Across the high reliefs of Cameroon — and tropical mountains generally — soil weathering ranges from base-retaining, smectite-bearing assemblages to extremely desilicated, gibbsite-dominated mantles, and the study that established this pattern (Tsozué et al., 2020) attributed the variation to topography, elevation, rainfall and temperature purely qualitatively, from three lithologically dissimilar single-pit sites. The open problem is to quantify the relative roles of these four controls on desilication-allitisation intensity — as measured by Si/Al ratios, the chemical index of alteration (CIA), and the kaolinite-versus-gibbsite mineralogical balance — separating their effects from one another and controlling for parent-material lithology. By the 'relative role' of a factor is meant a quantitative attribution of that factor's contribution to an indicator under an explicitly stated partitioning or causal model — for example, the share of variation in the indicator attributable to that factor — reported together with the uncertainty of the separation; because different models define different quantities, an attribution is meaningful only together with the model that defines it. No multivariate or process-model apportionment of these controls for this setting is known to date, so the problem remains open.

Background

Chemical weathering in the humid tropics strips soluble bases and, in its most intense form, silica itself from the soil profile, leaving aluminium and iron concentrated in residual minerals. Geochemists describe this with Pédro's pathway framework: monosiallitisation (leaching that removes bases and most silica, forming the 1:1 clay kaolinite), bisiallitisation (weaker leaching that retains enough silica and bases to form 2:1 clays such as smectites), and allitisation (extreme desilication, in which even kaolinite loses silica and the aluminium hydroxide gibbsite forms). The intensity of this desilication-allitisation is conventionally tracked by three complementary measures: the molecular Si/Al ratio of the soil, the chemical index of alteration, CIA = 100·Al₂O₃/(Al₂O₃ + CaO* + Na₂O + K₂O) on molar oxide contents (CaO* being calcium in silicates only), which rises from ~50 for fresh rock toward ~100 as labile cations are leached, and the secondary-mineral assemblage itself (smectite → kaolinite/halloysite → gibbsite). In humid tropical mountains these measures vary sharply over short distances because rainfall, temperature (which falls ~6 °C per km of elevation), and drainage all change with elevation and slope position.

The motivating study is Tsozué, Basga and Nzeugang Nzeukou (2020), who compared one soil pit at a high topographic position in each of three Cameroonian high-relief sites spanning the bioclimatic gradient: Mbalam (humid tropical rainforest at 840 m, garnet-rich micaschist; ~10 m mantle; kaolinite + gibbsite; CIA 52–71), Meleta (pseudotropical mountainous zone at 2740 m, sanidine-bearing trachyte; extreme aluminium enrichment, Al₂O₃ reaching ~41 wt% in the saprolite; gibbsite + halloysite; CIA up to ~95) and Secande (dry tropical zone at 640 m, microgabbro; montmorillonite + kaolinite + calcite, no gibbsite; CIA 51–55). They concluded that topography, elevation, rainfall and temperature are the main factors responsible for the spatial variation of weathering processes — but stated this attribution qualitatively, with no multivariate analysis, and the three-site design confounds climate with parent lithology (three different rocks) and omits slope-position variability (one pit per site).

Adjacent work quantifies only single axes of the problem. Along rainfall gradients, gibbsite occurrence and Si/Al loss have been quantified for northern Thailand (Herrmann et al., 2007, semi-quantitative X-ray diffraction, XRD, along an elevation transect), across tropical rainfall zones in general (Bremer & Sander, 2011), for Guadeloupe Andosols (soils developed on volcanic ash, highly weathered and desilicated here) whose topsoil Si/Al falls from ~1.06 to ~0.37 between ~2650 and ~4400 mm/yr rainfall (Vander Linden et al., 2021), and for Hawaiian volcanic soils (Chadwick et al., 2003; Kramer & Chadwick, 2016, high-elevation gradient). Climosequences (soil sequences differing only in climate) on constant lithology, such as the northeast-Brazil granite transect of da Silva et al. (2016), quantify CIA change along one climate axis, and Bétard's (2012) study of the Baturité massif — the closest design analogue to the Cameroon work — correlates weathering-process zonation with a water-balance index but again qualitatively. The multivariate machinery needed for a genuine apportionment demonstrably exists: Zhang et al. (2024) used redundancy analysis with variance partitioning to quantify the climatic share of weathering-degree variance for granites across temperate-to-subtropical China, and Alsalam et al. (2020) quantified chemical weathering rates along an altitudinal transect in Turkey. What no study does is partition the joint, collinear influences of temperature, rainfall, elevation and topography on the desilication-allitisation metrics of tropical high-relief soils while controlling for lithology. A key subtlety: in mountain terrain elevation is not an independent cause — it acts on weathering through elevation-linked changes in temperature and rainfall and, together with slope position, through drainage, erosion and profile residence time — so raw collinear correlations with 'elevation' do not by themselves identify a causal role.

Problem Statement

Determine, quantitatively, the relative roles of temperature, rainfall, elevation, and topography (slope position and drainage) in controlling the intensity of desilication-allitisation in tropical high-relief soils spanning the bioclimatic gradient (humid tropical, pseudotropical mountainous, dry tropical), as expressed by (i) soil Si/Al ratios, (ii) the chemical index of alteration (CIA, in the CaO* convention defined in the Background), and (iii) the balance among secondary minerals — kaolinite/halloysite versus gibbsite, with 2:1 clays marking the weakly desilicated end. The influences of the four factors must be separated from one another and from parent-material lithology. Elevation and slope position act on weathering through linked mechanisms (elevation-linked temperature and rainfall gradients; drainage, erosion and profile residence time with slope position and elevation), so raw collinear correlations do not by themselves identify the role of any single factor. The target is the general tropical high-relief setting across the full bioclimatic contrast; the tropical high reliefs of Cameroon are the motivating case, whose transect documents monosiallitisation with allitisation (kaolinite + gibbsite) in the humid tropical zone, extreme allitisation with gibbsite and halloysite at the 2740 m pseudotropical mountainous site, and weakly desilicated bisiallitisation (montmorillonite + kaolinite + calcite, no gibbsite) in the dry tropical zone. A solution is a quantitative attribution — for each of the three indicators separately (Si/Al, CIA, and the mineral balance), with the uncertainty of the separation stated — of the four factors and of parent-material lithology for the general tropical high-relief setting, with the Cameroon transect serving as the motivating case; a study-design proposal alone, an attribution of only one of the three indicators, or a merely qualitative ranking does not by itself resolve the task.

References

  1. Désiré Tsozué, Simon Djakba Basga, Aubin Nzeugang Nzeukou (2020), 'Spatial variation of soil weathering processes in the tropical high reliefs of Cameroon (Central Africa)', Eurasian Journal of Soil Science 9(2):92–104, DOI 10.18393/ejss.659830, https://doi.org/10.18393/ejss.659830
  2. François Bétard (2012), 'Spatial variations of soil weathering processes in a tropical mountain environment: The Baturité massif and its piedmont (Ceará, NE Brazil)', Catena 93:18–28, DOI 10.1016/j.catena.2012.01.013, https://doi.org/10.1016/j.catena.2012.01.013
  3. L. Herrmann, N. Anongrak, M. Zarei, U. Schuler, K. Spohrer (2007), 'Factors and processes of gibbsite formation in Northern Thailand', Catena 71:279–291, DOI 10.1016/j.catena.2007.01.007, https://doi.org/10.1016/j.catena.2007.01.007
  4. Hanna Bremer, Heinz Sander (2011), 'Clays a tool for tropical geomorphology', Zeitschrift für Geomorphologie 55:95–107, DOI 10.1127/0372-8854/2011/0055-0041, https://doi.org/10.1127/0372-8854/2011/0055-0041
  5. Oliver A. Chadwick, Robert T. Gavenda, Eugene F. Kelly, Karen Ziegler, Carolyn G. Olson, W. Crawford Elliott, David M. Hendricks (2003), 'The impact of climate on the biogeochemical functioning of volcanic soils', Chemical Geology 202(3–4):195–223, DOI 10.1016/j.chemgeo.2002.09.001, https://doi.org/10.1016/j.chemgeo.2002.09.001
  6. Marc G. Kramer, Oliver A. Chadwick (2016), 'Controls on carbon storage and weathering in volcanic soils across a high-elevation climate gradient on Mauna Kea, Hawaii', Ecology 97(9):2384–2395, DOI 10.1002/ecy.1467, https://doi.org/10.1002/ecy.1467
  7. Ygor Jacques Agra Bezerra da Silva, Clístenes Williams Araújo do Nascimento, Caroline Miranda Biondi, Peter van Straaten, Valdomiro Severino de Souza, Tiago Osório Ferreira (2016), 'Weathering rates and carbon storage along a climosequence of soils developed from contrasting granites in northeast Brazil', Geoderma 284:1–12, DOI 10.1016/j.geoderma.2016.08.009, https://doi.org/10.1016/j.geoderma.2016.08.009
  8. Charles Vander Linden, Zimin Li, Anne Iserentant, Eric Van Ranst, Félix de Tombeur, Bruno Delvaux (2021), 'Rainfall is the major driver of plant Si availability in perudic gibbsitic Andosols', Geoderma 404:115295, DOI 10.1016/j.geoderma.2021.115295, https://doi.org/10.1016/j.geoderma.2021.115295
  9. Shu Zhang, Yong Zhang, Gang Huang, Bo Zhang, Yichan Li, Xin Chen, Junkang Xu, Yujie Wei (2024), 'Linkage between Granite Weathering and Gully Erosion in Subtropical Region', Water 16(5):751, DOI 10.3390/w16050751, https://doi.org/10.3390/w16050751
  10. Omar Alsalam, Cevdet Şeker, Mert Dedeoğlu (2020), 'Quantifying the role of chemical weathering rates on soil developed along an altitudinal transect in the mountainous environments, Turkey', Eurasian Journal of Soil Science 9(2):140–150, DOI 10.18393/ejss.689428, https://doi.org/10.18393/ejss.689428