Promoter: Prof. Dr. Eric Van Ranst Tutor(s): Mathijs Dumon Master dissertation submitted in partial fulfilment of the requirements for the degree of Master of Science in Physical Land Resources By: Joseph Tamale Academic Year: 2013-2014 INTERUNIVERSITY PROGRAMME IN PHYSICAL LAND RESOURCES Ghent University Vrije Universiteit Brussel Belgium Soil characterization in the landslide-prone area Bumwalukani, Eastern Uganda.
91
Embed
Soil characterization in the landslide-prone area ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Promoter: Prof. Dr. Eric Van Ranst
Tutor(s): Mathijs Dumon
Master dissertation submitted in partial
fulfilment of the requirements for the
degree of Master of Science in Physical
Land Resources
By: Joseph Tamale
Academic Year: 2013-2014
INTERUNIVERSITY PROGRAMME IN
PHYSICAL LAND RESOURCES Ghent University
Vrije Universiteit Brussel Belgium
Soil characterization in the landslide-prone area
Bumwalukani, Eastern Uganda.
i
COPYRIGHT
This is an unpublished MSc dissertation and is not prepared for further distribution. The
author and the promoter give the permission to use this Master dissertation for
consultation and to copy parts of it for personal use. Every other use is subject to the
copyright laws more specifically the source must be extensively specified when using results
from this Master dissertation.
Gent,
Author:
Joseph Tamale
Promoter:
Prof. Dr. Eric VanRanst
ii
ACKNOWLEDGEMENT
I would like to express my sincere appreciation and thanks to my promoter, Prof. Dr. Eric
Van Ranst, for being an awesome supervisor and mentor. I would like to thank you for the
encouragement throughout this period and also allowing me an opportunity to grow as a
soil scientist. Your advice both on research and my career has been priceless and will
forever be appreciated.
I also want to thank members of the Jury for my thesis, Prof. Dr. Peter Finke and Prof. Dr.
Jozef (Seppe) Deckers for accepting to read my work.
I wish also to extend my gratitude to my tutor, Drs. Mathijs Dumon for the time spared to
review my text and guide me throughout the preparation of my thesis. Thank you deeply.
I would also like to thank the great team of Laboratory staff, Nicole Vindevogel and Veerle
Vandenhende for the exceptional guidance during the analysis of my soil and rock samples. I
extend special thanks to Mr. Jan Jurcica for his effort investment in the timely preparation
of thin sections and Dr. Florias Mees for the help extended in interpretation of slides.
Many thanks to my colleague, Amaury Defrére from the Catholic University of Leuven for
the great team spirit and wonderful collaboration exhibited that enabled us to gather
enough field data within a short time frame.
A special thanks to my family; words cannot express how grateful I am to my mother, Miss
Erioth Matovu for all the sacrifices you’ve made on my behalf in order to see me through
school. Your prayers and encouragement are what have kept me going till this end. I would
also like to thank all my friends for the enormous support and incredible encouragement
during the entire period of my research.
I wish to categorically state that none of the people mentioned above is accountable for any
of the inaccuracies that may arise in this piece of work apart from the author himself.
iii
DEDICATION
I dedicate this piece of work to my friends and all those who have passion for soil science as
a discipline.
iv
ABSTRACT
Landslides have recently become more frequent and recurrent worldwide almost occurring
every year with devastating socio-economic problems. Research studies undertaken in
different landslide-prone areas in the world indicate a positive correlation between soil
properties and slope failures. In Uganda, however, information on soil properties in general
is to date extremely limited owing to a few general soil characterization studies so far
carried out and thus makes it hard to draw similar correlations. It is against this back ground
that a soil characterization study was done in the landslide-prone area Bumwalukani,
Eastern Uganda in order to contribute to a better understanding of soils in this area that
have been generally described in previous studies.
Three profiles P1, P2, P3 at the upper slope, middle slope, and valley bottom respectively
were selected for the study. Profile 2 (P2) was in close proximity to a scar left behind after
the occurrence of the 2010 Bududa landslide. Both rock and soil samples were obtained
from P1 and P2, but only soil samples were taken from profile 3. These were then analysed
to determine their physico-chemical, mineralogical and micmorphological properties at the
Laboratory of Soil Science, Ghent University, Belgium.
Chemical and mineralogical analyses of the selected rocks show that they are basic in nature
due to their low silica content (<50%). The rocks also have an appreciable amount of
amphiboles, feldspars, kaolinite, quartz and mica. Owing to their weathered state, rock 1
can be called amphibolitic while rock 2 and 3 are phonolitic. In contrast, soils from the three
profiles are deep, yellowish brown (5YR-7YR) in colour, have a moderately acidic to neutral
pH (5.5-6.5), high water dispersible clay content( >10%), high amount of clay and slit (>40%)
for P1 and P2, and a high sand content (>50%) (only for P3). Selective dissolution analysis of
the soil samples shows relative accumulation of crystalline Fe oxides over non-crystalline
(short range order) Fe oxides. The silt and clay fractions of the soils are dominated by
kaolinite, mica and quartz.
Additionally, there are significant differences between the mineralogical composition of the
silt and clay fractions which is attributed to both physical and chemical weathering.
Leaching, clay illuviation, goethite and hematite formation, and bioturbation are the main
soil-forming processes indicated in thin sections. Furthermore, it is evident that soils are
v
highly leached and weathered, with a high spatial variability in soil properties (i.e. Cambisol
at the upperslope, Nitisol at midslope, and Fluvisol at valley bottom) due to differences in
parent materials and topographic positions. Profile 2 (close to the scar left behind by the
occurrence of 2010 Bududa landslide) has a small amount of swelling minerals in its Bt
horizon owing to low intensity peaks from XRD analysis. However, this soil profile (Nitisol)
also has a good internal drainage and therefore infiltration of rainwater to the depth with
swelling minerals can consequently lead to liquefaction, creating a sliding plane at that
depth and thus occurrence of a landslide. Therefore presence of some swelling minerals in
a soil coupled with good internal drainage increases sensitivity to landslides.
APPENDICES ................................................................................................................................ I
viii
LIST OF FIGURES
Figure 1. A profile of a Ferralsol (Oxisol) with indication of different horizons ....................... 5
Figure 2. Schematic representation of the effects of bioturbation on pedogenesis (Wilkinson et al., 2009). ............................................................................................. 7
Figure 3. A fresh scar left behind by the occurrence of 2010 landslide in Bumwalukani (photo by Joseph Tamale, 2013). ............................................................................ 14
Figure 4. A map of Uganda with a window showing the study area ...................................... 15
Figure 5. (a) Cracks in weathered amphibolitic matrix (plane-polarised light); (b) cracks in weathered amphibolitic matrix (crossed-polarised light), (c) grey amphibole matrix with cracks filled with unknown material (crossed-polarised light), (d) laminated illuvial clay infilling in amphibolitic matrix (plane-polarised light), (e) advanced weathering stage of amphibolitic rock, with fragments of goethite, anatase and quartz (plane-polarised light), (f) advanced weathering stage of amphibolitic rock, with fragments of goethite, anatase and quartz (crossed-polarised light) for the bed rock in profile 1 ................................................................................................ 26
Figure 6. (a) Pseudomorph after nepheline replaced by clay (plane-polarised light); (b) pseudomorph after nepheline replaced by clay (crossed-polarised light), (c) plagioclase feldspar (plane-polarised light), (d) plagioclase feldspar (crossed-polarised light) for the bed rock in profile 2 ............................................................ 27
Figure 7. Evolution of O.C (a), pH (b), and CEC (c) with depth for the selected soil profiles. . 32
Figure 8. XRD patterns of silt powders for the respective horizons for profile 1 (all d-spacing of the peaks are in nm) ............................................................................................. 37
Figure 9. XRD patterns of silt powders for the respective horizons for profile 2 (all d-spacing of the peaks are in nm ............................................................................................. 38
Figure 10. XRD patterns of silt powders for the respective horizons for profile 3 (all d-spacing of the peaks are in nm) ............................................................................. 39
Figure 11. (a) Clay illuviation in the saprolite (plane-polarised light); (b) illuvial clay with incorporated coarse kaolinite aggregates (plane-polarised light), (c) illuvial clay along planar voids in soil matrix (plane-polarised light), (d) quartz (white)- amphibole (green) dominated rock fragment (plane-polarised light), (e) fibrous
ix
goethite coating (plane-polarised light), (f) goethite coating consisting of radial aggregates (plane-polarised light) from Profile 1 .................................................. 41
Figure 12. (a) Laminated clay infillings (plane-polarised light); (b) infilling with crescent structure related to termite activity (plane-polarised light), (c) angular feldspar dominated rock fragment (crossed-polarised light), (d) angular feldspar dominated rock fragment (plane-polarised light), (e) fragmented fibrous goethite coating (plane-polarised light), (f) fragmented fibrous goethite coating (plane-polarised light) from Profile 2 ................................................................... 42
Figure 13. Bowen reaction series (1956) ............................................................................... 52
Figure 14. XRD patterns of oriented clay samples of the B2t horizon of profile 2 after K+ and Ca2+ saturation. ...................................................................................................... 53
LIST OF TABLES
Table 1. General site information ............................................................................................ 18
Table 2. Total elemental composition of rock samples related to the selected soil profiles . 25
Table 3. Morphological and physical properties of selected profiles .................................... 30
Table 4. Chemical properties of the selected soil profiles ...................................................... 33
Table 5. Selective dissolution analysis of the selected soil profiles (expressed in %) ............. 35
Table 6. Qualitative mineralogical composition of the silt fraction based on XRD analysis .. 40
Table 7. Qualitative mineralogical composition of the clay fraction based on XRD analysis.. 40
Table 8. Geological classification of rocks (Tilley, 2010). ........................................................ 43
x
ABBREVIATIONS AND ACRONYMS
a.s.l Above sea level
AEC Anion exchange capacity
BS Base saturation
CEC Cation exchange capacity
FAO Food and Agriculture Organisation
GoU Government of Uganda
LOI Loss on ignition
N Normal
NEMA National Environmental Management Authority
O.C Organic carbon
rpm Rounds per minute
WDC Water dispersible clay
WRB World Reference Base
1
CHAPTER ONE: INTRODUCTION
1.1. Background of the study
Landslides rank 10th among the most devastating natural disasters in the world occurring
almost across all terrains with steep slopes singled out as most susceptible to sliding
(Schuster and Highland, 2003; Highland and Bobrowsky, 2008; Leroy and Grachera, 2013).
They occur almost every year leading to significant property damages and deaths (Huabin et
al., 2005; Kirschbaum et al., 2010; Msilimba and Holmes, 2010). Within the period of 2004–
2011, more than 32,000 landslide-related fatalities were recorded worldwide with the 2010-
Bududa landslide in Eastern Uganda being among the major landslides documented there
(Pankow et al., 2014). This landslide event destroyed a lot of valuable property, claimed
over 300 lives and rendered about 5000 people homeless. It was the most devastating
catastrophe ever recorded in the history of Uganda (Gorokhovich et al., 2013). Mugagga et
al. (2012) noted that this landslide event was attributed to recurrent heavy rainfall and
slope failure due to overloading. The latter is a consequence of human development
expansion on unstable hilly slopes increasing their vulnerability to sliding (Smyth and Royle,
2000; Huabin et al., 2005).
However, Highland and Bobrowsky (2008) noted that most of the slope failures in landslide-
prone areas are by far linked to the nature of slope materials (rock and soil). Soil materials in
landslide-prone areas have considerable amounts of swelling minerals in the fine earth
fraction and therefore seasonally swell and shrink following wet and dry periods (Yalcin,
2007). Upon drying up, cracks develop which grow with subsequent periods of swelling and
shrinking. These cracks increase preferential entry of water in soils and consequently lead to
early saturation following a rainfall event causing liquefaction incase of a trigger like an
earthquake (Galeandro et al., 2011). If cracked fine-grained soils are overlying a more
permeable material, water flow through this type of soil system can be affected by closing
of cracks due to swelling of overlying fine-grained soils. Rainfall infiltration in such soil
systems can consequently lead to a dramatic decrease in suction of unsaturated soils
triggering various instability phenomena, such as the slip of steep surface soil layers.
2
Basharat et al. (2014) additionally, suggested earthquakes as a trigger for landslides
resulting into rock falls, debris falls, rockslides and rock avalanches.
Unlike developed countries, e.g. China and USA, where enormous efforts have been
invested in research about landslides (e.g. Hua-xi and Kun-long (2014) in China, Chung et al.
(2014); Tao and Barros (2014) in the USA) to better understand this phenomenon and foster
adoption of feasible mitigation measures in case of an occurrence, developing countries
(e.g. Uganda) still lag behind despite the socio-economic threats posed by the occurrence of
this catastrophe (landslide) on their populations.
In Uganda, a few studies have been carried out in the frame work of landslides with major
emphasis put on evaluation of causal factors and physico-chemical properties of the soils.
However, mineralogy of the soils was completely neglected despite its strong influence on
stability of slopes. Moreover, the two studies i.e. Kitutu et al. (2009) and Muggaga et al.
(2011) that focussed on physico-chemical properties were incomprehensive and don’t give a
complete picture about the nature of the soils. To highlight, Kitutu et al. (2009)
characterized and classified these soils as Cambisols, Ferralsols, Lixisols and Nitisols.
However, the classification was general and inconclusive since it was based on incomplete
analytical data. No soil mineralogical analysis was done at all yet it’s useful in understanding
the type of clay minerals (e.g. swelling minerals) present in the soils. Additionally, no
information on the presence or absence of weatherable minerals can be traced from their
study. In regard to this, one wonders how soils were classified as Ferralsols when analytical
data required in verifying presence or absence of a Ferralic diagnostic horizon was never
obtained. Mugagga et al. (2011) only focussed on particle-size distribution of the soils in the
region and used this to extrapolate and draw conclusions about their mineralogy. Therefore
there is still a big information vacuum on the properties of soils in Bumwalukani which
needs to be filled. In this study, a complete and comprehensive analysis of the physico-
chemical, mineralogical and micromorphological parameters of both bulk and undisturbed
soil samples from Bumwalukani, Eastern Uganda, was done.
3
1.2. Justification
It’s been said that scientists know more about soils of Mars than those of Africa (Sanchez et
al., 2010). This to a great extent explains why Uganda to-date still uses the conventional less
detailed and out-dated soil map of 1967 made at a scale of 1:1,500,000. Isabirye et al.
(2004) mentioned scarcity of information about soils in Uganda as a major hindrance and
obstacle in sustainable management of this resource. The little information that is available
is also not digital. Therefore there is a need to do more soil characterization and
classification to contribute to the soil resources information data base of Uganda. This
study was oriented towards collection of a considerable amount of information on the
properties of soils in the landslide-prone area and grouping studied profiles into their
respective reference groups. Additionally, results from the study would be used as a basis
for drawing linkages between soil types and landslide risk in future research studies.
1.3. Objectives
The main objective of the study was soil characterization in the landslide-prone area,
Bumwalukani in Eastern Uganda.
The specific objectives of the study were:
(1) to identify soil-forming processes using observable micromorpholgical features
indicated in thin sections; and
(2) to determine physico-chemical and mineralogical properties of the soils in order to
understand better their sensitivity to land sliding.
4
CHAPTER TWO: LITERATURE REVIEW
2.1. Soil-forming processes in Mount Elgon Region, Eastern Uganda
Soil formation is a result of soil forming processes whose rate is influenced by the intensity
of soil forming factors like climate, living organisms, parent material, topography and time
(Brady and Weil, 2002). In regard to Eastern Uganda, the following soil-forming processes
have been reported and they include; ferralitization, biotic activity, erosion, eluviation,
leaching and illuviation.
2.1.1. Ferralitization
Ferralitization process is associated with strong weathering leading to formation of soils
with a ferralic B horizon (i.e. Oxisols (USDA) or Ferralsols (FAO)). Oxisols (USDA) or Ferralsols
(FAO) (Figure 1) are extremely weathered soils with diffuse horizon boundaries (van
Breemen and Burman, 2002). Additionally, these soils are dominated by gibbsite, Fe oxides
and kaolinite in the clay fraction and as a consequence have a low CEC (<16 cmol(+)/kg at
pH7) and low cation retention (<10 cmol(+)/kg at soil pH). Neufeldt et al. (1999) noted that
ferralitization is a common soil-forming process in humid and subhumid tropics because of
the extended periods of wet and hot climate. However, it can also occur in ustic and drier
climates provided there is a wet period in between a long dry period. Ferralitization
(residual accumulation of (hydr)oxides of Fe and Al) is preceded by desilication (net loss of Si
from primary silicates and quartz) and kaolinitization (formation of kaolinite) (Neufeldt et
al., 1999). Desilication (chemical weathering of silicates) typically involves hydrolysis as the
most important reaction. It involves splitting of water molecules into their H (often
substitutes a cation from the mineral lattice) and OH components. Van Breemen and
Buurman (2002) used hydrolysis of feldspars (highly abundant Al-silicate minerals) to explain
both desilication and kaolinitization reactions in soils. Feldspars are more stable in the
earth‘s interior than at the surface because of favourable conditions (high temperatures and
little water) there compared to the earth‘s surface (low temperature, abundant water)
(Hieronymus et al., 1990; Hai and Egashira, 2008). The hydrolysis of feldspar involves H+ ions
from water reacting with Oxygen of feldspars which leads to disintegration of the primary
mineral thus desilication. The K and part of the silica (Si) are removed from the soil resulting
into formation of kaolinite (secondary clay mineral) hence kaolinitization. However, both
5
processes are extremely slow (Preetz, 2008). Conversely, residual accumulation of
sesquioxides (Al/Fe oxides) is a result of strong depletion of basic cations owing to their high
solubility leaving Al/Fe oxides to precipitate and accumulate (Schwertmann, 1985). As a
result of strong weathering of primary minerals in Ferralsols, their silt content is relatively
low (i.e. silt/clay ratio in most Ferralsols is <0.15) (van Breemen and Buurman, 2002).
Figure 1. A profile of a Ferralsol (Oxisol) with indication of different horizons.
From the physico-chemical results and thin sections of the selected profiles, the following
soil-forming processes can be identified;
5.2.1.1. Leaching and clay illuviation
Leaching refers to loss of mineral and organic solutes together with percolating water in the
soil. It is often predicted by fraction of clay that disperses in water (WDC). High WDC values
indicate high susceptibility to leaching (Kjaergaard et al., 2004). Additionally, evidence of
leaching can be reflected by the amount of exchangeable base cations present on the soil
exchange complex. Results of the study show the relative abundance of respective base
cations on the soil exchange complex as follows: Ca2+ >Mg2+> K+> Na+. This implies that
divalents are more common than monovalents. Abundance of exchangeable base cations on
the exchange complex is largely affected by their solubility. In humid tropics, solubility is
linked to hydration energy (heat energy released when new bonds are formed between ions
and water molecules) and lattice energy (forces holding the crystals together) of the cations.
The extent to which hydration energy is greater than lattice energy determines whether or
not the cation will be soluble. Hydration energies of base cations are in the order; Na+>
K+>Mg2+>Ca2+. This implies that Ca2+ is sparingly soluble (low hydration energy) while Na+ is
highly soluble (high hydration energy) and this explains the high amount of Ca2+ on the
exchange complex compared to Na+ in all the profiles (Table 4). Additionally, the base
saturation of P1 and P2 located at middle and upper slope positions (where drainage is
good) is low (<50%) compared to high BS values (>50%) in P3 located at the valley bottom
(where drainage is impeded). The difference in BS is good indicator about the soil leaching
regime. Profile 1 and 2 are highly leached compared to P3 owing to low and high BS values
respectively.
There is clay illuviation in the Bt (B1t and B2t) horizons of profiles 2 evidenced by presence
of clay coatings along planar voids (Figure 11 and 12) and an increase in clay content with
depth (Table 3). The mechanical migration of clay from surface to deep horizons in a soil
profile is attributed to mobilisation of clay by percolating atmospheric precipitation and
subsequent infiltration in form of suspensions through macro soil voids. When the
45
suspensions reach deep horizons where the soil is dry, water in them is suctioned out
through micro voids of surrounding areas leading to formation of fine clay skins/argillans
having particles arranged parallel to each other and also parallel to the walls of voids. It is
these clay skins that coat walls of macro voids (Rawling, 2000). Similarly, when the
suspensions reach aggregates, water deposits clay particles on their surface as it goes
towards the interior of edaphic aggregates forming illuviation argillans that cover the
aggregate (Fedoroff, 1997). Clay migration is favoured by slightly acidic to neutral pH typical
of the studied profiles (Table 4). Under these pH conditions, clay particles are easily
dispersed irrespective of the concentration of Ca2+ on the exchange complex (Greene et al.,
1988). Furthermore, the bimodal climate (characterised by distinct dry and wet seasons) in
the area highly favours precipitation and translocation of clay during dry and wet periods
respectively. In P2, surface horizons have low bulk densities compared to subsurface
horizons (Table 3) which could as well indicate clay accumulation in the latter. The high bulk
density in subsurface can be linked to filling of pores with illuvial clay which consequently
clogs fine pores leading to compaction and reduction in pore space. All these observation
indicate that clay illuviation is an important soil forming process in the study area.
5.2.1.2. Formation of goethite and hematite
Thin sections of subsurface horizons of P1 and P2 show presence of goethite in form of
fibrous and radial coated aggregates (Figure 11 e, f; and 12 e, f). Goethite (α-FeOOH) is a
crystalline Fe oxide formed by weathering of iron bearing minerals (van der Zee et al., 2003).
The formation and transformation of Fe oxides is based on numerous chemical reactions but
the most pronounced is hydrolytic and oxidative decomposition of lithogenic containing
primary minerals mainly Fe (II) silicates (Sposito, 1989) illustrated by the equation below;
Fe2SiO4 + 1/2O2 + 3H2O → 2FeOOH + H4SiO4
In the soil environment this reaction is irreversible. The degree to which Fe (II) silicates are
transformed to goethite or hematite reflects the degree of weathering of a soil and varies
widely between weakly and strongly developed soils. The extent of transformation of Fe (II)
silicates to Fe (III) oxides can be measured using ratios of Fe in oxides and total Fe through
selective dissolution analysis (Schwertmann, 1989). The formation of crystalline Fe oxides
(usually goethite and hematite) is not practically achieved in a single step instead follows a
46
chain of reactions which yield intermediate non-crystalline Fe oxides (ferrihydrite) before
formation of crystalline Fe oxides. Ferrihydrite (a poorly crystalline Fe oxide) is the initial
product of weathering of iron rich minerals but re-crystallizes under favourable conditions
forming goethite (Schwertmann, 1985). The high amount of ferrihydrite and oxalate-
extractable Fe in P3 than dithionite-extractable Fe (Table 5), suggests that the soils have
more non-crystalline Fe oxides than crystalline Fe oxides. This is because soils of P3 are
young (Fluvisol) owing to fresh sediments received from the seasonal flooding of River
Tsutsu. However, an opposite trend is observed in P1 and P2 since their ferrihydrite values
are low probably because most of it has been converted to goethite/hematite owing to a
higher degree of weathering. Moreover, there is more dithionite-extractable Fe than
oxalate extractable Fe (Table 5) in P1 and P2 implying that crystalline Fe (e.g.
goethite/hematite) is more abundant than amorphous Fe in these soils. Goethite formation
is favoured by warm temperatures (typical for this area) and low organic carbon
(characteristic of BC and Bt horizons of P1 and P2 respectively). Additionally, the low
Feo/Fed ratio (< 0.1) in the two profiles also suggests high abundance of crystalline Fe oxide
(Torrent, 1994).
Both hematite and goethite are formed from ferrihydrite. However, the mechanism of
formation of these two Fe oxides is significantly different. Goethite crystals form in solution
from dissolved Fe (III) ions produced by the dissolution of ferrihydrite, whereas hematite
forms through an internal dehydration and rearrangement within the ferrihydrite
aggregates (Schwertmann, 1985). Therefore, goethite formation is favoured by an increase
in the concentration of Fe (III) ions in equilibrium with ferrihydrite while hematite is
favoured by a decrease in concentration. The concentration and form of Fe (III) ions in
equilibrium with ferrihydrite are strongly dependent on pH. Goethite formation is favoured
by pH values ranging between 4 and 7 while hematite is favoured by pH (>8) (Ebinger and
Schulze, 1990). Indeed the results of soil pH for the respective profiles (Table 4) are within
the pH range favouring formation of goethite and thus the yellowish brown colour of the
soils.
47
5.2.1.3. Bioturbation
Infillings with a crescent structure present in profile 2 are attributed to termite activity
(Figure 12 c) and thus illustrate the influence of soil fauna activity on soil properties.
Termites and other soil fauna play an important role in transporting and altering soil
components like soil structure formation and organic matter decomposition thereby
influencing physical and chemical processes in the soil (Castellanos-Navarrete et al., 2012).
However, effects of soil fauna on soils can be multifold and complex e.g. one group alone for
example earthworms, termites or molluscs can produce over 50 different features (Stoops,
2010). Preferential transportation of fine materials from deeper horizons to the surface by
termites during nest building creates fine textured soils at the surface and coarse textured
soils in the subsurface horizons (Van Wambeke, 1992). He further noted that termite
activity could lead to formation of well sorted soil materials, free of gravel and with a
mineralogical composition similar to that of the underlying rock. Generally, the contribution
of termite activity to soil textural evolution in the studied profiles is complex to quantify
despite the evidence of crescent infillings in the thin sections.
5.2.2. Evolution of soil morphological and physico-chemical properties in relation to
weathering
Soil morphological properties
Colour
There is a clear distinction between colour of surface and subsurface horizons of the studied
profiles. Surface horizons are darker owing to a higher O.C content than subsurface horizons
(Table 4). Conversely, later (subsurface horizons) are more yellowish brown in colour than
the former (surface horizons) (Table 3). Yellowish brown colour in soils is associated with
liberation and accumulation of free iron oxides (goethite) upon chemical weathering. This is
predominant in areas which receive high precipitation totals with alternating dry and wet
periods (typical of the study area) (Van Wambeke, 1992). Additionally, the increase in both
oxalate and dithionite-extractable-Fe contents with depth (Table 5) can also be associated
with precipitation and accumulation of Fe oxides after soluble cations have been leached to
deeper horizons. Translocation of soluble elements is enhanced by a good micro and macro
pore network system, and deep drainage typical of P1 and P2. In the past, several
48
hypotheses were suggested to account for yellowish brown and reddish colour in soils e.g.
Van Wambeke (1992) suggested that red colour in soils was strongly related to hematite
content but the difference in intensity of redness was attributed to different
hematite/goethite ratios. Additionally, he hypothesised that the red colour was linked to
preferential dissolution of hematite from existing hematite/goethite mixtures.
Soil physico-chemical properties
Particle size distribution
Textural evolution in soils is largely dependent upon the type of parent material and stage
of weathering. Parent materials are often characterized by their clay producing capacity. A
basic rock will produce more clay upon weathering than an acid rock. Soils (derived from
basic rocks) of P1 and P2 are characterized by a clayey texture (in A and B horizons) and silt
clay (in BC or CB horizon). However, soils of P3 (young) have a significant amount of sand
which is linked to high amount of coarse grained weatherable primary minerals
(amphiboles) in the profile. The abrupt decrease in clay content close to the saprolite is
attributed to high weatherability of primary minerals in the parent rock.
Soil pH
According to Landon (2014), the pH of the three profiles can be described as near to neutral.
There is a general increase in pH with depth (Figure 7b) since upper horizons receive more
precipitation than deeper horizons thus are more leached (i.e. the former has low amount
of base cations compared to the later) (Sposito, 1989). The amount of acidity in soils
depends on the stage of evolution and mineralogy of soils (Van Ranst, 1991). Conversely, in
soils with high variable charge, amount of acidity depends on ‘point of zero charge’ (PZNC)
i.e. pH at which the net total surface charge is zero. The difference (∆pH) between pH
determined in a KCl solution and distilled water (pH KCl - pH H2O) is used to estimate PZNC.
∆pH is either positive or negative but the smaller the difference, the lower the amount of
negative charge. Variation in ∆pH is a good indicator of the stage of weathering in tropical
soils. Baert (1995) suggested the following pH ranges for soils derived from basic rocks in
relation to weathering; recent stage of weathering: high pH-H2O values (6.4 to > 7) and a
∆pH ranging from -1.0 to -2.1 in the subsoil; intermediate stage of weathering: pH-H2O
49
values range from 5.5 to 6.5 and a ∆pH between -0.4 to -0.9; ultimate stage of weathering:
pH-H2O values range from 4.2 to 5.5 and a positive ∆pH. If these pH and ∆pH ranges are
considered in the determination of the stage of weathering of the studied soils, all soil
profiles would be classified as young soils but this is not true since they are deeply leached
and weathered with significant amount of Fe oxides typical of old soils . Therefore the stage
of weathering of the soils in the study area can’t be adequately explained by the obtained
pH and ∆pH values.
5.2.3. Mineralogical composition of silt and clay fraction
Clay fraction
Profile 2 (close to the scar left behind by the occurrence of 2010 Bududa landslide) was
selected for a detailed qualitative mineralogical analysis to identify specific clay minerals in
this profile. XRD patterns of oriented clay samples (from the respective horizons of P2)
saturated with K+ (and after heating at 350 and 500 °C) and Ca2+ (air dry and glycolated)
(Figure 14 and Appendix III) were used in the confirmation of constituent minerals. The
following minerals were identified;
Kaolinite
Kaolinite is commonly identified by the 0.72-0.360 nm XRD maxima if neither chlorite nor
Mg-saturated vermiculite is present. All diffractograms show rational series of OOI
reflections starting with 0.72 nm (1st order), 0.360 nm (2nd order), 0.240 (3rd order). After
heating at 550 °C, all these peaks disappear because the lattice is completely destroyed as a
result of dehydroxylation and therefore this confirmed presence of kaolinite in this soil
profile.
Mica
Micas are characterized mainly by two intense peaks in the region of 1.0 and 0.33 nm and a
relatively weaker peak at 0.5 nm. All XRD patterns show a strong first order reflection (1.0
nm) and weak second order at 0.5 nm and these are unaltered upon glycerol solvation, K
saturation and heating up to 500 °C. The weaker second order reflection might suggest the
presence of some Fe-for-Al substitution in the crystal lattice of a phyllosilicate. Fe influences
50
the relative intensities of the (OOI) diffraction lines particularly the 0.5 nm line. Most likely
the mica is type of a muscovite.
Quartz
XRD patterns show well defined, sharp and symmetric peaks at 0.334 and 0. 426 nm
confirming presence of quartz in the profile. All the profiles have high amounts of quartz
(Table 6 and 7) and there is uniform distribution with depth.
Silt fraction
XRD patterns of the silt powder (Figure 8, 9 and 10), show quartz as the dominant non-clay
mineral in all the profiles. However, profile 1 and 2 have appreciable amounts amphiboles
with traces of anatase (weak peak of 0.350nm) in P1. Anatase is usually common in strongly
weathered soils because it easily accumulates due to its low solubility.
5.2.4. Evolution of mineralogy of silt and clay fractions in relation to weathering
The presence or absence of various minerals in the silt and clay fractions of the selected
profile (P1, P2, and P3) can be attributed to weathering. It involves disintegration of primary
minerals and their reformation into new (secondary) minerals. The latter are usually formed
by dissolution or chemical alteration of minerals. The weathering process is favoured by
availability of water (medium for chemical and physical reactions), oxygen (for oxidation-
reduction reaction) and presence of cleavage planes and fissures (allow translocation of
dissolved minerals). All these factors are in abundance in the environment where the soil
samples were obtained from. The silt fraction is essentially dominated by inherited primary
minerals while the clay fraction is dominated by secondary minerals either formed in-situ or
transported from other environments. However, it is not unusual to find feldspars, mica and
quartz in the clay fraction noted Karathanasis (2006) (Table7) since they are relatively stable
to chemical weathering but can be broken down into clay sized fractions by physical
weathering. Bowen (1956) suggested a schematic series to explain evolution of minerals
(Figure 13). The minerals at the top (also referred to as the zone of rock formation) of the
illustration are first to crystallize. Similarly, the temperature gradient can be read to be from
high to low with the high temperature minerals being on the top and the low temperature
51
at the bottom. The chart can be used as an indication for the stability of minerals i.e. the
ones at the bottom being more stable and the ones at top being more susceptible to
weathering. This is because minerals are most stable in the conditions closest to those
under which they were formed i.e. minerals formed at high temperatures are more stable at
high temperatures and those formed at low temperatures are more stable at low
temperatures, and because of differences in composition i.e. minerals containing more base
cations (like Ca2+, Mg2+, K+) are more susceptible to weathering compared to minerals
having relatively more silica and or aluminium. Aside from these general principles, the
crystal structure obviously plays a very important role. The high stability of quartz and mica
(i.e. most likely muscovite) explains why they are present in the silt and clay fraction of the
studied profile. The presence of both minerals in the silt and clay fractions suggests that
they are susceptible to physical weathering but more resistant to chemical weathering (Van
Ranst, 1991). Very few feldspars are seen in the silt fraction and completely absent in the
clay fraction since they are quickly weathered before arriving in the clay fraction.
Amphiboles are present in profile 3 (Fluvisol) probably because it receives fresh sediments
rich in these minerals from seasonal flooding of Tsutsu River during rainy periods, as this
minerals should normally weather faster compared to feldspars or micas.
There are noticeable mineralogical differences between the composition of silt and clay
fractions in the studied profiles and are as follows; silt fraction; quartz > kaolinite > feldspars
> amphiboles > mica > open 2:1 phyllosilicates; clay fraction; mica, kaolinite > quartz > open
2:1 phyllosilicates. Presence of open 2:1 phyllosilicates in both silt and clay fraction suggests
chemical alteration of primary minerals. The swelling minerals seen both in profile 2 (Nitisol)
and 3 (Fluvisol) (Table 6 and 7) could be as a result of transformation of mica (biotite) to
trioctahedral smectites or weathering of amphiboles and ferromagnesian minerals in the
respective profiles.
52
Figure 13. Bowen reaction series (1956)
53
Figure 14. XRD patterns of oriented clay samples of the B2t horizon of profile 2 after K+
and Ca2+
saturation.
54
5.2.5. Classification of the selected soil profiles
Classification of all the profiles was based on the guidelines proposed in the World
Reference Base for Soil Resources (IUSS Working Group WRB, 2014).
Profile 1
Profile 1 has a deep (105 cm deep) diagnostic B (BC) horizon starting from 35 to 140cm.
Horizon BC contains some characteristics of the saprolite (C) but the properties of B
dominate over those of C. It has a clay content of 52%, CEC clay of 30.47 cmol (+) kg-1clay and
a base saturation of 43%. The silt to clay ratio is lower than 0.4. Accordingly, horizon BC is a
cambic horizon owing to evidence of alteration relative to underlying horizons.
Furthermore, the cambic horizon has silty clay to clay texture, shows evidence of
pedogenetic alteration, does not form part of another typical horizon and has a thickness of
more than 15 cm. Therefore Profile 1 is a Cambisol with dystric and rhodic suffix qualifiers.
The suffix qualifier dystric is used because the BS is less than 50% in the major part between
20 and 100 cm from the soil surface. Rhodic is used because the Munsell hue is redder than
2.5YR and the value is less than 3.5 i.e. the hue and value for the BC horizon is 5YR 3/6.
Therefore soil profile P1 is a Cambisol (Dystric, Rhodic).
Profile 2
Profile 2 has a deep (180cm deep) diagnostic Bt horizon (comprising of B1t and B2t) starting
from 70 to 250cm. The horizon has higher clay content (72%) than the overlying and
underlying horizons. Clay skins, shiny peds and strong angular blocky structure were
observed in the field indicating clay accumulation in this horizon. There is a 1% (< 20%)
change in clay content over 20 cm depth to layers immediately above and below Bt, a WDC
to total clay ratio of 0.03, a silt to clay ratio of 0.33 (<0.4), 7.2% (>4%) citrate-dithionite
extractable Fe in the fine earth fraction, 0.8% (>0.2%) acid oxalate extractable Fe in the fine
earth fraction, and 0.11 (>0.05%) ratio between active and free Fe. Accordingly, horizon Bt
is a nitic horizon since it meets all the criteria for a nitic horizon as per the guidelines of WRB
(2014). Profile 2 is therefore a Nitisol with dystric and rhodic suffix qualifiers. The suffix
qualifier dystric is used because the BS is less than 50% in the major part between 20 and
100 cm from the soil surface. Rhodic is used because the Munsell hue is redder than 2.5YR
55
and the value is less than 3.5 i.e. the hue and value for BC horizon is 5YR 3/6. Therefore soil
profile P2 is a Nitisol (Dystric, Rhodic).
Profile 3
Profile 3 has a high sand content (>45%) in the fine earth fraction because the area is
seasonally flooded when water spills over the banks of River Tsutsu during rainy seasons
thus allowing deposition of fresh coarse sediment. There is clear horizonation, gleyic color
pattern in the subsurface horizons, very high CECclay (>60 cmol (+) kg-1 clay), and high base
saturation (>50%) in all the horizons of P3. The fluctuating water table (found at 64cm)
normally creates reducing conditions seen in the profile. Profile 3 is a Gleyic Fluvisol (Eutric,
Loamic). The prefix qualifier gleyic is used due to evidence of reducing conditions at some
depth with in the profile and presence of a gleyic colour pattern. The suffix qualifier eutric is
used because the base saturation on average throughout the entire depth of the profile is
greater than 50%. Loamic is used due to a sandy loam texture in a layer ≥ 30 cm thick, within
≤ 100 cm of the mineral soil surface.
5.2.6. Implications for landslide susceptibility
Soil mineralogy
Downslope movement of slope materials (soils, rocks and debris) normally happens when
the force of gravity (major driving forces)/ shear stress exceeds the slope shear strength
(Highland and Bobrowsky, 2008). Shear strength (sum of all forces resisting downslope
movement of materials) can be attributed to frictional resistance and soil cohesion which
are both influenced by the mineralogy of the soil. Presence of swelling minerals in a soil
influences the stability and susceptibility of slope materials to sliding. Indeed XRD results
(Appendix II) for P2 (close to fresh scar from occurrence of landslide) and P3 in the valley
bottom reveal the presence of swelling minerals. Discussion of influence of mineralogy on
shear and frictional resistance of the soils is made in reference to observations of Fall et al.
(2006). Presence of swelling clay minerals in soils imposes a high plasticity on soils lowering
their resistance to deformation. Since plasticity is a result of adsorption of water in clayey
soils, heavy rainfalls can trigger higher plasticity and consequently lower shear strength.
56
Soils with swelling minerals undergo cyclic seasonal swelling and shrinking which is favoured
by wet and dry periods respectively (typical of the climate of the study area). The alternate
swelling and shrinking during wet and dry seasons creates slip planes (when moist) and
cracks (when dry). These cracks usually provide preferential infiltration of water and thus
saturation is easily reached in areas with such features. In addition, the sensitivity to
formation of erosional features (e.g. rills and gullies), becomes higher due to channelling of
runoff water. Once the soil materials become saturated, the cohesion is lowered and in case
of any trigger like an earthquake, the water pressure will increase to a point where soil
particles can readily slide relative to each other causing a landslide. This should make it clear
that landslides are complex phenomena involving an interaction of many factors and
therefore cannot be solely explained by mineralogy of slope materials. However, the
presence of swelling minerals can be a good indicator for the susceptibility of slopes to
failure.
57
CHAPTER SIX: CONCLUSIONS AND SUGGESTIONS FOR FURTHER RESEARCH
6.1. General conclusions
Soils in the landslide-prone area of Bumwalukani, Eastern Uganda, were investigated for
their physico-chemical, micromorphological and mineralogical properties to determine the
dominant soil-forming processes, and their sensitivity to landslides. The studied soil profiles
are within the same geographical setting, therefore the significant differences in their
properties cannot be linked to differences in climate. Younger soils were found close to river
in the valley bottom while well-developed soils were found in the middle and upper slope
topographic locations.
Leaching, clay eluviation and illuviation, goethite/hematite formation, and bioturbation are
the dominant soil forming processes. Accumulation of iron oxides in the middle and upper
slope profiles is an indicator of intense weathering. Additionally, the yellowish brown to
reddish brown colour of the soils can be attributed to high amounts of iron oxides (e.g.
goethite/hematite). Kaolinite and mica are the dominant minerals (typical of older soils) in
the clay fraction. Conversely, the silt fraction is dominated by quartz. Soils also have a high
clay content compared to silt and sand contents which indicates an advanced degree of
physical and chemical weathering.
Open 2:1 phyllosilicates observed in profile 2 (close to the scar of a recent land slide) and P3
suggest susceptibility of these soils to sliding following moderate to heavy rainfall events.
XRD peaks for open 2:1 phyllosilicates were not sharp in P2 possibly indicating a low amount
of these minerals in the soils. However, the fact that there is a scar from the occurrence of a
landslide in the vicinity of P2 suggests that soils can still be susceptible to sliding even at low
amounts of swelling minerals provided other contributing factors are present. Finally, the
presence of different soil types e.g. Cambisol (P1), Nitisol (P2), and Fluvisol (P3) within a
small geographic setting implies high spatial variability in soil properties within the study
area and therefore calls for caution when making generalized conclusions based on limited
soil data sets.
58
6.2. Recommendations for further research
Future studies should try to focus on the determination of area specific threshold values for
soil physical parameters (e.g. shear strength, plasticity index, and infiltration capacity) that
determine the risk of slope failures in relation to their mineralogy. Furthermore, materials
above, on, and below slip plane scars should be characterized for their physico-chemical and
mineralogical properties to gain insight on the factors leading to their formation.
59
REFERENCES
Arya, L. M., Leij, F. J., Shouse, P. J., and M. T. van Genuchten. 1999. Relationship between the hydraulic conductivity function and the particle-size distribution. Soil Science Society of America Journal, 63 (5), 1063-1070.
Baert, G. 1995. Properties and chemical management aspects of soils on different parent
rocks in the Lower Zaire. PhD thesis. Bagnavets, O.S. 1989. On the history of soil formation in the Holocene in the northern part
of the Pre-Volga upland. In: Anthropogennic and Natural Evolution of Soils and the Soil Cover. Pushchino, Moscow, pp.237–239.
Basharat, M., Rohn, J., Baig, M. S., Khan, M. R., and M. Schleier. 2014. Large scale mass
movements triggered by the Kashmir earthquake 2005, Pakistan. Journal of Mountain Science, 11, 19-30.
Blokhuis, W. A., Kooistra, M. J., and L. P. Wilding. 1990. Micromorphology of cracking clayey
soils (Vertisols). Developments in Soil Science, 19, 123-148. Bockheim, J. G., Marshall, J. G., and H. M. Kelsey. 1996. Soil-forming processes and rates on
uplifted marine terraces in South Western Oregon, USA. Geoderma, 73 (1), 39-62. Bowen, N. L. 1956. Biographical Memoirs of Fellows of the Royal Society, 3, 6–26. Brady, N. C., and R. R. Weil. 2002. The nature and properties of soils. Upper Saddle River.
Prentice Hall, New Jersey. Brubaker, S. C., Holzhey, C. S., and B. R. Brasher. 1992. Estimating the water-dispersible clay
content of soils. Soil Science Society of America Journal, 56 (4), 1226-1232. Bunnenberg, C. and M. Taeschner. 2000. Soil fauna transport versus radionuclide migration.
Radiation Protection Dosimetry, 92, 35–38. Castellanos-Navarrete, A., Rodriguez-Aragones, C., De Goede, R. G. M., Kooistra, M. J.,
Sayre, K. D., Brussaard, L., and M. M. Pulleman. 2012. Earthworm activity and soil structural changes under conservation agriculture in central Mexico. Soil and Tillage Research, 123, 61-70.
Cerda, A., and M. F. Jurgensen. 2008. The influence of ants on soil and water losses from an
orange orchard in eastern Spain. Journal of Applied Entomology, 132 (4), 306-314. Chaopricha, N. T., and E. Marín-Spiotta. 2014. Soil burial contributes to deep soil organic
carbon storage. Soil Biology and Biochemistry, 69, 251-264.
Cheng, G. Z., Fray, D. J., and T. W. Farthing, 2000. Direct electrochemical reduction of titanium dioxide to titanium in molten calcium chloride. Nature, 407 (6802), 361-364.
Chenu, C., Le Bissonnais, Y., and D. Arrouays. 2000. Organic matter influence on clay
wettability and soil aggregate stability. Soil Science Society of America Journal, 64 (4), 1479-1486.
Chung, J., Rogers, D., and C. M. Watkins. 2014. Estimating severity of seismically induced
landslides and lateral spreads using threshold water levels." Geomorphology, 204, 31-41.
Deckers, J., Driessen, P., Nachtergaele, F., and O. Spaargaren. 2001. Tropical Soils in the
Classification Systems of USDA, FAO and WRB. In: Evolution of Tropical Soil Science: Past and Future (Workshop, Brussels, 6 March 2002), Royal Academy of Overseas Sciences: 79-94.
DIN ISO 90.60 14869-2: 2002. Soil quality-Dissolution for the determination of total element
content-Part 2: Dissolution by alkaline fusion, International Organizatuion for Standardization, Geneva, Switzerland.
Dultz, S. 2000. Effects of silicate weathering and lessivage on K‐content in forest soils
derived from Pleistocene sediments. Journal of Plant Nutrition and Soil Science, 163 (3), 299-305.
Ebinger, M. H., and D. G. Schulze. 1990. The influence of pH on the synthesis of mixed Fe-
Mn oxide minerals. Clay Minerals, 25 (4), 507-518. Fall, M., Azzam, R., and C. Noubactep. 2006. A multi-method approach to study the stability
of natural slopes and landslide susceptibility mapping. Engineering Geology, 82 (4), 241-263.
FAO/ISRIC/ISSS. 1998. World reference base for Soil Resources. World Soil Resources Report
No. 84, FAO, Rome, pp. 88. Fedoroff, N. 1997. Clay illuviation in Red Mediterranean soils. Catena, 28 (3), 171-189. Galeandro, A.,. Doglioni, A., Simeone, V., and J. Šimůnek. 2011. Analysis of infiltration
processes into fractured and swelling soils as triggering factors of landslides. Environmental Earth Sciences, pp.1-13.
Gorokhovich, Y., Doocy, S., Walyawula, F., Muwanga, A., and F. Nardi. 2013. Landslides in
Bududa, Eastern Uganda: Preliminary Assessment and Proposed Solutions." In Landslide Science and Practice, pp. 145-149.
GoU. 1996. Mount Elgon National Park Biodiversity Report, Forest Department, Kampala,
Uganda.
61
Greene, R. S. B., Rengasamy, P., Ford, G. W., Chartres, C. J., and J. J. Millar. 1988. The effect
of sodium and calcium on physical properties and micromorphology of two red‐brown earth soils. Journal of Soil Science, 39 (4), 639-648.
Hai, N. Q., and K. Egashira. 2008. Clay mineralogy of ferralitic soils derived from igneous
rocks in Vietnam. Clay Science, 13 (6), 189-197. Hieronymus, B., Kotschoubey, B., Boulegue, J., Benedetti, M., Godot, J. M., and W.
Truckenbrodt. 1990. Aluminum behaviour in some alterites of eastern Amazonia (Brazil). Chemical Geology, 84 (1), 74-77.
Highland. L., and P. T. Bobrowsky. 2008. The landslide handbook: a guide to understanding
landslides. Reston, VA, USA: US Geological Survey. Huabin, W., Gangjun, L., Weiya, X., and W. Gonghui. 2005. GIS-based landslide hazard
assessment: an overview. Progress in Physical Geography, 29, 548-567. Hua-xi, G., and Y. Kun-long. 2014. Study on spatial prediction and time forecast of landslide.
Natural Hazards, pp. 1-14. Isabirye, M., Mwesige, D., Sali, H., Magunda, M., and J. Lwasa. 2004. Soil resouce
information and linkages to agricultural production. Uganda Journal of Agricultural Sciences, 9, 215-22.
Karathanasis, A. D. 2006. Soil Mineralogy. Land Use and Land Cover, from Encyclopedia of
Life Support Systems (EOLSS), Developed under the Auspices of the UNESCO, Eolss Publishers, Oxford, UK,[http://www. eolss. net].
Kirschbaum, D. B., Adler, R., Hong, Y., Hill, S., and A. Lerner-Lam. 2010. A global landslide
catalog for hazard applications: method, results, and limitations. Natural Hazards, 52, 561-575.
Kitutu, M. G., Muwanga, A., Poesen, J., and J. A. Deckers. 2009. Influence of soil properties
on landslide occurrences in Bududa district, Eastern Uganda. African Journal of Agricultural Research, 4, 611-620.
Kitutu, M. G., Poesen, A. M. J., and J, Deckers. 2011. Farmer's perception on landslide
occurrences in Bududa District, Eastern Uganda. African Journal of Agricultural Research, 6 (1), 7-18.
Kjaergaard C., de Jonge, L.W., Moldrup, P., and P. Schjønning. 2004. Water-dispersible
colloids: effects of measurement method, clay content, initial soil matric potential, and wetting rate. Vadose Zone Journal, 3, 403-412.
Knapen, A. 2003. Influence of soil properties, landuse and deforestation on the occurrence
of landslides in Manjiya county, Mount Elgon area, East Uganda. Phd Thesis.
62
Knapen, A., Kitutu, G., Poesen, M., Breugelmans, J., Deckers, J., and A. Muwanga. 2006.
Landslides in a densely populated county at the footslopes of Mount Elgon (Uganda): characteristics and causal factors." Geomorphology, 73, 149-165.
Laffan, M. D., Daly, B. K., and J. S. Whitton. 1989. Soil patterns in weathering, clay
translocation and podzolisation on hilly and steep land at Port Underwood, Marlborough Sounds, New Zealand: classification and relation to landform and altitude. Catena, 16 (3), 251-268.
Landon, J. R. 2014. Booker tropical soil manual: a hand book for soil survey and agricultural
land evaluation in the tropics and subtropics. Routledge. Leroy, S. A., and R. Gracheva. 2013. Historical Events. In Encyclopedia of Natural Hazards
Springer Netherlands, pp. 452-471. Loveland, P., and J. Webb. 2003. Is there a critical level of organic matter in the agricultural
soils of temperate regions: a review. Soil and Tillage Research, 70 (1), 1-18. Mehra, O.P., and M. L. Jackson. 1960. Iron Oxide Removal from Soils and Clay by Dithionite-
Citrate Systems buffered with Sodium Carbonate. Clays and Clay Minerals, 7, 317-327.
Michael, H., Elke. D., and G. Agola. 2003. International Centre for Research in Agroforestry
Kenya. Miedema, R., Koulechova, I. N., and M. I. Gerasimov., 1999. Soil formation in Greyzems in
Msilimba, G. G., and P. J. Holmes. 2010. Landslides in the Rumphi District of Northern
Malawi: characteristics and mechanisms of generation. Natural Hazards, 54, 657-677.
Mugagga, F., Buyinza , M., and V. Kakembo. 2010. Livelihood diversification strategies and
soil erosion on Mount Elgon. Eastern Uganda: a socio-economic perspective. Environmental Research Journal, 4, 272–280.
Mugagga, F., Kakembo, V., and M. Buyinza. 2011. A characterisation of the physical
properties of soil and the implications for landslide occurrence on the slopes of Mount Elgon, Eastern Uganda. Natural Hazards, 60, 1113-1131.
Mugagga, F., Kakembo, V., and M. Buyinza. 2012. Land use changes on the slopes of Mount
Elgon and the implications for the occurrence of landslides. Catena, 90, 39- 46.
63
Mujinya, B. B., Van Ranst, E., Verdoodt, A., Baert, G., and L. M. Ngongo. 2010. Termite bioturbation effects on electro-chemical properties of Ferralsols in the Upper Katanga (DR Congo). Geoderma, 158 (3), 233-241.
Murphy, C. P. 1986. Thin Section Preparation of Soils and Sediments. A B Academic
Publishers, Berkhamsted. National Environment Management Authority. 1997. District State of Environment Report,
Mbale. Uganda, pp. 128. Neufeldt, H., de Oliveira Schneider, M., and W. Zech. 1999. Oxisol development along a
compound catena of the Araguari river, Central Brazil. Sustainable Land Management for the Oxisols of the Latin American Savannas. CIAT publication, (312), 10-21.
Ollier, C.D., and J. F. Harrop. 1959. Memoirs of the research division, series1-soils. The soil
of the Eastern province of Uganda. Kawanda Research Station, Kampala. Soils Report No. 2.
Pankow, K. L., Moore, J. R., Hale, J. M., Koper, K. D., Kubacki, T., Whidden, K. M., McCarter,
M. K., Stokes, P. J., Levine, R., and K. W. Flessa. 2014 Massive landslide at Utah copper mine generates wealth of geophysical data. GSA Today. pp.24.
Preetz, H., Altfelder, S., and J. Igel. 2008. Tropical soils and landmine detection—an
approach for a classification system. Soil Science Society of America Journal, 72 (1), 151-159.5-364.
Quénard, L., Samouëlian, A., Laroche, B., and S. Cornu. 2011. Lessivage as a major process of
soil formation: A revisitation of existing data. Geoderma, 167, 135-147. Rawling, J. E. 2000. A review of lamellae. Geomorphology, 35 (1-2), 1-9. Reedman, J.H. 1973. Potash Ultra-fenites at the Butiriku carbonatite complex in South-East,
Uganda. University of Leeds, Research Institute of African Geology. 17th Annual Report. pp. 78-81.
Sanchez, P. A., Sachs, J., Remans, R., Smukler, S., Winowiecki, L., Andelman, S. J., and K. G.
Cassman. 2010. Monitoring the world's agriculture. Nature, 466 (7306), 558-560. Schuster, R. L., and L. M. Highland. 2003. Impact of landslides and innovative landslide-
mitigation measures on the natural environment. Geologic Hazards Team, US Geological Survey, Denver, Colorado.
Schwertmann, U. 1985. The effect of pedogenic environments on iron oxide minerals.
Advanced Soil Science, 1, 171-200.
64
Schwertmann, U., and R.M. Taylor. 1989. Iron oxides. In: Dixon, J.B., Weed, S.B. (Eds.), Minerals in Soil Environments, 2nd, Ed. Soil Science Society of America, Madison, Wisconsin, USA, pp. 379 –438.
Scott, P. 1994. An assessment of natural resource use by communities from Mt. Elgon
National Park. UNDP/Technical Report No. 15. Seta, A. K., and A. D. Karathanasis. 1996. Water dispersible colloids and factors influencing
their dispersibility from soil aggregates. Geoderma, 74 (3), 255-266. Sharma, J. P., Landey, R. J., Kalbande, A. R. and Mandal. C. 2001a, Characteristics and
classification of soils of Kathiawar region of Gujarat as influenced bytopography. Agropedology, 11, 83-90.
Smyth, C. G., and S. A. Royle. 2000. Urban landslide hazards: incidence and causative factors
in Niteroi, Rio de Janeiro State, Brazil. Applied Geography, 20 (2), 95-118. Sposito, G. 1989. The Chemistry of Soils. New York: Oxford University Press. pp. 277. Stoops, G. 2003. Guidelines for analysis and description of soil regolith thin sections. Soil
Science Society of America, Inc. Stoops, G. 2010. Interpretation of micromorphological features of soils and regoliths.
Elsevier. Synnott, T. J. 1968. Working Plan for Mount Elgon Central Forest Reserve.1st Revision. Period
1968 to 1978. Uganda Forest Department, Entebbe. Szott, L. T., Palm, C. A., and P. A. Sanchez. 1991. Agroforestry in acid soils of the humid
tropics. Advances in Agronomy, 45, 275-301. Tao, J., and A. P. Barros. 2014. Coupled prediction of flood response and debris flow
initiation during warm-and cold-season events in the Southern Appalachians, USA. Hydrology and Earth System Sciences, 18, 367-388.
Tilley, C. 2010. Interpreting Landscapes: Geologies, Topographies, Identities; Explorations in
Landscape Phenomenology 3 (Vol. 3). Left Coast Press. Tonneijck, F. H., and A. G.Jongmans. 2008. The influence of bioturbation on the vertical
distribution of soil organic matter in volcanic ash soils: a case study in northern Ecuador. European Journal of Soil Science, 59 (6), 1063-1075.
Torrent, J. 1994. Iron oxides in Mediterranean soils: properties and influence on soil
behaviour. Vol 8a, pp. 2-14 in Transactions, 15th World Congress of Soil Science. ISS, The Mexican Society of Soil Science; Acapulco.
Van Breemen, N., and P. Buurman. 2002. Ferralitization. Soil Formation, 2, 291-312.
65
van der Zee, C., Roberts, D. R., Rancourt, D. G., and Slomp, C. P. 2003. Nanogoethite is the
dominant reactive oxyhydroxide phase in lake and marine sediments. Geology, 31 (11), 993-996.
van Heist, M. 1994. Land Unit Map of Mount Elgon National Park, IUCN technical report.
Unpublished. Van Ranst, E. 1991. Soils of the tropics and subtropics: geography, classification, properties
and management. Lecture notes. Ghent University, Belgium. pp 292. Van Reeuwijk, L. P. 1993. Procedures for soil analysis. 6th edition, International Soil Referene
and Information Centre (ISRIC) FAO, Wageningen, The Netherlands. Van Wambeke. A. 1992. Soils of the tropics: properties and appraisal. McGraw Hill. Walkley, A., and I. A. Black. 1934. An examination of digestion methods for soil organic
matter determination and a proposed modification of the chronic acid titration. Soil Science, 37, 29-38.
Wilkinson, M. T., Richards, P. J., and G. S. Humphreys. 2009. Breaking ground: pedological,
geological, and ecological implications of soil bioturbation. Earth-Science Reviews, 97 (1), 257-272.
WRB. 2006. World reference base for soil resources 2006. World Soil Resources Reports,
(103). Yalcin, A. 2007. The effects of clay on landslides: A case study. Applied Clay Science, 38 (1),
77-85. Zaidel’man, F. R. 2007. Lessivage and its relation to the hydrological regime of soils. Eurasian
Soil Science, 40 (2), 115-125.
I
APPENDICES
Appendix I: FAO soil profile description
Appendix I-1: Profile P1
General site information
Author(s): Joseph Tamale and Amaury Defrère
Date: 07/09/2013
Profile number: UG/BU/BUM-P1
Sprofile description status: 2
Location: Bumaraka site is found in Uganda, Bududa district, Bulucheke subcounty, Bumwalukani parish
Atmospheric climate: The mean maximum and minimum temperatures are 23°C and 15 °C respectively. The wettest period of the year is from March till October, while the dry season starts from November till February with a short dry period around July. The mean annual rainfall is 1500 mm.
Present weather conditions: Sunny/clear (SU) at the time the profile description was made.
Former weather conditions: Rainy without heavy rain in the last 24 hours (WC4)
Landform and topography
Major landform: High gradient hill (TH)
Position: Crest (summit, CR)
Slope gradient and orientation: Flat
Land use and vegetation
Land use: Settlement, transport (ST)
Human influence: Ploughing (PL), Levelling (LV)
Vegetation: Short grass (HS)
II
Soil horizon description for profile 1
Horizon Description
C1 (0-13 cm) 7,5YR 4/6 (brown) clay loam, fine to medium sized granules, very friable when moist, soft consistence when dry, moist soil water status, common very fine to medium roots, many biological activity (earthworms and ants), medium to high porosity, very fine voids (vughs), many very fine to medium pores, clear and wavy boundary
Apb (13-35 cm) 5YR 2/3 (dark brown) sandy clay loam, few rock fragments (coarse gravel), fine to medium sized granules, friable when moist, soft consistence when dry, moist soil water status, many very fine to medium roots, common biological activity (ants and other insects), medium to high porosity, common very fine to fine pores, very fine voids (vughs), some brick pieces and charcoal artefacts, clear and wavy boundary
BC (35-140 cm) 5YR 3/6 (dark reddish brown) clay, common rock fragments (stones+coarse gravel), angular-blocky structure, firm when moist, hard consistence when dry, moist soil water status, common very fine to fine roots, common biological activity (termite nests and ant barrels), many clay coatings, compacted not cemented, fine voids (fauna-channels), low porosity, very few very fine pores, diffuse, broken and discontinuous boundary
CB (140-225 cm) 5YR 4/4 (dull reddish brown) sandy clay loam, dominant rock (saprolite) fragments, rock structure intermingled with angular blocky structured soil, extremely hard consistence when dry, very firm when moist, moist soil water status, no roots, very low porosity, fine voids (channels), very few fine pores, few clay coatings, broken compactation, abrupt, irregular boundary
R (225+ cm) Rock layer, clear and broken boundary
Profile P1 with corresponding horizons and exact location of Kubiëna box sampling
III
Appendix I-2: Profile P2
General site information
Author(s): Joseph Tamale and Amaury Defrère
Date: 17//08/2013
Profile number: UG/BU/BUN-P2
Soil profile description status: 2
Location: Bunamurembe site is found in Uganda, Bududa district, Bulucheke subcounty, Bumwalukani parish
Atmospheric climate: The mean maximum and minimum temperatures are 23°C and 15 °C respectively. The wettest period of the year is from March till October, while the dry season occurs from November till February with a short dry period around July. The mean annual rainfall is 1500 mm.
Present weather conditions: Sunny/clear (SU) at the time the profile description was made.
Former weather conditions: Heavier rain for some days (WC5)
Landform and topography
Major landform: High gradient hill (TH)
Position: Middle slope (MS)
Slope form: Concave + convex (CV)
Slope gradient and orientation: 35%
Land use and vegetation
Land use: Rainfed arable cultivation (AA4) and non-irrigated cultivation (AP1)
Crops: Coffee (LuCo), avocado , beans (PuBe)
Human influence: Ploughing (PL)
IV
Soil horizon description of profile 2
Horizon Description
Ap (0-15 cm) 7,5YR 4/4 (brown) clay, fine to medium sized granules, friable when moist, soft consistence when dry, moist soil water status, many very fine to medium roots, many biological activity (earthworms and ants) , medium to high porosity, common very fine to medium pores, clear and wavy boundary
A (15-70 cm) 5YR 2/3 (very dark reddish brown) clay, fine to medium sized granules, friable when wet, slightly hard consistence when dry, moist soil water status, common fine to medium roots, common biological activity, medium to high porosity, common fine to medium pores, some plastics and charcoal artefacts, clear and wavy boundary
B1t (70-155 cm) 2,5YR 3/4 (dark reddish brown) clay, fine to medium sized granules, friable when wet, slightly hard consistence when dry, moist soil water status, common fine to coarse roots, common biological activity (termite nests), medium to high porosity, common fine to medium pores, some charcoal artefacts, smooth and diffuse boundary
B2t (155-250 cm) 5YR 4/4 (dull reddish brown) clay, fine to medium sized granules, friable when wet, soft consistence when dry, moist soil water status, common fine to coarse roots, charcoal artefacts, low porosity, few fine pores, compacted and not cemented, clear and wavy boundary
BC (250+ cm) 7,5YR 4/6 (brown) and clay, granular to massive structure, hard when dry, very friable consistence when wet, moist soil water status, very few fine to coarse roots, termite nests, very low porosity, very few very fine pores, compacted and not cemented, clay coatings, clear and wavy boundary
Profile P2 with corresponding horizons and exact location of Kubiëna box sampling.
V
Appendix I-3: Profile P3
General site information Author(s): Joseph Tamale and Amaury Defrère
Date: 19/08/2013
Profile number UG/BU/MAT-P3
Soil profile description status 2
Location Mataya site is found in Uganda, Bududa district, Nakatsi subcounty, Bumwalukani parish, on the boundary with Bulucheke subcounty, 3 meters from Zuzu-river.
Atmospheric climate The mean maximum and minimum temperatures are 23° and 15 °C respectively. The wettest period of the year is from March till October, while the dry season occurs from November till February with a short dry period around July. The mean annual rainfall is 1500 mm.
Present weather conditions Sunny/clear (SU) at the time the profile description was made.
Former weather conditions Heavier rain for some days (WC5)
Landform and topography
Major landform Valley-floor (LV)
Position Bottom (BO)
Land use and vegetation
Land use Perennial field cropping (AP1)
Crops Sugar cane (OtSc)
Human influence Burning (BR) and Ploughing (PL)
Parent material Unconsolidated fluvial material (UF1)
Age of the land surface Young (Yn)
VI
Soil horizon description of profile 3
Horizon Description
Ap (0-15 cm) 10 YR 3/4 (dark brown) sandy clay loam, fine to coarse gravel, abundant small to big rocks, granular structure, friable when moist, soft consistence when dry, moist soil water status, many very fine to medium roots, many biological activity, charcoal artefacts, medium to high porosity, fine to coarse voids (vughs), common very fine to medium pores, smooth and clear
C1 (15-29 cm) 7,5YR 4/1 (brownish gray) sandy loam, dominant rock fragments, fine to medium sized gravel, granular structure, loose when moist, loose consistence when dry, moist soil water status, few very fine to fine roots, few biological activity (earthworm channels), high porosity, many fine to medium pores, fine voids (vughs), continuous compaction, smooth and clear boundary
C2 (29-54 cm) 7,5YR 3/4 (dark brown) sandy loam, fine to medium gravel, granular structure, loose when moist, loose consistence when dry, moist soil water status, very few, very fine roots, very high porosity, many fine to medium pores, fine voids (vughs), clay coatings on gravels, smooth and diffuse boundary
C3 (54-64 cm) 10YR 5/4 (dull yellowish brown) sandy clay loam, granular structure, friable when miost, soft consistence when dry, moist soil water status, common fine to coarse roots, very fine to fine voids (vughs), high porosity, common pores, common clay coatings, wavy abrupt boundary
C4 (64+ cm) 10YR 5/6 (yellowish brown) sandy loam, dominant rock structure, fine to medium gravel, granular structure, loose when dry, loose consistence when moist, wet water status, very few very fine roots, high porosity, many fine to coarse pores, fine to coarse voids (vughs), clay coatings, clear and wavy boundary
Profile P3 with the corresponding horizons and exact location of Kubiëna box sampling.
VII
Appendix II: XRD patterns of clay powders for the respective horizons with all d-spacings indicated in nm
Appendix II-1: Profile P1
VIII
Appendix II-2: Profile P2
IX
Appendix II-3: Profile P3
X
Appendix III: Oriented XRD patterns of the K+ and Ca2+-saturated clay samples of profile P2 (All d-spacings are indicated in nm)
Appendix III-1: Horizon Ap of P2
XI
Appendix III-2: Horizon A of P2
XII
Appendix III-3: Horizon B1t of P2
XIII
Appendix III-4: Horizon B2t of P2
XIV
Appendix IV: XRD patterns of rock powders (all d-spacings are indicated in nm)