Martin Fey

Unravelling the effects of soil properties on water infiltration : segmented quantile regression on a large data set from arid south-west Africa

Relationships were sought between infiltrability and the properties of hundreds of surface soils (pedoderms) sampled across Namibia and western South Africa. Infiltrability was determined using a laboratory method, calibrated against a rainfall simulator, which measures the passage of a suspension of soil in distilled water through a small column packed with the same soil. Other properties determined were EC, pH, water-soluble cations and anions, ammonium acetate-extractable cations, organic C, total N, a 7-fraction particle size distribution, water-dispersible silt and clay, and clay mineral composition. Our objective was to ascertain whether general principles pertaining to infiltrability can be deduced from an analysis of a wide diversity of soils. To achieve this we compared correlation analysis, generalised linear models (GLMs), and generalised additive models (GAMs) with a segmented quantile regression approach, in which parametric regression lines were fitted to the 0.9 and 0.1 quantile values of equal subpopulations based on the x variable. Quantile regression demarcated relational envelopes enclosing four-fifths of the data points. The envelopes revealed ranges for soil properties over which infiltrability is potentially maximal (spread over a wide range) or predictably minimal (confined to small values). The r2 value of the 0.9 quantile regression line was taken as an index of reliability in being able to predict limiting effects on infiltrability associated with a variety of soil properties. Prediction of infiltration was most certain from textural properties, especially the content of water-dispersible silt (r2 = 0.96, n = 581), water-dispersible clay (0.88, n = 581), very fine sand (0.86, n = 174), and medium sand (0.84, n = 174). Chemical properties such as EC, sodium status, organic C content, and clay mineralogy were less clearly related to infiltrability than was texture. The role of fine-particle dispersion in blocking pores was highlighted by the stronger prediction in all statistical analyses provided by the water-dispersible as opposed to total content of silt and clay. All the statistical analyses revealed a probable skeletal role of medium and fine sand fractions in shaping pores and a plasmic (mobile) role of finer fractions in blocking pores. A noteworthy discovery was an apparent switch in role from skeletal to plasmic at a particle diameter of about 0.1 mm (i.e. between fine and very fine sand).

Transformation of thicket to savanna reduces soil quality in the Eastern Cape, South Africa

Xeric succulent thicket in the Eastern Cape, South Africa has been used for farming goats since the early 1900s. This habitat is characterised by a dense cover of the succulent bush Portulacaria afra and by a warm, semi-arid climate with evenly distributed annual rainfall of 250–400 mm. Heavy browsing by goats results in the loss of P. afra and transforms the thicket to an open savanna dominated by annual grasses. Eight fence-line comparisons between thicket and savanna were used to investigate differences in soil quality associated with the vegetation change. Composite soil samples were taken to a depth of 10 cm from 1 ha plots on either side of the fence-line. Associated with the change from thicket to savanna, a significant decrease (paired t-test, P < 0.05) was found in total C (respective means of 5.6 vs. 3.0%), total N (0.33 vs. 0.24%), labile C (2.8 vs. 1.5%), CO2 flux (1.9 vs. 0.5 µmol m−2 s−1), soil respiration in the laboratory (144 vs. 79 ng C kg−1 s−1), (NH4)OAc-extractable Mg (55 vs. 28 mmolc kg−1), and laboratory infiltration rate (51 vs. 19 mm h−1). In the same direction there was a similarly significant increase in modulus of rupture (16 vs. 34 kPa), water-soluble Ca (2.3 vs. 3.4 mmolc kg−1) and pH (6.7 vs. 7.7). The soil C content of 5.6% in thicket is surprisingly high in this warm, semi-arid climate and suggests that the dense P. afra bush strongly regulates soil organic matter through microclimate, erosion control, litter quantity and perhaps chemistry. Savanna soils had a greater tendency to crust (as evident in a lower rate of laboratory infiltration and greater modulus of rupture) than thicket soils. This was attributed to their lower organic matter content, which probably reduced aggregate stability. Savannas are likely to be more prone to runoff and erosion not only because of sparser vegetation but also because of a decline in soil quality.

Ecosystem carbon storage under different land uses in three semi-arid shrublands and a mesic grassland in South Africa

Carbon (C) storage in biomass and soils is a function of climate, vegetation type, soil type and land management. Carbon storage was examined in intact indigenous vegetation and under different land uses in thicket (250–400 mm mean annual precipitation), xeric shrubland (350 mm), karoo (250 mm), and grassland (900–1200 mm). Carbon storage was as follows: (i) mean soil C (0–50 cm): thicket (T) = grassland (G) > xeric shrubland on Dwyka sediments (XS) > xeric shrubland on dolerite (XSD) > karoo (K) (168, 164, 65, 34 & 26 t ha−1, respectively); (ii) mean root C: T > G > XS = XSD (25.4, 11.4, 7.2 & 7.1 t ha−1); (iii) mean above-ground C including leaf litter: T>XS>G>K> XSD (51.6, 12.9, 2.0, 1.7 & 1.51 ha−1). Carbon stocks in intact indigenous vegetation were related more to woodiness of vegetation and frequency of fire than to climate. Biomass C was greatest in woody thicket and soil C stocks were greatest in thicket and grassland. Total C storage of 245 t ha−9 in thicket is exceptionally high for a semi-arid region and is comparable with mesic forests. Soil C dominated ecosystem C storage in grassland and was influenced more by soil parent material than land use. The semi-arid sites (xeric shrubland and thicket) were more sensitive to effects of land use on C storage than the grassland site. Effects of land use on C stocks were site- and land use-specific and defied prediction in many instances. The results suggest that modelling of national C stocks would benefit from further research on the interactions between C storage, land use, and soil properties.

Declining soil quality in South Africa: effects of land use on soil organic matter and surface crusting

Soil conservation in South Africa has historically focussed on preventing soil erosion. Effective maintenance of the soil resource requires, in addition to erosion control, an understanding of how land use practices affect more subtle indicators of soil quality. This review outlines how land use in South Africa can rapidly result in a marked reduction in soil organic matter (SOM) content and a greater tendency of soils to crust. Removal of a cover of vegetation whether by ploughing, grazing or burning tends to reduce SOM due to reduced organic matter inputs and enhanced soil microbe activity. Loss of SOM, particularly from the top few centimetres of soil (named here the pedoderm) has a disproportionately large effect on soil infiltrability and nutrient supply. The mineralogy of the clay fraction also has great bearing on the response of soil to land use effects. The unexpected role of quartz in soil dispersion and crusting in South Africa has only recently been unveiled. Apart from SOM effects, land use can lead to subtle changes in soil chemistry. Plantation forestry has resulted in an increase in soil nitrate in many areas, possibly due to greater mineralisation under forests than grasslands. Annual burning in the Kruger National Park bushveld has been shown to increase clay dispersibility and crusting of the pedoderm, which was ascribed to a reduction in electrical conductivity and SOM as well as an increase in the exchangeable sodium percentage. Soil quality is a multifaceted concept. One aspect stands out, however, as critical and that is the conservation and replenishment of nitrogen which is all important for retaining humus and maintaining soil quality.

Soil infiltrability as a driver of plant cover and species richness in the semi-arid Karoo, South Africa

Data from 199 plots in the semi-arid Karoo showed that relationships between soil infiltrability and plant cover/species richness, as depicted by boundary lines, yielded ecological insights not evident if only commonly measured soil properties such as pH, electrical conductivity and the content of clay, silt, sand, nitrogen and carbon were considered. For example, the common grass Stipagrostis obtusa, herb Lepidium africanum and shrub Pentzia incana showed potentially maximal cover at high, low and intermediate infiltrability, respectively (r2 > 0.65 for boundary lines derived by segmented quantile regression), but did not show distinct boundary lines for sand, silt or clay content data (r2 < 0.5). Potentially maximal species richness was revealed under soil conditions of low infiltrability, high nitrogen content and low pH. Distinct boundary lines suggested that the drivers of species richness at any particular point in the Karoo landscape may operate in opposing directions simultaneously.

SOIL carbon and nitrogen in five contrasting biomes of South Africa exposed to different land uses

Stocks of soil C to a depth of 50 cm in untransformed, indigenous veld ranged from 21 t ha-1 in karoo to 168 t ha-1 in thicket and stocks of N ranged from 3.41 ha-1 in karoo to 12.8 t ha-1 in grassland. Mean soil C in thicket (5.6%, 0–10 cm) was approximately five times greater than expected for a semi-arid region. Removal of vegetation due to cultivation, grazing or burning reduced soil C and N at all sites. Soil C under intact thicket was greater than at sites degraded by goats (71 vs 40 t ha-1, 0–10 cm). Restoration of thicket could potentially sequester -40 t C ha-1. The sale of this sequestered carbon to the international market may make restoration of thousands of hectares of degraded thicket financially feasible. Soil C under plant cover was greater than In exposed soil in renosterveld (28 vs 15 t ha-1) and in karoo (7 vs 5 t ha-1). Parent material was also related to soil C content. In grassland, soil C was greater in dolerite-derived than sandstone-derived soils (54 vs 271 ha-1); and in bushveld it was greater in basalt-derived than granite-derived soils (28 vs 14 t ha-1 in unburnt plots). Annual burning in bushveld reduced soil C, particularly at the surface. Soil C in the 0–1 cm layer of unburnt plots was 2 to 3 times greater than in burnt plots.

Effects of sodium sulphate, sodium chloride and manganese sulphate on kikuyu (Pennisetum clandestinum) growth and ion uptake

Irrigation with saline water is increasingly practised yet an understanding of plant response to different salt types remains largely elusive. Our study investigated the effects of irrigation with simulated effluent containing three salts on the growth and composition of a common South African pasture grass. We treated pots of kikuyu (Pennisetum clandestinum Hochst, ex Chiov.) in a controlled environment with solutions of sodium sulphate (Na2SO4) or chloride (NaCI) at saturated soil solution electrical conductivity (EC) of 0, 5,10 and 20 dS m-1, combined with 0, 900 or 1800 mg kg-1 Mn as MnSO4. Kikuyu yields in the Na2SO4 treatments were significantly lower than in corresponding NaCI treatments. This could be explained by increased ion imbalance and osmotic stress as a result of higher ionic strength and a greater concentration of neutral ion pairs in the Na2SO4 system at similar EC. An apparent Na2SO4-induced Ca deficiency was attributed to suppressed Ca2+ activity through ion pairing at high ionic strengths. Under the conditions of this experiment ionic strength appeared to be more effective than EC as a measure of salinity stress when comparing the effects of SO4 and CI on growth of kikuyu.

Hydrosalinity fluxes in a small scale catchment of the Berg River (South Africa).

The occurrence of dryland salinity is widespread throughout semi-arid regions of the world. The sources of salts may be either rock weathering or rain deposition. Clearing of natural scrubland to make way for cultivated crops and pastures may also change the water balance, trigger salt mobilization and increase the salinity of water resources. These processes are suspected to be the main cause for salinization of the Berg river catchment (South Africa). The objective of this study was to determine the hydrosalinity fluxes associated with overland and subsurface (vadose zone) flow for different soils and land uses. For this purpose, the following data were collected during 2006 in a typical small scale catchment (~40 ha) located near the town of Riebeek West: weather data, hydrological and water quality measurements, soil water contents and chemistry. The climate of the area is Mediterranean with winter rainfall of about 350 mm a. The chemical speciation of water and soil solution in the catchment is conservative, with Na and Cl being the dominant ions. Soil water and salt contents varied seasonally. Due to the typical low intensity of rainfall, the fluxes of salts during individual runoff events were steady. Fluctuations in salinity due to local processes were buffered at a catchment scale. Uncultivated (bare) soil produced more runoff and higher salinity compared to pasture land. Overland flow varied between 3 and 18% of rainfall, mobilizing up to 24 g m of salts during 2006, depending on soil properties, slopes, rainfall intensity and duration, and antecedent moisture conditions. Subsurface fluxes of water and salts were estimated with the HYDRUS-2D model. Management practices at farm scale are required in order to reduce salt mobilization and salinization at catchment scale.

Soils of South Africa: Animals in soil environments

INTRODUCTION Although soil scientists are aware of the role that animals play in forming soil and the reciprocal role of soil in providing for animals, they tend to think of animals as being of somewhat less importance than geology, climate, topography and vegetation in shaping the character of soil over time. Since human disturbance has greatly diminished populations of wild animals interacting with soil we tend to underestimate the intensity of soil-animal relationships in undisturbed ecosystems. On the other hand there are some who consider the upper part of soil primarily as a biomantle (Johnson, 1990; Johnson et al ., 2005). South African soils are – or have been – affected by animals in many different taxa exhibiting an extraordinary range of body size, population and activity. Some animals that lack size often make up for it in numbers. Soil in turn may benefit animals in a variety of ways. It functions as shelter, breeding ground, waste dump, source of food, dietary supplement and means of bodily hygiene. In exploiting soil, animals change it. Mixing, segregation and nutrient cycling take place. The consequences for porosity, aeration, water storage, drainage, density, erodibility and nutrient status can be profoundly important, both ecologically and economically. In deciding on material for this chapter we have chosen conspicuous examples and have not attempted to describe a full or balanced spectrum of animal-soil interactions. The objective is to heighten awareness of how fauna affect soil properties and how soil affects the welfare of animals.

Relationships between soil particle size fractions and infiltrability

Abstract The influence of particle size fractions on infiltrability was investigated in soils sampled across Namibia and western South Africa. Infiltrability was determined using a laboratory technique calibrated with rainfall simulation, which measures the passage of a suspension of soil particles through a packed soil column. Water-dis-persible soil particle size fractions were determined using a high definition digital laser particle size analyser. Total (calgon-dispersed) particle size fractions were determined by hydrometer. Dispersion of soil particles resulting in crust formation on the soil surface appeared to be a main mechanism reducing infiltrability. Water-dispersible clay and fine silt determined by laser analyser showed higher correlation with infiltrability (r2 = −0.43 for clay and −0.47 for fine silt) than total clay and fine silt determined by hydrometer (r2 = −0.30 and −0.28, respectively). Clay, fine silt, coarse silt, very fine sand and fine sand fractions (<120 µm) showed a probable plasmic role in soil crusts. At a content of these fractions > 5% infiltrability was inevitably restrained. The 120–200 µm fraction showed no clear relationship with infiltrability. It played either a plasmic or skeletal role, depending on its ratio to the <120 µm and >200 µm fractions. Fine, medium and coarse sand fractions (>200 µm) showed a probable skeletal role in soil crusts, i.e. forming pores that enhanced infiltrability. At levels >50% of these fractions, infiltrability was potentially maximal. This potentially maximal infiltrability was also explained by the concomitant decrease in plasmic fraction content with an increase of the skeletal fraction.