOM%), while feces were scanned for NIR spectra. NIRS equations were developed from data using modified partial least square (MPLS) regression. Coefficients of determination (R2) of CP for cattle, sheep, and goats were 0.92, 0.95, and 0.97, with corresponding standard errors of calibration (SEC) being 0.90, 0.79, and 0.80, respectively. Standard errors of cross validation (SECV) for CP were 1.12%, 1.08%, and 1.03% for cattle, sheep, and goats, respectively. R2 and SEC values for DOM were 0.88, 0.94, 0.94 and 2.82%, 1.68%, and 2.65%, for cattle, sheep, and goats, respectively. Corresponding SECV values for DOM were 3.26%, 2.07%, and 3.30%, respectively. The statistics reported were within the acceptable limits for NIRS calibrations. The results indicate that dietary CP and DOM of free-ranging SSA livestock can be predicted with the same precision as that of conventional wet chemistry methods. The cattle equation was used to predict cattle fecal samples collected, from February to August 2000, from selected households located within the northern Ghana savanna. The predicted CP% and DOM% were used with Normalized Differential Vegetation Index (NDVI) data, and cokriging technique to create diet quality maps for March and July 2000 for the northern Ghana savanna. Cross validation results indicated a moderate capability of cokriging to estimate predicted CP% for March (r2 = 0.687, SEp = 1.736) and July (r2 = 0.513, SEp = 1.558). Cokriged-estimated DOM value for July was above average (r2 = 0.584, SEp = 3.611), while March DOM% estimation was rather poor (r2 = 0.132, SEp = 3.891). The techniques of cokriging and creation of diet quality maps were moderately successful in this study.

O presente estudo teve como objetivo avaliar o teor de Cu, Cr, Ni, Zn, Pb, Mn, Al e Fe e sua partição geoquímica em sedimentos da bacia hidrográfica do rio dos Sinos, Sul do Brasil. Esta bacia apresenta elevada densidade populacional e industrial, especialmente dos setores de metalurgia, galvanoplastia, siderurgia, petroquímica e curtumes. As concentrações totais de metais, determinadas por espectroscopia de fluorescência de raios X, e as concentrações nas fases dos sedimentos, determinadas através da aplicação do esquema de extração sequencial BCR-701 proposto pelo Community Bureau of Reference, da Comunidade Europeia, utilizando ICP/OES, foram avaliadas nos sedimentos com granulometria <63 μm. Também foram considerados os teores extraídos através de uma solução de HCl 0,5 mol/L nesta fração dos sedimentos. Os resultados na região industrializada e com alta densidade populacional mostraram um aumento nas concentrações totais, para Cu, Cr, Zn e Pb, e nas frações potencialmente móveis para os metais Cu, Cr, Zn e Ni extraídos sequencialmente. O Cr destacou-se especialmente no arroio Portão, apresentando uma concentração total de 1286 mg/kg, devido à influência de curtumes localizados nesta região. Os sedimentos também foram caracterizados através de espectroscopia de reflectância, método rápido e não destrutivo de caracterização dos diferentes materiais, na região entre 400 e 2500 nm. Nos espectros de reflectância das amostras avaliadas observou-se que as principais feições espectrais são resultantes das absorções decorrentes da água, das ligações Al-OH e dos teores de ferro presentes. Ainda ficou evidenciada a influência da granulometria, dos teores de ferro e matéria orgânica sobre os espectros de reflectância resultantes. Não foi possível observar as feições espectrais decorrentes dos metais pesados, provavelmente devido às altas concentrações de ferro observadas nas amostras e devido ao fato destes metais apresentarem as suas feições espectrais nas mesmas regiões de comprimento de onda onde ocorrem as feições do ferro. / The present study aimed to evaluate the content of Cu, Cr, Ni, Zn, Pb, Mn, Al, and Fe as well as their geochemical partitioning in sediments of the Sinos river basin in southern Brazil. This basin has a high population density and a great number of industries, especially metallurgy, electroplating works, steel mills, petrochemicals, and tanneries. The total metal concentrations, determined by X-ray fluorescence, and the concentrations in the sediment phases, determined by applying the BCR-701 sequential extraction method proposed by the Community Bureau of Reference of the European Community, using ICP/OES, were evaluated in the sediment fraction <63 μm. The contents extracted by a solution of 0.5 mol/L HCl were also considered in this fraction. Results in the industrialized and densely populated region showed an increase in the total concentrations for Cu, Cr, Zn, and Pb, and in the potentially mobile fractions for the sequentially extracted metals Cu, Cr, Zn and Ni. Cr stood out particularly in the Portão stream, showing a total concentration of 1,286 mg/kg, due to the influence of tanneries in this region. The sediments were also characterized by reflectance spectroscopy, a rapid and nondestructive method for characterization of different materials, in the region between 400 and 2500 nm. The reflectance spectra of the samples showed that the main spectral features are the result of the absorptions due to water, Al-OH bonds and contents of iron. The influences of particle size, organic matter and mineral composition on the absorption features and on the reflection coefficient of the spectra were also evidenced. It was not possible to observe the spectral features resulting from metals, probably due to high iron concentrations observed in the samples, and because these metals make their spectral features on the same wavelength regions where there are the iron spectral features.

Submitted by DANILO ALMEIDA BALDO DO CARMO (danilo_baldo@hotmail.com) on 2018-05-23T15:53:30Z No. of bitstreams: 1 Tese_DaniloBaldo.pdf: 4225878 bytes, checksum: f6162ae37b5db55be022b39f8669ba7e (MD5) / Approved for entry into archive by Alexandra Maria Donadon Lusser Segali null (alexmar@fcav.unesp.br) on 2018-05-23T16:34:17Z (GMT) No. of bitstreams: 1 carmo_dab_dr_jabo.pdf: 4225878 bytes, checksum: f6162ae37b5db55be022b39f8669ba7e (MD5) / Made available in DSpace on 2018-05-23T16:34:17Z (GMT). No. of bitstreams: 1 carmo_dab_dr_jabo.pdf: 4225878 bytes, checksum: f6162ae37b5db55be022b39f8669ba7e (MD5) Previous issue date: 2018-04-05 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A agricultura moderna no contexto internacional propõe inovações para o agronegócio, em especial para as práticas agrícolas. No entanto, este avanço deverá auxiliar o empresário rural a aumentar a eficiência, utilizando de forma sustentável os recursos naturais: água, terra e energia. Desse modo, objetivou-se, neste estudo, verificar o potencial dos sensores para obter os dados espectrais de um argissolo com cultivo de café. Utilizou-se de um conjunto de dados de um grid com 51 amostras de solo, em 39 hectares, com cultivo de café (Catuaí Amarelo, linhagem IAC 62). A obtenção das variáveis da cor: matiz, valor, croma, red, green e blue foi determinada por espectroscopia de reflectância difusa (ERD) e sensor próximo móbile (SPM). Os resultados mostraram relação com os atributos do solo nos seguintes casos: redSPM e argila: (R2 = 0,70, p < 0,01), greenSPM e CTC (R2 = 0,73, p < 0,01), redSPM e P (R2 = 0,64, p < 0,05), blueSPM e EC (R2 = 0,96, p < 0,01) e, greenSPM e Fe (R2 = 0,75, p < 0,01). Para os atributos da planta, foram: greenSPM e P17 (R2 = 0,70, p < 0,01), defeitos com regressões variando de 0,71 - 0,84 e, redSPM e produtividade (R2 = 0,95, p < 0,01). Os resultados mostraram que o uso da cor determinada pelo red, green e blue permitiu identificar atributos do solo e áreas com potenciais produtivos e qualitativos do café. Para o estudo no Planalto Ocidental Paulista, realizou-se um grid em área de 13 milhões de hectares, totalizando 600 amostras na profundidade de 0,00-0,20 metros. Nestas amostras, foram determinadas as variáveis da cor. Observou-se semelhança no uso dos métodos ERD e SPM. Os semivariogramas da ERD tiveram alcance para as variáveis matiz, valor, croma, red, green e blue de 112 km, 216 km, 19 km, 207 km, 183 km e 158 km, respectivamente. Com o uso do SPM, os alcances foram de: 19 km, 226 km, 30 km, 160 km, 218 km e 75 km, respectivamente. Os semivariogramas cruzados estimaram os atributos físico-químicos (Argila, capacidade de troca catiônica - CTC, fósforo - P e hematita - Hm). Os semivariogramas cruzados explicaram a distância máxima de correlação observada entre as variáveis. Os resultados do SPM mostraram que a cor nos sistemas red, green e blue pode ser utilizada na predição do teor de argila (77%), CTC (99%), P (90%) e Hm (70%). Com o uso da ERD, os sistemas red, green e blue predizem o teor de argila (79%), CTC (94%), P (86%) e Hm (67%). Estes resultados mostram que a cor é um indicativo seguro dos atributos do solo. Como alternativa de baixo custo, o SPM mostrou ser uma ferramenta eficiente que possibilita a avaliação destes atributos. / Modern agriculture in the international context proposes innovations for agribusiness, especially for agricultural practices. However, this progress must help the rural entrepreneur to increase efficiency, using sustainable natural resources: water, land and energy. Thus, the objective of this study was, to verify the potential of the sensors to obtain the spectral data of an argisol with coffee cultivation. We used a set of data with a grid 51 soil samples on 39 hectares, with coffee cultivation (Yellow Catuaí, lineage IAC 62). Obtaining the color variables: hue, value, chroma, red, green and blue was determined by diffuse reflectance spectroscopy (DRS) and mobile near sensor (MNS). The results were related to soil attributes in the following cases: redMNS and clay: (R2 = 0.70, p <0.01), greenMNS and CTC (R2 = 0.73, p <0.01), redMNS and P (R2 = 0.64, p <0.05), blueMNS and EC (R2 = 0.96, p <0.01) and, greenMNS and Fe (R2 = 0.75, p <0.01). For the plant atributes, were: greenMNS and P17 (R2 = 0.70, p <0.01), defects with regressions ranging from 0.71 - 0.84 and, redMNS and productivity (R2 = 0.95, p <0.01). The results showed that the use of the color determined by red, green and blue allowed to identify soil attributes and areas with productive and qualitative potentials of coffee. For the study in the Midwest Paulista Plateau, a grid was realized in an area of 13 million hectares, totaling 600 samples in the depth of 0.00-0.20 meters. In these samples, the color variables were determined. Similarity was observed in the use of DRS and MNS methods. The DRS semivariograms were range for the hue, value, chroma, red, green and blue variables of 112 km, 216 km, 19 km, 207 km, 183 km and 158 km, respectively. With the use of MNS, the ranges were: 19 km, 226 km, 30 km, 160 km, 218 km and 75 km, respectively. The crossed semivariograms estimated the physical-chemical attributes (clay, cation exchange capacity - CTC, phosphorus - P and hematite-Hm). Crossed semivariograms explained the maximum correlation distance observed between the variables. The results of the MNS showed that the color in the red, green and blue systems can be used to predict the content of clay (77%), CTC (99%), P (90%) and Hm (70%). With the use of DRS, red, green and blue systems predict the content of clay (79%), CTC (94%), P (86%) and Hm (67%). These results show that color is a safe indication of soil attributes. As an alternative of low cost, the MNS has shown to be an efficient tool that allows the evaluation of these attributes.

O presente estudo teve como objetivo avaliar o teor de Cu, Cr, Ni, Zn, Pb, Mn, Al e Fe e sua partição geoquímica em sedimentos da bacia hidrográfica do rio dos Sinos, Sul do Brasil. Esta bacia apresenta elevada densidade populacional e industrial, especialmente dos setores de metalurgia, galvanoplastia, siderurgia, petroquímica e curtumes. As concentrações totais de metais, determinadas por espectroscopia de fluorescência de raios X, e as concentrações nas fases dos sedimentos, determinadas através da aplicação do esquema de extração sequencial BCR-701 proposto pelo Community Bureau of Reference, da Comunidade Europeia, utilizando ICP/OES, foram avaliadas nos sedimentos com granulometria <63 μm. Também foram considerados os teores extraídos através de uma solução de HCl 0,5 mol/L nesta fração dos sedimentos. Os resultados na região industrializada e com alta densidade populacional mostraram um aumento nas concentrações totais, para Cu, Cr, Zn e Pb, e nas frações potencialmente móveis para os metais Cu, Cr, Zn e Ni extraídos sequencialmente. O Cr destacou-se especialmente no arroio Portão, apresentando uma concentração total de 1286 mg/kg, devido à influência de curtumes localizados nesta região. Os sedimentos também foram caracterizados através de espectroscopia de reflectância, método rápido e não destrutivo de caracterização dos diferentes materiais, na região entre 400 e 2500 nm. Nos espectros de reflectância das amostras avaliadas observou-se que as principais feições espectrais são resultantes das absorções decorrentes da água, das ligações Al-OH e dos teores de ferro presentes. Ainda ficou evidenciada a influência da granulometria, dos teores de ferro e matéria orgânica sobre os espectros de reflectância resultantes. Não foi possível observar as feições espectrais decorrentes dos metais pesados, provavelmente devido às altas concentrações de ferro observadas nas amostras e devido ao fato destes metais apresentarem as suas feições espectrais nas mesmas regiões de comprimento de onda onde ocorrem as feições do ferro. / The present study aimed to evaluate the content of Cu, Cr, Ni, Zn, Pb, Mn, Al, and Fe as well as their geochemical partitioning in sediments of the Sinos river basin in southern Brazil. This basin has a high population density and a great number of industries, especially metallurgy, electroplating works, steel mills, petrochemicals, and tanneries. The total metal concentrations, determined by X-ray fluorescence, and the concentrations in the sediment phases, determined by applying the BCR-701 sequential extraction method proposed by the Community Bureau of Reference of the European Community, using ICP/OES, were evaluated in the sediment fraction <63 μm. The contents extracted by a solution of 0.5 mol/L HCl were also considered in this fraction. Results in the industrialized and densely populated region showed an increase in the total concentrations for Cu, Cr, Zn, and Pb, and in the potentially mobile fractions for the sequentially extracted metals Cu, Cr, Zn and Ni. Cr stood out particularly in the Portão stream, showing a total concentration of 1,286 mg/kg, due to the influence of tanneries in this region. The sediments were also characterized by reflectance spectroscopy, a rapid and nondestructive method for characterization of different materials, in the region between 400 and 2500 nm. The reflectance spectra of the samples showed that the main spectral features are the result of the absorptions due to water, Al-OH bonds and contents of iron. The influences of particle size, organic matter and mineral composition on the absorption features and on the reflection coefficient of the spectra were also evidenced. It was not possible to observe the spectral features resulting from metals, probably due to high iron concentrations observed in the samples, and because these metals make their spectral features on the same wavelength regions where there are the iron spectral features.

O presente estudo teve como objetivo avaliar o teor de Cu, Cr, Ni, Zn, Pb, Mn, Al e Fe e sua partição geoquímica em sedimentos da bacia hidrográfica do rio dos Sinos, Sul do Brasil. Esta bacia apresenta elevada densidade populacional e industrial, especialmente dos setores de metalurgia, galvanoplastia, siderurgia, petroquímica e curtumes. As concentrações totais de metais, determinadas por espectroscopia de fluorescência de raios X, e as concentrações nas fases dos sedimentos, determinadas através da aplicação do esquema de extração sequencial BCR-701 proposto pelo Community Bureau of Reference, da Comunidade Europeia, utilizando ICP/OES, foram avaliadas nos sedimentos com granulometria <63 μm. Também foram considerados os teores extraídos através de uma solução de HCl 0,5 mol/L nesta fração dos sedimentos. Os resultados na região industrializada e com alta densidade populacional mostraram um aumento nas concentrações totais, para Cu, Cr, Zn e Pb, e nas frações potencialmente móveis para os metais Cu, Cr, Zn e Ni extraídos sequencialmente. O Cr destacou-se especialmente no arroio Portão, apresentando uma concentração total de 1286 mg/kg, devido à influência de curtumes localizados nesta região. Os sedimentos também foram caracterizados através de espectroscopia de reflectância, método rápido e não destrutivo de caracterização dos diferentes materiais, na região entre 400 e 2500 nm. Nos espectros de reflectância das amostras avaliadas observou-se que as principais feições espectrais são resultantes das absorções decorrentes da água, das ligações Al-OH e dos teores de ferro presentes. Ainda ficou evidenciada a influência da granulometria, dos teores de ferro e matéria orgânica sobre os espectros de reflectância resultantes. Não foi possível observar as feições espectrais decorrentes dos metais pesados, provavelmente devido às altas concentrações de ferro observadas nas amostras e devido ao fato destes metais apresentarem as suas feições espectrais nas mesmas regiões de comprimento de onda onde ocorrem as feições do ferro. / The present study aimed to evaluate the content of Cu, Cr, Ni, Zn, Pb, Mn, Al, and Fe as well as their geochemical partitioning in sediments of the Sinos river basin in southern Brazil. This basin has a high population density and a great number of industries, especially metallurgy, electroplating works, steel mills, petrochemicals, and tanneries. The total metal concentrations, determined by X-ray fluorescence, and the concentrations in the sediment phases, determined by applying the BCR-701 sequential extraction method proposed by the Community Bureau of Reference of the European Community, using ICP/OES, were evaluated in the sediment fraction <63 μm. The contents extracted by a solution of 0.5 mol/L HCl were also considered in this fraction. Results in the industrialized and densely populated region showed an increase in the total concentrations for Cu, Cr, Zn, and Pb, and in the potentially mobile fractions for the sequentially extracted metals Cu, Cr, Zn and Ni. Cr stood out particularly in the Portão stream, showing a total concentration of 1,286 mg/kg, due to the influence of tanneries in this region. The sediments were also characterized by reflectance spectroscopy, a rapid and nondestructive method for characterization of different materials, in the region between 400 and 2500 nm. The reflectance spectra of the samples showed that the main spectral features are the result of the absorptions due to water, Al-OH bonds and contents of iron. The influences of particle size, organic matter and mineral composition on the absorption features and on the reflection coefficient of the spectra were also evidenced. It was not possible to observe the spectral features resulting from metals, probably due to high iron concentrations observed in the samples, and because these metals make their spectral features on the same wavelength regions where there are the iron spectral features.