Brief report: Using canonical correlation analysis to understand the dynamics between texture parameters and chemical attributes in the Ceará State soils of the western coast region in Brazil

Brief report: Using canonical correlation analysis to understand the dynamics between texture parameters and chemical attributes in the Ceará State soils of the western coast region in Brazil | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Brief report: Using canonical correlation analysis to understand the dynamics between texture parameters and chemical attributes in the Ceará State soils of the western coast region in Brazil Julian Andres Castillo Vargas, Francisca Gomes Ferreira, Joaquim Emanuel Fernandes Gondim This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9683449/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study used canonical correlation analysis (CCA) to evaluate the strength of the multivariate relationship between soil inorganic fractions and chemical attributes associated with the fertility complex in soils of the western coast of the State of Ceará, Brazil. A dataset comprising 96 observations was constructed from publicly available data from the Medium-Intensity Soil Reconnaissance Survey of the State of Ceará. The dataset included soil particle size fractions (sand, silt, and clay) and chemical attributes (Ca²⁺, Mg²⁺, K⁺, Na⁺, Al³⁺, and pH). The following multivariate relationship was explored using CCA with the CANCOR procedure of SAS: soil texture fractions versus soil chemical attributes, with a significance level of 0.05. The Wilk’s Lambda test revealed two significant (P < 0.01) canonical correlations. However, the redundancy analysis revealed that the first one had the greatest explanatory power (60.2%). Therefore, only the first canonical correlation was considered for practical interpretations. Canonical cross-loadings for the first canonical correlation indicated that an increase in silt (0.824) and clay (0.446), along with a concomitant decrease in sand (− 0.789) may result mainly in an increase in Ca²⁺ (0.634), Mg²⁺ (0.695), and K⁺ (0.484) in the coastal soils explored. Hence, the multivariate approach revealed the strength and specific relationships between soil texture characteristics and key exchangeable cations, information that may be useful for developing future mechanistic models to support decision-making in western coast soils. Multivariate relationship pedology soils attribute soil dynamic Figures Figure 1 Introduction Soils are formed by an association of factors and pedogenetic processes of general and specific order that impart morphological, physical, chemical, mineralogical, geochemical, and biological attributes (Amundson 2021 ). Thus, they promote functions and ecosystem services essential to the maintenance of life forms on the planet (Rodrigues et al. 2021 ; Vieillard et al. 2024). In addition, they represent a central basis for the primary food production chain (Kopittke et al. 2021; Xing et al. 2025 ). The description of soils in the field, as well as the characterization of attributes, is necessary in relation to the proposition of techniques for use, management, and conservation, aiming to enhance agricultural production and environmental sustainability (Coelho et al. 2021 ). In this regard, a recent collaborative study provided a detailed pedological mapping of the main soil orders in the Ceará state of Brazil, as well as physical, chemical, and mineralogical characteristics (FUNCEME 2024). This study is the result of the collective efforts of Brazilian soil scientists over decades of work, culminating in the creation of a digital pedological map across the entire state of Ceará, providing valuable and updated information on Brazilian soils since the last compiled surveys in 1970 (BRASIL 1973 ). The Ceará state has a high pedodiversity (natural variability of soil orders) due to several factors, including geology (parent material) and topography (BRASIL 1973 ; FUNCEME 2024). Thus, the exploration of soil data characterization using quantitative statistical techniques could provide valuable information for optimizing agricultural production in that region. In particular, the soils in the extreme west region are characterized by a predominantly sandy texture, fitting into the classes of quartzarenic neosols (Lima et al. 2002 ). These soils have desirable physical characteristics, such as infiltration and effective depth, but low natural fertility. This presents an interesting scenario for understanding the impacts of texture parameters on chemical characteristics in coastal soils, particularly in semiarid Brazilian soils. For this purpose, canonical correlation analysis (CCA) could be suitable, as this technique allows the understanding of the intricate patterns between two groups of variables (Hair et al. 2019 ), thereby considering the biological dynamics relationships between soil characteristics. Hence, this type of statistical approach is relevant in fieldwork, where the design of classical univariate statistics does not seem favorable, given that a greater number of external factors converge to the natural variability of soils in the field. Thus, using CCA, we explored the multivariate relationship of texture parameters (sand, silt, and clay) with major minerals (Ca, Mg, K, Na, and Al) and acidity (pH) of soils in the Extreme West Region of Ceará state in Brazil. Materials and methods Study area The study was conducted by acquiring publicly available, open-access data from the Medium-Intensity Soil Recognition Survey of the State of Ceará (2024), which was published as an e-book (FUNCEME 2024). The primary criterion adopted was the identification of soil classes representative of the extreme western coastal region, based on the rural territorial identity classification stated by IPECE (2016). To conduct the analysis, the first pedogenetic horizon of each soil class (A horizon) was considered due to its natural depth variability. Geologically, the study area is located within a sedimentary rock domain encompassing the geomorphological units of the Coastal Plain, Fluvial Plains, and pre-coastal Coastal Tablelands associated with the Barreiras Formation (Bizzi et al. 2003 ). The quantitative information used to construct the aforementioned dataset was obtained through the “Solos Ceará” application, which is available on the Google and Apple Store® platforms. Subsequently, soil analysis reports were manually retrieved from the different sampling points across municipalities within the study region. The edaphic analytical data were systematized into a structured database using Microsoft Excel® software, followed by a triple-check validation process conducted independently by different analysts to ensure data accuracy and consistency. As a result, a database comprising 96 records was built, including the following quantitative edaphic variables: calcium (Ca²⁺; cmol c kg⁻¹), magnesium (Mg²⁺; cmol c kg⁻¹), potassium (K⁺; cmol c kg⁻¹), sodium (Na⁺; cmol c kg⁻¹), aluminum (Al³⁺; cmol c kg⁻¹), pH in KCl (1:2.5), total sand (0.05–2 mm), silt (0.05–0.02 mm), and clay (< 0.002 mm). All laboratories responsible for reporting the soil analyses employed standardized methodologies approved by FUNCEME (2024). Physical and chemical characteristics of soil Granulometric analysis was performed using the pipette method, with sodium hexametaphosphate as the chemical dispersant and gentle mechanical agitation for 16 h on a Wagner-type shaker (50 rpm). The sand fraction (2–0.05 mm) was separated by sieving, clay (< 0.002 mm) by sedimentation, and the silt fraction (0.05–0.002 mm) by the difference between the two other fractions (Teixeira et al. 2017 ). Chemical analyses of air-dried soil (ADS) included determination of the potential of hydrogen (pH) in 1 M KCl. Macronutrients such as exchangeable calcium (Ca 2+ ) and magnesium (Mg 2+ ) were quantified using a potassium chloride (KCl) extractor, whereas phosphorus (P), sodium (Na + ), and potassium (K + ) were obtained using the Mehlich-1 extractor. These macrominerals were determined using atomic absorption spectrophotometry, according to the methodologies proposed by Teixeira et al. ( 2017 ). Exchangeable aluminum (Al³⁺) was extracted using 1 mol L⁻¹ potassium chloride (KCl) at pH 7.0 and quantified by volumetric titration with 0.025 mol L⁻¹ sodium hydroxide (NaOH), following the methodology described by Teixeira et al. ( 2017 ). Statistical analysis Multivariate statistical analysis was conducted using the PROC CANCOR of the SAS software (Statistical Analysis System, SAS Institute Inc., 2017) to explore the multivariate relationship of soil texture parameters with chemical components and pH. The CCA estimates the multivariate associations between two groups of variables, where each group is defined by a canonical variate composed of a linear combination of their original variables (Hair et al. 2019 ). For this study, two canonical variates were defined as follows: U (chemical components) = a 1 × Ca + a 2 × Mg + a 3 × Na + a 4 × K + a 5 × Al + a 6 × pH; V (texture parameters) = b 1 × sand + b 2 × silt + b 3 × clay in which, a and b are the canonical weights. The multivariate relationship explored between the canonical variates was as follows: U i = V j The smaller set of variables included in the canonical variates determines the number of canonical correlations (Hair et al. 2019 ). Accordingly, three canonical correlations were obtained in this study. The general mathematical expression for the canonical correlation analysis is as follows: $$\:{R}_{i}=\frac{Côv\:({U}_{i},{V}_{j})}{\sqrt{\left(Vâr\right({U}_{i}\left)Vâr\right({V}_{j})}}$$ The canonical loadings and cross-loadings were estimated for each canonical correlation (Hair et al. 2019 ). Canonical loading refers to the correlation between an original variable and its corresponding canonical variate, whereas canonical cross-loading denotes the correlation between an original variable from one set and the canonical variate derived from the opposite set. The significance of the evaluated canonical correlations was tested using Wilk’s Lambda test. In addition, redundancy analysis was conducted to explore the quantitative influence and predictability power of texture parameters on soil mineral components and acidity. To determine the relative contributions of the original variables to their respective canonical variates, the canonical loadings were transformed into relative quantities (%) using the following equation: Relative contribution = canonical loadings (in absolute value) / sum of canonical loadings within the canonical variate (in absolute value) × 100. The extent to which the canonical variate U i was explained by the corresponding canonical variate V j was evaluated through the coefficient of determination (R²) and redundancy analysis. A significance threshold of α = 0.05 was adopted. Results and Discussion Table 1 presents the descriptive statistics of the edaphic composition in the explored region. The mean sand, clay, and silt fractions were 787.6 ± 172.8, 104.9 ± 67.8, and 107.7 ± 128.7 g kg − 1 respectively. These values were consistent with granulometric parameters observed in soils with predominantly sandy texture, especially quartzipsamments ( Neossolos quartzarênicos ), which are classified as light soils and are predominant along the coast of the Ceará state. In these soils, the sand content generally exceeds 70%, whereas the clay content is less than 15%, and is often characterized by low colloidal activity (Espindola et al., 2025 ). The particle size distribution of these soils is consistent with other studies, such as Lima et al. ( 2002 ), who investigated soils with cohesive characteristics along the Ceará coast, specifically Acrisols ( Argissolos ). Knowledge of the particle size distribution of these soils is a key factor, as it provides essential information for practical applications in land management and agricultural crop management according to land use (Schadosin et al. 2023 ). The mean values of Ca²⁺, Mg²⁺, Na⁺, K⁺, and Al³⁺ concentrations were 1.913 ± 2.770, 1.169 ± 2.225, 0.077 ± 0.232, 0.114 ± 0.123, and 0.411 ± 0.459 cmol c kg⁻¹, respectively. The mean soil pH was 4.25 ± 0.63, indicating acidic conditions. From a soil fertility perspective, the exchange complex exhibits limitations due to its low capacity to supply nutrients to plants, as evidenced by the strongly acidic soil pH (Neina 2023). If these soils are to be cultivated, practical measures must be implemented. Fertilization and soil amendment decisions should be based on soil analysis, and agricultural use must always be aligned with land suitability to prevent degradation (Dertli et al. 2024 ). The multivariate relationship of texture parameters with mineral components and pH in soils (i.e., U = V) revealed two significant canonical correlations (R 1 = 0.825; R 2 = 0.642; P < 0.01; Table 2 ). However, redundancy analysis (Table 2 ) demonstrated that the first canonical function accounted for a substantial proportion of significant shared variance between the two sets of variables, with the chemical variables explaining approximately 60.2% of their variance from the texture canonical variate. In contrast, the second canonical function contributed only marginally to the explained variance of 4.6% for chemical variables. These results indicate that the first canonical function captured a strong and biologically meaningful association between soil texture and chemical composition, suggesting that approximately 60% of the variability in mineral availability and soil acidity could be reliably predicted from the distribution of sand, silt, and clay fractions (Hair et al., 2019 ). Hence, only the first canonical correlation was explored for practical purposes. At this point, it is important to mention that this strong predictive relationship highlights the potential use of textural attributes in the development of mechanistic models aimed at estimating nutrient dynamics in coastal environments, particularly in regions where detailed chemical analyses are limited or costly. This information could be useful for suggesting fertilization strategies under these conditions. Canonical weights quantify the observed variables’ relative contribution to their respective canonical variates. Nevertheless, these coefficients are often characterized by substantial instability because they are highly susceptible to multicollinearity among predictors, which can compromise the precision and interpretability of the estimated variable contributions (Diel et al., 2020 ). Considering these methodological constraints, this study restricted the analysis to the examination of canonical loadings and cross-loadings associated with the first multivariate canonical function. However, canonical weights could be useful in future prediction models derived from multi-study designs (meta-analysis) to understand the impacts of texture on chemical constituents in soils (Iatrou et al., 2025 ). Hence, we decided to keep this information. Canonical loadings can be used to examine the multivariate correlations between the original variables and their corresponding canonical variates (Jaenanda et al. 2020). Moreover, they allow the assessment of each original variable’s relative contribution to its respective canonical variate. Based on this premise, the canonical loadings are presented in Table 3 and expressed in terms of the relative contributions (%) within their respective canonical variates (Fig. 1 ). These values revealed that U (chemical components) was mainly described by Ca 2+ (25.2%), Mg 2+ (27.6%), and K + (19.2%), with relatively lower contributions for Na + (14.7%), Al 3+ (12.7%), and pH - KCl (0.59%). Hence, the aforementioned results suggest that Ca 2+ , Mg 2+ , and K + are the main chemical components affecting the canonical variate U 1 . These results indicate that the region’s geological and mineralogical variability accounts for the predominance of these potential nutrients in the canonical loadings. The presence of primary minerals, such as feldspars, amphiboles, and micas, undergoes weathering over time, contributing to the release of Ca 2+, Mg 2+ and K + (Wilson, 2004 ) in less weathered soils, such as Quartzipsamments ( Neossolos Quartzarênicos ), as well as in more developed soils, such as Acrisols ( Argissolos ) (FUNCEME 2024). Nevertheless, according to Donagema et al. (2011) and Lima et al. ( 2002 ), coastal soils do not exhibit an exchange complex with an adequate number of base cations to readily supply agricultural crops due to their low pH and low cation exchange capacity (CEC). Regarding canonical variate V 1 (Table 3 ), data revealed that texture parameters had the following decreasing pattern (Fig. 1 ): silt (40.0%) > sand (38.3%) > clay (21.7%), suggesting that silt and sand were the most representative variables for this canonical variate. This confirms the lower contribution of inorganic fractions to soil fertility, reflecting the general particle size distribution observed in the region. The inorganic fractions, particularly sand and silt, have low nutrient reserves (Sarkar et al., 2019 ; Thabit et al., 2023 ). Canonical cross-loading revealed the multivariate association between the two explored canonical variates. In other words, the multivariate correlation between chemical constituents and texture parameters is explored. Table 3 and the major contributions of the original variables within the respective canonical variates (Fig. 1 ) suggest that an increase in silt (0.824) and clay (0.446) with a concomitant decrease in sand (-0.789) may result in an increase in Ca 2+ (0.634), Mg 2+ (0.695), and K + (0.484). This pattern is consistent because an increase in the silt fraction is accompanied by an increase in the specific surface area of the particles, given that silt particles (0.05–0.002 mm) are smaller than sand particles (2.00–0.05 mm) (Provens et al. 2022 ). In this context, electrical charge phenomena are more pronounced, enhancing the retention of a greater number of ions such as Ca²⁺, Mg²⁺, and K⁺. However, it should be noted that in the clay fraction (< 0.002 mm), these phenomena are further intensified, mainly due to the type of clay mineral present (Yu et al. 2025 ). Taking together, the results reported herein, revealed the strength of the relationship between soil texture (sand, silt, and clay) and the potential ions present in the soil (Ca²⁺, Mg²⁺, and K⁺) provides valuable insight into the electrical charge phenomena of coastal soils (Santos et al. 2026 ). In addition, the high redundancy observed (> 60%) suggest that textural attributes may serve as reliable predictors of soil chemical behavior, and this information could be incorporated into the development of prediction or mechanistic models for a better understanding of the mineral’s dynamics in this kind of soil, which remains insufficiently understood. As a result, fertilization practices could be recommended based on an understanding of soil charge characteristics and the types of ions present in association with soil texture (Besen et al. 2021 ). This is of practical importance, as soil components are key attributes governing water infiltration, movement, and retention, thereby directly influencing irrigation system management and selection (Shahadha and Wendroth 2025 ). Conclusion Canonical correlation and redundancy analyses revealed a strong multivariate relationship between soil texture and chemical composition, with more than 60% of the variability in mineral composition and soil acidity explained by textural attributes through the first canonical function. Increases in silt and clay fractions, combined with reductions in sand content, were associated with greater retention of Ca²⁺, Mg²⁺, and K⁺, highlighting the importance of particle-size distribution in regulating ion retention and electrical charge dynamics in coastal soils. These findings demonstrate the strong predictive potential of textural parameters and support their application in the development of predictive or mechanistic models aimed at estimating nutrient dynamics, soil acidity, and fertility patterns in coastal environments, particularly where detailed chemical analyses are limited or costly. Declarations Competing interests None Funding No external funding was received for this study. Author Contribution J.A.C.V.: Conceptualization, methodology, software, validation, formal analysis, investigation, data curation, writing—original draft preparation, and writing—review and editing. F.T.G.F.: Data curation and methodology; J.E.F.G.: Conceptualization, methodology, investigation, writing—original draft preparation, and writing—review and editing Acknowledgement The authors would like to thank the Universidade Estadual Vale do Acaraú (UVA) for providing institutional support in the form of salaries during the development of this research. The authors also acknowledge the Ceará State Foundation for the Support of Scientific and Technological Development (FUNCAP) for providing a scholarship (grant no. 2347). Data Availability The dataset used in this study is available upon reasonable request from the corresponding author. References Amundson R. Factors of soil formation in the 21st century. Geoderma. 2021;391:114960. https://doi.org/10.1016/j.geoderma.2021.114960 . Besen MR, Ribeiro RH, Esper Neto M, Minato EA, Coneglian CF, Kachinski WD, Tormena CA, Inoue TT, Batista MA. Lime and phosphogypsum application management: changes in soil acidity, sulfur availability and crop yield. Rev Bras Cienc Solo. 2021;45:e0200135. https://doi.org/10.36783/18069657rbcs20200135 . Bizzi LA, Schobbenhaus C, Gonçalves JH, Baars FJ, Delgado IM, Abram MB, Leão Neto R, Matos GMM, Santos JOS, editors. Geologia, tectônica e recursos minerais do Brasil: texto, mapas e SIG. Brasília: CPRM – Serviço Geológico do Brasil; 2003. p. 673. BRASIL. Levantamento exploratório - reconhecimento de solos do Estado do Ceará. Recife: DPP-MA/DNPEA-SUDENE; 1973. (Boletim Técnico nº 28). Coelho FF, Giasson E, Campos AR, Tiecher T, Costa JJF, Coblinski JA. Digital soil class mapping in Brazil: a systematic review. Scientia Agricola. 2021;78(5):e20190227. https://doi.org/10.1590/1678-992X-2019-0227 . Dertli HI, Hayes DB, Zorn TG. Effects of multicollinearity and data granularity on regression models of stream temperature. J Hydrol. 2024;639(1):131572. https://doi.org/10.1016/j.jhydrol.2024.131572 . Diel MI, Lúcio ADC, Lambrecht DM, Pinheiro MVM, Sari BG, Olivoto T, Valera OVS, Melo PJ, Tartaglia FL, Tischler AL, Schmidt D. Canonical correlations in agricultural research: Method of interpretation used leads to greater reliability of results. Int J Innov Educ Res. 2020;8:171–81. https://doi.org/10.31686/ijier.vol8.iss7.2464 . Donagemma GK, Freitas PL, Balieiro FC, Fontana A, Spera ST, Lumbreras JF, Viana JHM, Araujo Filho JC, Santos HG, Albuquerque MR, Carvalho Junior W, Calderano SB. Caracterização, potencial agrícola e perspectivas de manejo de solos leves no Brasil. Pesquisa Agropecuária Brasileira. 2016;51(9):1003–20. https://doi.org/10.1590/S0100-204X2016000900001 . Espindola C, Pinheiro M, Nakashima M. Proposal for taxonomic revision of thick and highly weathered sandy soils in the Brazilian soil classification system. Discover Soil. 2025;2:124. https://doi.org/10.1007/s44378-025-00151-3 . FUNCEME – Fundação Cearense de Meteorologia e Recursos Hídricos. (2024) Levantamento de reconhecimento de média intensidade dos solos do Estado do Ceará [electronic book]. Fortaleza, CE: Coletivo Duas Catitas, ePub. ISBN 978-65-983255-0-3. Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate data analysis. 8th ed. Boston: Cengage; 2019. p. 813. IPECE – Instituto de Pesquisa e Estratégia Econômica do Ceará. (2016) Territórios rurais de identidade. In: Atlas do Ceará. Fortaleza: IPECE. Available at: http://www2.ipece.ce.gov.br/atlas/capitulo1/11/pdf/Territorios_rurais_identidade.pdf Jaenanda M, Miranda P, Pardo L, Zografos K. An approach to canonical correlation analysis based on Rényi’s pseudodistances. Entropy. 2024;25(5):713. https://doi.org/10.3390/e25050713 . Kopittke PM, Menzies NW, Wang P, McKenna BA, Lombi E. Soil and the intensification of agriculture for global food security. Environ Int. 2019;138:105078. https://doi.org/10.1016/j.envint.2019.105078 . Iatrou M, Tziachris P, Bilias F, Kekelis P, Pavlakis C, Theofilidou A, Papadopoulos I, Strouthopoulos G, Giannopoulos G, Arampatzis D, et al. Data-Driven and Mechanistic Soil Modeling for Precision Fertilization Management in Cotton. Nitrogen. 2025;6(2):29. https://doi.org/10.3390/nitrogen6020029 . Lima ACC, Oliveira FNS, Aquino ARL. (2002) Limitações do uso dos solos do Estado do Ceará por suscetibilidade à erosão. Fortaleza: Embrapa Agroindústria Tropical, 19 p. (Documentos, 54). Mota LHSO, Valladares GS. Vulnerabilidade à degradação dos solos da Bacia do Acaraú, Ceará. Revista Ciência Agronômica. 2011;42(1):39–50. https://doi.org/10.1590/S1806-66902011000100006 . Neina D. (2019) The role of soil pH in plant nutrition and soil remediation. Applied and Environmental Soil Science 2019: 5794869. https://doi.org/10.1155/2019/5794869 Provens ECP, Arrarás IR, Santos JMD. Estimating textural fractions of the USDA using those of the International System: a quantile approach. Geoderma. 2022;416:115782. https://doi.org/10.1016/j.geoderma.2022.115783 . Rodrigues AF, Latawiec AE, Reid BJ, Solórzano A, Schuler AE, Lacerda C, Fidalgo ECC, Scarano FR, Tubenchlak F, Pena I, Vicente-Vicente JL, Korys KA, Cooper M, Fernandes NF, Prado RB, Maioli V, Dib V, Teixeira WG. Systematic review of soil ecosystem services in tropical regions. R Soc Open Sci. 2021;8(3):201584. https://doi.org/10.1098/rsos.201584 . Santos AB, Silva AC, Tabarelli M, Barros MF, Franca EJ, Lins SRM, Vanderlei RS. Soil attributes are key drivers of calcium, potassium, and magnesium concentrations and stocks in a tropical dry forest. J Arid Environ. 2026;235:105604. https://doi.org/10.1016/j.jaridenv.2026.105604 . Sarkar I, Hanif M, Khan MZ. Soil organic fractions in cultivated and uncultivated soils of coastal area in Bangladesh. J Agricultural Chem Environ. 2019;8(3):129–44. https://doi.org/10.4236/jacen.2019.83011 . SAS Institute. Base SAS 9.4 procedures guide: Statistical procedures. SAS institute; 2017. Schadosin J, Saab SC, Brinatti AM, Pires LF. Quantitative and qualitative characterization of the granulometry of an Inceptisol. Quím Nova. 2023;46(4):329–35. https://doi.org/10.21577/0100-4042.20230018 . Shahadha SS, Wendroth O. Modeling the effect of tillage and irrigation management on water and barley productivity for different soil textures. Soil Tillage Res. 2025;250:106505. https://doi.org/10.1016/j.still.2025.106505 . Teixeira PC, Donagemma GK, Fontana A, Teixeira WG, editors. (2017) Manual de métodos de análise de solo. 3rd edn. rev. e ampl. Brasília, DF: Embrapa, 573 p. Thabit FN, El-Shater AH, Soliman W. Role of silt and clay fractions in organic carbon and nitrogen stabilization in soils of some old fruit orchards in the Nile floodplain, Sohag Governorate, Egypt. J Soil Sci Plant Nutr. 2023;23:2525–44. https://doi.org/10.1007/s42729-023-01209-3 . Vieilard C, Vidal-Beaudet L, Dagois R, Lothode M, Vadepied F, Gontier M, Schwartz C, Ouvrard S. Impacts of soil de-sealing practices on urban land-uses, soil functions and ecosystem services in French cities. Geoderma Reg. 2024;e00854. https://doi.org/10.1016/j.geodrs.2024.e00854 . Wilson MJ. Weathering of the primary rock-forming minerals: processes, products and rates. Clay Miner. 2004;39(3):233–66. https://doi.org/10.1180/0009855043930133 . Xing Y, Wang X, Mustafa A. Exploring the link between soil health and crop productivity. Ecotoxicol Environ Saf. 2025;289:117703. https://doi.org/10.1016/j.ecoenv.2025.117703 . Yu L, He A, Li H. Soil particle size and Hofmeister effects on soil colloidal aggregation. Appl Clay Sci. 2025;278:108018. https://doi.org/10.1016/j.clay.2025.108018 . Tables Tables are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Vargasetal.2026tables.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9683449","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":638510283,"identity":"42865256-bc77-4c8d-93b4-c9c41d3e9568","order_by":0,"name":"Julian Andres Castillo Vargas","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIiWNgGAWjYBACxgbmhgMIbgVcGJ8WRmQtZ+CC+O1BYrcRoYV5RmLjgZ87GBL7Zx+/+PHnvMNARvPxB4w77uG2Y0Ziw8HeMwyJM87lFEvzbjucOOPOscQGxjPFeLUc4G1jMGY4w5MgzbjtsDHDjRzDBsa2BPy2/AVqkT/Dk/zz55zDxvI38j8S1HIYaIucwRn2YxK8DYflDG7kMOLX0vOw4bAsUIvhGR42a55j6XKGN9IMZySewa3FsD358Me3bQw8cmfYH9/8UWPNI3cj+cGHjzvwaGkAU/+BmMcAIYxbAwODPILJ/gCPulEwCkbBKBjJAAD8lF4rkN8+nAAAAABJRU5ErkJggg==","orcid":"","institution":"State University of Vale do Acaraú","correspondingAuthor":true,"prefix":"","firstName":"Julian","middleName":"Andres Castillo","lastName":"Vargas","suffix":""},{"id":638510284,"identity":"f779a272-b4c2-4b49-a713-91446b7fe12e","order_by":1,"name":"Francisca Gomes Ferreira","email":"","orcid":"","institution":"State University of Vale do Acaraú","correspondingAuthor":false,"prefix":"","firstName":"Francisca","middleName":"Gomes","lastName":"Ferreira","suffix":""},{"id":638510285,"identity":"4492c394-d313-4ce4-b4bc-ee217dc88a24","order_by":2,"name":"Joaquim Emanuel Fernandes Gondim","email":"","orcid":"","institution":"State University of Vale do Acaraú","correspondingAuthor":false,"prefix":"","firstName":"Joaquim","middleName":"Emanuel Fernandes","lastName":"Gondim","suffix":""}],"badges":[],"createdAt":"2026-05-11 18:54:48","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9683449/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9683449/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109165989,"identity":"8d1807ba-8c30-4b83-8996-02297dd43124","added_by":"auto","created_at":"2026-05-13 08:16:00","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":84131,"visible":true,"origin":"","legend":"\u003cp\u003eContributions of chemical components (canonical variate U) and texture parameters (canonical variate V) to their respective canonical variates in the first significant canonical correlation. R = canonical correlation (i.e., multivariate correlation between U and V canonical variates). The Wilks’ Lambda test was used to test the significance of the canonical correlation, and the confidence level adopted was P \u0026lt; 0.05. Values next to arrows correspond to variable contributions (expressed as canonical loading in absolute value/sum of canonical loadings in absolute value for all variables evaluated × 100) to the respective canonical variate.\u003c/p\u003e","description":"","filename":"Screenshot20260513133953.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9683449/v1/abad76ba18d3ba11e5ca4d9e.jpg"},{"id":109205204,"identity":"251d887a-8e9c-449a-89e9-0a28464e2444","added_by":"auto","created_at":"2026-05-13 15:03:45","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":268390,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9683449/v1/f8013f87-f8c7-4a13-8e19-bdd4a7bfbd22.pdf"},{"id":109165924,"identity":"72d56241-932a-4f9a-a2d3-19cdd5c666a8","added_by":"auto","created_at":"2026-05-13 08:15:46","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":22599,"visible":true,"origin":"","legend":"","description":"","filename":"Vargasetal.2026tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-9683449/v1/8bb62e0df0dca3013378481f.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Brief report: Using canonical correlation analysis to understand the dynamics between texture parameters and chemical attributes in the Ceará State soils of the western coast region in Brazil","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSoils are formed by an association of factors and pedogenetic processes of general and specific order that impart morphological, physical, chemical, mineralogical, geochemical, and biological attributes (Amundson \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Thus, they promote functions and ecosystem services essential to the maintenance of life forms on the planet (Rodrigues et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Vieillard et al. 2024). In addition, they represent a central basis for the primary food production chain (Kopittke et al. 2021; Xing et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe description of soils in the field, as well as the characterization of attributes, is necessary in relation to the proposition of techniques for use, management, and conservation, aiming to enhance agricultural production and environmental sustainability (Coelho et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In this regard, a recent collaborative study provided a detailed pedological mapping of the main soil orders in the Cear\u0026aacute; state of Brazil, as well as physical, chemical, and mineralogical characteristics (FUNCEME 2024). This study is the result of the collective efforts of Brazilian soil scientists over decades of work, culminating in the creation of a digital pedological map across the entire state of Cear\u0026aacute;, providing valuable and updated information on Brazilian soils since the last compiled surveys in 1970 (BRASIL \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1973\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe Cear\u0026aacute; state has a high pedodiversity (natural variability of soil orders) due to several factors, including geology (parent material) and topography (BRASIL \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1973\u003c/span\u003e; FUNCEME 2024). Thus, the exploration of soil data characterization using quantitative statistical techniques could provide valuable information for optimizing agricultural production in that region. In particular, the soils in the extreme west region are characterized by a predominantly sandy texture, fitting into the classes of quartzarenic neosols (Lima et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). These soils have desirable physical characteristics, such as infiltration and effective depth, but low natural fertility. This presents an interesting scenario for understanding the impacts of texture parameters on chemical characteristics in coastal soils, particularly in semiarid Brazilian soils.\u003c/p\u003e \u003cp\u003eFor this purpose, canonical correlation analysis (CCA) could be suitable, as this technique allows the understanding of the intricate patterns between two groups of variables (Hair et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), thereby considering the biological dynamics relationships between soil characteristics. Hence, this type of statistical approach is relevant in fieldwork, where the design of classical univariate statistics does not seem favorable, given that a greater number of external factors converge to the natural variability of soils in the field. Thus, using CCA, we explored the multivariate relationship of texture parameters (sand, silt, and clay) with major minerals (Ca, Mg, K, Na, and Al) and acidity (pH) of soils in the Extreme West Region of Cear\u0026aacute; state in Brazil.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy area\u003c/h2\u003e \u003cp\u003eThe study was conducted by acquiring publicly available, open-access data from the Medium-Intensity Soil Recognition Survey of the State of Cear\u0026aacute; (2024), which was published as an e-book (FUNCEME 2024). The primary criterion adopted was the identification of soil classes representative of the extreme western coastal region, based on the rural territorial identity classification stated by IPECE (2016). To conduct the analysis, the first pedogenetic horizon of each soil class (A horizon) was considered due to its natural depth variability. Geologically, the study area is located within a sedimentary rock domain encompassing the geomorphological units of the Coastal Plain, Fluvial Plains, and pre-coastal Coastal Tablelands associated with the Barreiras Formation (Bizzi et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2003\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe quantitative information used to construct the aforementioned dataset was obtained through the \u0026ldquo;Solos Cear\u0026aacute;\u0026rdquo; application, which is available on the Google and Apple Store\u0026reg; platforms. Subsequently, soil analysis reports were manually retrieved from the different sampling points across municipalities within the study region. The edaphic analytical data were systematized into a structured database using Microsoft Excel\u0026reg; software, followed by a triple-check validation process conducted independently by different analysts to ensure data accuracy and consistency.\u003c/p\u003e \u003cp\u003eAs a result, a database comprising 96 records was built, including the following quantitative edaphic variables: calcium (Ca\u0026sup2;⁺; cmol\u003csub\u003ec\u003c/sub\u003e kg⁻\u0026sup1;), magnesium (Mg\u0026sup2;⁺; cmol\u003csub\u003ec\u003c/sub\u003e kg⁻\u0026sup1;), potassium (K⁺; cmol\u003csub\u003ec\u003c/sub\u003e kg⁻\u0026sup1;), sodium (Na⁺; cmol\u003csub\u003ec\u003c/sub\u003e kg⁻\u0026sup1;), aluminum (Al\u0026sup3;⁺; cmol\u003csub\u003ec\u003c/sub\u003e kg⁻\u0026sup1;), pH in KCl (1:2.5), total sand (0.05\u0026ndash;2 mm), silt (0.05\u0026ndash;0.02 mm), and clay (\u0026lt;\u0026thinsp;0.002 mm). All laboratories responsible for reporting the soil analyses employed standardized methodologies approved by FUNCEME (2024).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePhysical and chemical characteristics of soil\u003c/h3\u003e\n\u003cp\u003eGranulometric analysis was performed using the pipette method, with sodium hexametaphosphate as the chemical dispersant and gentle mechanical agitation for 16 h on a Wagner-type shaker (50 rpm). The sand fraction (2\u0026ndash;0.05 mm) was separated by sieving, clay (\u0026lt;\u0026thinsp;0.002 mm) by sedimentation, and the silt fraction (0.05\u0026ndash;0.002 mm) by the difference between the two other fractions (Teixeira et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eChemical analyses of air-dried soil (ADS) included determination of the potential of hydrogen (pH) in 1 M KCl. Macronutrients such as exchangeable calcium (Ca\u003csup\u003e2+\u003c/sup\u003e) and magnesium (Mg\u003csup\u003e2+\u003c/sup\u003e) were quantified using a potassium chloride (KCl) extractor, whereas phosphorus (P), sodium (Na\u003csup\u003e+\u003c/sup\u003e), and potassium (K\u003csup\u003e+\u003c/sup\u003e) were obtained using the Mehlich-1 extractor. These macrominerals were determined using atomic absorption spectrophotometry, according to the methodologies proposed by Teixeira et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Exchangeable aluminum (Al\u0026sup3;⁺) was extracted using 1 mol L⁻\u0026sup1; potassium chloride (KCl) at pH 7.0 and quantified by volumetric titration with 0.025 mol L⁻\u0026sup1; sodium hydroxide (NaOH), following the methodology described by Teixeira et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eMultivariate statistical analysis was conducted using the PROC CANCOR of the SAS software (Statistical Analysis System, SAS Institute Inc., 2017) to explore the multivariate relationship of soil texture parameters with chemical components and pH.\u003c/p\u003e \u003cp\u003eThe CCA estimates the multivariate associations between two groups of variables, where each group is defined by a canonical variate composed of a linear combination of their original variables (Hair et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). For this study, two canonical variates were defined as follows:\u003c/p\u003e \u003cp\u003eU (chemical components) = a\u003csub\u003e1\u003c/sub\u003e \u0026times; Ca\u0026thinsp;+\u0026thinsp;a\u003csub\u003e2\u003c/sub\u003e \u0026times; Mg\u0026thinsp;+\u0026thinsp;a\u003csub\u003e3\u003c/sub\u003e \u0026times; Na\u0026thinsp;+\u0026thinsp;a\u003csub\u003e4\u003c/sub\u003e \u0026times; K\u0026thinsp;+\u0026thinsp;a\u003csub\u003e5\u003c/sub\u003e \u0026times; Al\u0026thinsp;+\u0026thinsp;a\u003csub\u003e6\u003c/sub\u003e \u0026times; pH;\u003c/p\u003e \u003cp\u003eV (texture parameters) = b\u003csub\u003e1\u003c/sub\u003e \u0026times; sand\u0026thinsp;+\u0026thinsp;b\u003csub\u003e2\u003c/sub\u003e \u0026times; silt\u0026thinsp;+\u0026thinsp;b\u003csub\u003e3\u003c/sub\u003e \u0026times; clay in which, a and b are the canonical weights.\u003c/p\u003e \u003cp\u003eThe multivariate relationship explored between the canonical variates was as follows:\u003c/p\u003e \u003cp\u003eU\u003csub\u003ei\u003c/sub\u003e = V\u003csub\u003ej\u003c/sub\u003e\u003c/p\u003e \u003cp\u003eThe smaller set of variables included in the canonical variates determines the number of canonical correlations (Hair et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Accordingly, three canonical correlations were obtained in this study. The general mathematical expression for the canonical correlation analysis is as follows:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{R}_{i}=\\frac{C\u0026ocirc;v\\:({U}_{i},{V}_{j})}{\\sqrt{\\left(V\u0026acirc;r\\right({U}_{i}\\left)V\u0026acirc;r\\right({V}_{j})}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThe canonical loadings and cross-loadings were estimated for each canonical correlation (Hair et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Canonical loading refers to the correlation between an original variable and its corresponding canonical variate, whereas canonical cross-loading denotes the correlation between an original variable from one set and the canonical variate derived from the opposite set.\u003c/p\u003e \u003cp\u003eThe significance of the evaluated canonical correlations was tested using Wilk\u0026rsquo;s Lambda test. In addition, redundancy analysis was conducted to explore the quantitative influence and predictability power of texture parameters on soil mineral components and acidity.\u003c/p\u003e \u003cp\u003eTo determine the relative contributions of the original variables to their respective canonical variates, the canonical loadings were transformed into relative quantities (%) using the following equation:\u003c/p\u003e \u003cp\u003eRelative contribution\u0026thinsp;=\u0026thinsp;canonical loadings (in absolute value) / sum of canonical loadings within the canonical variate (in absolute value) \u0026times; 100.\u003c/p\u003e \u003cp\u003eThe extent to which the canonical variate U\u003csub\u003ei\u003c/sub\u003e was explained by the corresponding canonical variate V\u003csub\u003ej\u003c/sub\u003e was evaluated through the coefficient of determination (R\u0026sup2;) and redundancy analysis. A significance threshold of α\u0026thinsp;=\u0026thinsp;0.05 was adopted.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results and Discussion","content":"\u003cp\u003eTable \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the descriptive statistics of the edaphic composition in the explored region. The mean sand, clay, and silt fractions were 787.6\u0026thinsp;\u0026plusmn;\u0026thinsp;172.8, 104.9\u0026thinsp;\u0026plusmn;\u0026thinsp;67.8, and 107.7\u0026thinsp;\u0026plusmn;\u0026thinsp;128.7 g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e respectively. These values were consistent with granulometric parameters observed in soils with predominantly sandy texture, especially quartzipsamments (\u003cem\u003eNeossolos quartzar\u0026ecirc;nicos\u003c/em\u003e), which are classified as light soils and are predominant along the coast of the Cear\u0026aacute; state. In these soils, the sand content generally exceeds 70%, whereas the clay content is less than 15%, and is often characterized by low colloidal activity (Espindola et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The particle size distribution of these soils is consistent with other studies, such as Lima et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2002\u003c/span\u003e), who investigated soils with cohesive characteristics along the Cear\u0026aacute; coast, specifically Acrisols (\u003cem\u003eArgissolos\u003c/em\u003e). Knowledge of the particle size distribution of these soils is a key factor, as it provides essential information for practical applications in land management and agricultural crop management according to land use (Schadosin et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eThe mean values of Ca\u0026sup2;⁺, Mg\u0026sup2;⁺, Na⁺, K⁺, and Al\u0026sup3;⁺ concentrations were 1.913\u0026thinsp;\u0026plusmn;\u0026thinsp;2.770, 1.169\u0026thinsp;\u0026plusmn;\u0026thinsp;2.225, 0.077\u0026thinsp;\u0026plusmn;\u0026thinsp;0.232, 0.114\u0026thinsp;\u0026plusmn;\u0026thinsp;0.123, and 0.411\u0026thinsp;\u0026plusmn;\u0026thinsp;0.459 cmol\u003csub\u003ec\u003c/sub\u003e kg⁻\u0026sup1;, respectively. The mean soil pH was 4.25\u0026thinsp;\u0026plusmn;\u0026thinsp;0.63, indicating acidic conditions. From a soil fertility perspective, the exchange complex exhibits limitations due to its low capacity to supply nutrients to plants, as evidenced by the strongly acidic soil pH (Neina 2023). If these soils are to be cultivated, practical measures must be implemented. Fertilization and soil amendment decisions should be based on soil analysis, and agricultural use must always be aligned with land suitability to prevent degradation (Dertli et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eThe multivariate relationship of texture parameters with mineral components and pH in soils (i.e., U\u0026thinsp;=\u0026thinsp;V) revealed two significant canonical correlations (R\u003csub\u003e1\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.825; R\u003csub\u003e2\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.642; P\u0026thinsp;\u0026lt;\u0026thinsp;0.01; Table \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). However, redundancy analysis (Table \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) demonstrated that the first canonical function accounted for a substantial proportion of significant shared variance between the two sets of variables, with the chemical variables explaining approximately 60.2% of their variance from the texture canonical variate. In contrast, the second canonical function contributed only marginally to the explained variance of 4.6% for chemical variables. These results indicate that the first canonical function captured a strong and biologically meaningful association between soil texture and chemical composition, suggesting that approximately 60% of the variability in mineral availability and soil acidity could be reliably predicted from the distribution of sand, silt, and clay fractions (Hair et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Hence, only the first canonical correlation was explored for practical purposes. At this point, it is important to mention that this strong predictive relationship highlights the potential use of textural attributes in the development of mechanistic models aimed at estimating nutrient dynamics in coastal environments, particularly in regions where detailed chemical analyses are limited or costly. This information could be useful for suggesting fertilization strategies under these conditions.\u003c/p\u003e\n\u003cp\u003eCanonical weights quantify the observed variables\u0026rsquo; relative contribution to their respective canonical variates. Nevertheless, these coefficients are often characterized by substantial instability because they are highly susceptible to multicollinearity among predictors, which can compromise the precision and interpretability of the estimated variable contributions (Diel et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Considering these methodological constraints, this study restricted the analysis to the examination of canonical loadings and cross-loadings associated with the first multivariate canonical function. However, canonical weights could be useful in future prediction models derived from multi-study designs (meta-analysis) to understand the impacts of texture on chemical constituents in soils (Iatrou et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Hence, we decided to keep this information.\u003c/p\u003e\n\u003cp\u003eCanonical loadings can be used to examine the multivariate correlations between the original variables and their corresponding canonical variates (Jaenanda et al. 2020). Moreover, they allow the assessment of each original variable\u0026rsquo;s relative contribution to its respective canonical variate. Based on this premise, the canonical loadings are presented in Table \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and expressed in terms of the relative contributions (%) within their respective canonical variates (Fig. \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). These values revealed that U (chemical components) was mainly described by Ca\u003csup\u003e2+\u003c/sup\u003e (25.2%), Mg\u003csup\u003e2+\u003c/sup\u003e (27.6%), and K\u003csup\u003e+\u003c/sup\u003e (19.2%), with relatively lower contributions for Na\u003csup\u003e+\u003c/sup\u003e (14.7%), Al\u003csup\u003e3+\u003c/sup\u003e (12.7%), and pH - KCl (0.59%). Hence, the aforementioned results suggest that Ca\u003csup\u003e2+\u003c/sup\u003e, Mg\u003csup\u003e2+\u003c/sup\u003e, and K\u003csup\u003e+\u003c/sup\u003e are the main chemical components affecting the canonical variate U\u003csub\u003e1\u003c/sub\u003e.\u003c/p\u003e\n\u003cp\u003eThese results indicate that the region\u0026rsquo;s geological and mineralogical variability accounts for the predominance of these potential nutrients in the canonical loadings. The presence of primary minerals, such as feldspars, amphiboles, and micas, undergoes weathering over time, contributing to the release of Ca\u003csup\u003e2+,\u003c/sup\u003e Mg\u003csup\u003e2+\u003c/sup\u003e and K\u003csup\u003e+\u003c/sup\u003e (Wilson, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2004\u003c/span\u003e) in less weathered soils, such as Quartzipsamments (\u003cem\u003eNeossolos Quartzar\u0026ecirc;nicos\u003c/em\u003e), as well as in more developed soils, such as Acrisols (\u003cem\u003eArgissolos\u003c/em\u003e) (FUNCEME 2024). Nevertheless, according to Donagema et al. (2011) and Lima et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2002\u003c/span\u003e), coastal soils do not exhibit an exchange complex with an adequate number of base cations to readily supply agricultural crops due to their low pH and low cation exchange capacity (CEC).\u003c/p\u003e\n\u003cp\u003eRegarding canonical variate V\u003csub\u003e1\u003c/sub\u003e (Table \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), data revealed that texture parameters had the following decreasing pattern (Fig. \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e): silt (40.0%) \u0026gt; sand (38.3%) \u0026gt; clay (21.7%), suggesting that silt and sand were the most representative variables for this canonical variate. This confirms the lower contribution of inorganic fractions to soil fertility, reflecting the general particle size distribution observed in the region. The inorganic fractions, particularly sand and silt, have low nutrient reserves (Sarkar et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Thabit et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eCanonical cross-loading revealed the multivariate association between the two explored canonical variates. In other words, the multivariate correlation between chemical constituents and texture parameters is explored. Table \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and the major contributions of the original variables within the respective canonical variates (Fig. \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) suggest that an increase in silt (0.824) and clay (0.446) with a concomitant decrease in sand (-0.789) may result in an increase in Ca\u003csup\u003e2+\u003c/sup\u003e (0.634), Mg\u003csup\u003e2+\u003c/sup\u003e (0.695), and K\u003csup\u003e+\u003c/sup\u003e (0.484). This pattern is consistent because an increase in the silt fraction is accompanied by an increase in the specific surface area of the particles, given that silt particles (0.05\u0026ndash;0.002 mm) are smaller than sand particles (2.00\u0026ndash;0.05 mm) (Provens et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In this context, electrical charge phenomena are more pronounced, enhancing the retention of a greater number of ions such as Ca\u0026sup2;⁺, Mg\u0026sup2;⁺, and K⁺. However, it should be noted that in the clay fraction (\u0026lt;\u0026thinsp;0.002 mm), these phenomena are further intensified, mainly due to the type of clay mineral present (Yu et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eTaking together, the results reported herein, revealed the strength of the relationship between soil texture (sand, silt, and clay) and the potential ions present in the soil (Ca\u0026sup2;⁺, Mg\u0026sup2;⁺, and K⁺) provides valuable insight into the electrical charge phenomena of coastal soils (Santos et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). In addition, the high redundancy observed (\u0026gt;\u0026thinsp;60%) suggest that textural attributes may serve as reliable predictors of soil chemical behavior, and this information could be incorporated into the development of prediction or mechanistic models for a better understanding of the mineral\u0026rsquo;s dynamics in this kind of soil, which remains insufficiently understood. As a result, fertilization practices could be recommended based on an understanding of soil charge characteristics and the types of ions present in association with soil texture (Besen et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This is of practical importance, as soil components are key attributes governing water infiltration, movement, and retention, thereby directly influencing irrigation system management and selection (Shahadha and Wendroth \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eCanonical correlation and redundancy analyses revealed a strong multivariate relationship between soil texture and chemical composition, with more than 60% of the variability in mineral composition and soil acidity explained by textural attributes through the first canonical function. Increases in silt and clay fractions, combined with reductions in sand content, were associated with greater retention of Ca\u0026sup2;⁺, Mg\u0026sup2;⁺, and K⁺, highlighting the importance of particle-size distribution in regulating ion retention and electrical charge dynamics in coastal soils. These findings demonstrate the strong predictive potential of textural parameters and support their application in the development of predictive or mechanistic models aimed at estimating nutrient dynamics, soil acidity, and fertility patterns in coastal environments, particularly where detailed chemical analyses are limited or costly.\u003c/p\u003e"},{"header":"Declarations","content":" \u003cp\u003e \u003cstrong\u003eCompeting interests\u003c/strong\u003e \u003cp\u003eNone\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eNo external funding was received for this study.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eJ.A.C.V.: Conceptualization, methodology, software, validation, formal analysis, investigation, data curation, writing\u0026mdash;original draft preparation, and writing\u0026mdash;review and editing. F.T.G.F.: Data curation and methodology; J.E.F.G.: Conceptualization, methodology, investigation, writing\u0026mdash;original draft preparation, and writing\u0026mdash;review and editing\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors would like to thank the Universidade Estadual Vale do Acara\u0026uacute; (UVA) for providing institutional support in the form of salaries during the development of this research. The authors also acknowledge the Cear\u0026aacute; State Foundation for the Support of Scientific and Technological Development (FUNCAP) for providing a scholarship (grant no. 2347).\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe dataset used in this study is available upon reasonable request from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAmundson R. Factors of soil formation in the 21st century. Geoderma. 2021;391:114960. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.geoderma.2021.114960\u003c/span\u003e\u003cspan address=\"10.1016/j.geoderma.2021.114960\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBesen MR, Ribeiro RH, Esper Neto M, Minato EA, Coneglian CF, Kachinski WD, Tormena CA, Inoue TT, Batista MA. Lime and phosphogypsum application management: changes in soil acidity, sulfur availability and crop yield. Rev Bras Cienc Solo. 2021;45:e0200135. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.36783/18069657rbcs20200135\u003c/span\u003e\u003cspan address=\"10.36783/18069657rbcs20200135\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBizzi LA, Schobbenhaus C, Gon\u0026ccedil;alves JH, Baars FJ, Delgado IM, Abram MB, Le\u0026atilde;o Neto R, Matos GMM, Santos JOS, editors. Geologia, tect\u0026ocirc;nica e recursos minerais do Brasil: texto, mapas e SIG. Bras\u0026iacute;lia: CPRM \u0026ndash; Servi\u0026ccedil;o Geol\u0026oacute;gico do Brasil; 2003. p. 673.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBRASIL. Levantamento explorat\u0026oacute;rio - reconhecimento de solos do Estado do Cear\u0026aacute;. Recife: DPP-MA/DNPEA-SUDENE; 1973. (Boletim T\u0026eacute;cnico n\u0026ordm; 28).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoelho FF, Giasson E, Campos AR, Tiecher T, Costa JJF, Coblinski JA. Digital soil class mapping in Brazil: a systematic review. Scientia Agricola. 2021;78(5):e20190227. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1590/1678-992X-2019-0227\u003c/span\u003e\u003cspan address=\"10.1590/1678-992X-2019-0227\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDertli HI, Hayes DB, Zorn TG. Effects of multicollinearity and data granularity on regression models of stream temperature. J Hydrol. 2024;639(1):131572. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2024.131572\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2024.131572\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiel MI, L\u0026uacute;cio ADC, Lambrecht DM, Pinheiro MVM, Sari BG, Olivoto T, Valera OVS, Melo PJ, Tartaglia FL, Tischler AL, Schmidt D. Canonical correlations in agricultural research: Method of interpretation used leads to greater reliability of results. Int J Innov Educ Res. 2020;8:171\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.31686/ijier.vol8.iss7.2464\u003c/span\u003e\u003cspan address=\"10.31686/ijier.vol8.iss7.2464\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDonagemma GK, Freitas PL, Balieiro FC, Fontana A, Spera ST, Lumbreras JF, Viana JHM, Araujo Filho JC, Santos HG, Albuquerque MR, Carvalho Junior W, Calderano SB. Caracteriza\u0026ccedil;\u0026atilde;o, potencial agr\u0026iacute;cola e perspectivas de manejo de solos leves no Brasil. Pesquisa Agropecu\u0026aacute;ria Brasileira. 2016;51(9):1003\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1590/S0100-204X2016000900001\u003c/span\u003e\u003cspan address=\"10.1590/S0100-204X2016000900001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEspindola C, Pinheiro M, Nakashima M. Proposal for taxonomic revision of thick and highly weathered sandy soils in the Brazilian soil classification system. Discover Soil. 2025;2:124. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s44378-025-00151-3\u003c/span\u003e\u003cspan address=\"10.1007/s44378-025-00151-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFUNCEME \u0026ndash; Funda\u0026ccedil;\u0026atilde;o Cearense de Meteorologia e Recursos H\u0026iacute;dricos. (2024) Levantamento de reconhecimento de m\u0026eacute;dia intensidade dos solos do Estado do Cear\u0026aacute; [electronic book]. Fortaleza, CE: Coletivo Duas Catitas, ePub. ISBN 978-65-983255-0-3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHair JF, Black WC, Babin BJ, Anderson RE. Multivariate data analysis. 8th ed. Boston: Cengage; 2019. p. 813.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIPECE \u0026ndash; Instituto de Pesquisa e Estrat\u0026eacute;gia Econ\u0026ocirc;mica do Cear\u0026aacute;. (2016) Territ\u0026oacute;rios rurais de identidade. In: Atlas do Cear\u0026aacute;. Fortaleza: IPECE. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www2.ipece.ce.gov.br/atlas/capitulo1/11/pdf/Territorios_rurais_identidade.pdf\u003c/span\u003e\u003cspan address=\"http://www2.ipece.ce.gov.br/atlas/capitulo1/11/pdf/Territorios_rurais_identidade.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJaenanda M, Miranda P, Pardo L, Zografos K. An approach to canonical correlation analysis based on R\u0026eacute;nyi\u0026rsquo;s pseudodistances. Entropy. 2024;25(5):713. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/e25050713\u003c/span\u003e\u003cspan address=\"10.3390/e25050713\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKopittke PM, Menzies NW, Wang P, McKenna BA, Lombi E. Soil and the intensification of agriculture for global food security. Environ Int. 2019;138:105078. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.envint.2019.105078\u003c/span\u003e\u003cspan address=\"10.1016/j.envint.2019.105078\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIatrou M, Tziachris P, Bilias F, Kekelis P, Pavlakis C, Theofilidou A, Papadopoulos I, Strouthopoulos G, Giannopoulos G, Arampatzis D, et al. Data-Driven and Mechanistic Soil Modeling for Precision Fertilization Management in Cotton. Nitrogen. 2025;6(2):29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/nitrogen6020029\u003c/span\u003e\u003cspan address=\"10.3390/nitrogen6020029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLima ACC, Oliveira FNS, Aquino ARL. (2002) Limita\u0026ccedil;\u0026otilde;es do uso dos solos do Estado do Cear\u0026aacute; por suscetibilidade \u0026agrave; eros\u0026atilde;o. Fortaleza: Embrapa Agroind\u0026uacute;stria Tropical, 19 p. (Documentos, 54).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMota LHSO, Valladares GS. Vulnerabilidade \u0026agrave; degrada\u0026ccedil;\u0026atilde;o dos solos da Bacia do Acara\u0026uacute;, Cear\u0026aacute;. Revista Ci\u0026ecirc;ncia Agron\u0026ocirc;mica. 2011;42(1):39\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1590/S1806-66902011000100006\u003c/span\u003e\u003cspan address=\"10.1590/S1806-66902011000100006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeina D. (2019) The role of soil pH in plant nutrition and soil remediation. Applied and Environmental Soil Science 2019: 5794869. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1155/2019/5794869\u003c/span\u003e\u003cspan address=\"10.1155/2019/5794869\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProvens ECP, Arrar\u0026aacute;s IR, Santos JMD. Estimating textural fractions of the USDA using those of the International System: a quantile approach. Geoderma. 2022;416:115782. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.geoderma.2022.115783\u003c/span\u003e\u003cspan address=\"10.1016/j.geoderma.2022.115783\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodrigues AF, Latawiec AE, Reid BJ, Sol\u0026oacute;rzano A, Schuler AE, Lacerda C, Fidalgo ECC, Scarano FR, Tubenchlak F, Pena I, Vicente-Vicente JL, Korys KA, Cooper M, Fernandes NF, Prado RB, Maioli V, Dib V, Teixeira WG. Systematic review of soil ecosystem services in tropical regions. R Soc Open Sci. 2021;8(3):201584. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rsos.201584\u003c/span\u003e\u003cspan address=\"10.1098/rsos.201584\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantos AB, Silva AC, Tabarelli M, Barros MF, Franca EJ, Lins SRM, Vanderlei RS. Soil attributes are key drivers of calcium, potassium, and magnesium concentrations and stocks in a tropical dry forest. J Arid Environ. 2026;235:105604. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jaridenv.2026.105604\u003c/span\u003e\u003cspan address=\"10.1016/j.jaridenv.2026.105604\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSarkar I, Hanif M, Khan MZ. Soil organic fractions in cultivated and uncultivated soils of coastal area in Bangladesh. J Agricultural Chem Environ. 2019;8(3):129\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4236/jacen.2019.83011\u003c/span\u003e\u003cspan address=\"10.4236/jacen.2019.83011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSAS Institute. Base SAS 9.4 procedures guide: Statistical procedures. SAS institute; 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchadosin J, Saab SC, Brinatti AM, Pires LF. Quantitative and qualitative characterization of the granulometry of an Inceptisol. Qu\u0026iacute;m Nova. 2023;46(4):329\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21577/0100-4042.20230018\u003c/span\u003e\u003cspan address=\"10.21577/0100-4042.20230018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShahadha SS, Wendroth O. Modeling the effect of tillage and irrigation management on water and barley productivity for different soil textures. Soil Tillage Res. 2025;250:106505. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.still.2025.106505\u003c/span\u003e\u003cspan address=\"10.1016/j.still.2025.106505\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeixeira PC, Donagemma GK, Fontana A, Teixeira WG, editors. (2017) Manual de m\u0026eacute;todos de an\u0026aacute;lise de solo. 3rd edn. rev. e ampl. Bras\u0026iacute;lia, DF: Embrapa, 573 p.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThabit FN, El-Shater AH, Soliman W. Role of silt and clay fractions in organic carbon and nitrogen stabilization in soils of some old fruit orchards in the Nile floodplain, Sohag Governorate, Egypt. J Soil Sci Plant Nutr. 2023;23:2525\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s42729-023-01209-3\u003c/span\u003e\u003cspan address=\"10.1007/s42729-023-01209-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVieilard C, Vidal-Beaudet L, Dagois R, Lothode M, Vadepied F, Gontier M, Schwartz C, Ouvrard S. Impacts of soil de-sealing practices on urban land-uses, soil functions and ecosystem services in French cities. Geoderma Reg. 2024;e00854. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.geodrs.2024.e00854\u003c/span\u003e\u003cspan address=\"10.1016/j.geodrs.2024.e00854\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilson MJ. Weathering of the primary rock-forming minerals: processes, products and rates. Clay Miner. 2004;39(3):233\u0026ndash;66. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1180/0009855043930133\u003c/span\u003e\u003cspan address=\"10.1180/0009855043930133\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXing Y, Wang X, Mustafa A. Exploring the link between soil health and crop productivity. Ecotoxicol Environ Saf. 2025;289:117703. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecoenv.2025.117703\u003c/span\u003e\u003cspan address=\"10.1016/j.ecoenv.2025.117703\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu L, He A, Li H. Soil particle size and Hofmeister effects on soil colloidal aggregation. Appl Clay Sci. 2025;278:108018. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.clay.2025.108018\u003c/span\u003e\u003cspan address=\"10.1016/j.clay.2025.108018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Multivariate relationship, pedology, soils attribute, soil dynamic","lastPublishedDoi":"10.21203/rs.3.rs-9683449/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9683449/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study used canonical correlation analysis (CCA) to evaluate the strength of the multivariate relationship between soil inorganic fractions and chemical attributes associated with the fertility complex in soils of the western coast of the State of Cear\u0026aacute;, Brazil. A dataset comprising 96 observations was constructed from publicly available data from the Medium-Intensity Soil Reconnaissance Survey of the State of Cear\u0026aacute;. The dataset included soil particle size fractions (sand, silt, and clay) and chemical attributes (Ca\u0026sup2;⁺, Mg\u0026sup2;⁺, K⁺, Na⁺, Al\u0026sup3;⁺, and pH). The following multivariate relationship was explored using CCA with the CANCOR procedure of SAS: soil texture fractions versus soil chemical attributes, with a significance level of 0.05. The Wilk\u0026rsquo;s Lambda test revealed two significant (P\u0026thinsp;\u0026lt;\u0026thinsp;0.01) canonical correlations. However, the redundancy analysis revealed that the first one had the greatest explanatory power (60.2%). Therefore, only the first canonical correlation was considered for practical interpretations. Canonical cross-loadings for the first canonical correlation indicated that an increase in silt (0.824) and clay (0.446), along with a concomitant decrease in sand (\u0026minus;\u0026thinsp;0.789) may result mainly in an increase in Ca\u0026sup2;⁺ (0.634), Mg\u0026sup2;⁺ (0.695), and K⁺ (0.484) in the coastal soils explored. Hence, the multivariate approach revealed the strength and specific relationships between soil texture characteristics and key exchangeable cations, information that may be useful for developing future mechanistic models to support decision-making in western coast soils.\u003c/p\u003e","manuscriptTitle":"Brief report: Using canonical correlation analysis to understand the dynamics between texture parameters and chemical attributes in the Ceará State soils of the western coast region in Brazil","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-13 08:14:02","doi":"10.21203/rs.3.rs-9683449/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2e6ee1b6-bcb2-4584-a3cd-8d6c33027b28","owner":[],"postedDate":"May 13th, 2026","published":true,"recentEditorialEvents":[{"type":"submitted","content":"Discover Soil","date":"2026-05-11T18:40:57+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-13T08:14:02+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-13 08:14:02","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9683449","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9683449","identity":"rs-9683449","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}