Introduction
Soils in sub-Saharan Africa are in most cases low in nutrient availability (Smaling and Braun, Reference Smaling and Braun1996). Such low nutrient availability in soils is known to lead to low plant biomass production (Du et al., Reference Du, Terrer, Pellegrini, Ahlström, Van Lissa, Zhao, Xia, Wu and Jackson2020). Nutrient availability is not only determining the biomass production in grasslands but also the nutritional value of the plant biomass as food for example migratory herd of large herbivores (McNaughton, Reference McNaughton1990) leading to migration to fodder with specific nutrition content. A specific and well-studied area for such migratory herds of large herbivores to meet their nutritional demands is the Serengeti National Park in Tanzania (Seagle and McNaughton, Reference Seagle and McNaughton1992). There is ample data about the different nutrient content of grassland biomass within the Serengeti National Park in Tanzania (McNaughton, Reference McNaughton1990; Seagle and McNaughton, Reference Seagle and McNaughton1992). However, there is less knowledge about the soil nutrient gradient from the Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti National Park in Tanzania (Eckmeier et al., Reference Eckmeier, Kübler, Meya and Mathai Rucina2020) being the basis for the well-known differences in nutrient content of grassland biomass within this area, beside elements like nitrogen (N) or phosphorus (P) (Ruess and McNaughton, Reference Ruess and McNaughton1987; Ruess and Seagle, Reference Ruess and Seagle1994; Anderson et al., Reference Anderson, Ritchie and McNaughton2007).
High content of silicon (Si) and calcium (Ca) in plant material is known to deter feeding by for example insects (Schaller et al., Reference Schaller, Hodson and Struyf2017) or by the megafauna as shown for Si in the Serengeti (McNaughton and Tarrants, Reference McNaughton and Tarrants1983; McNaughton et al., Reference McNaughton, Tarrants, McNaughton and Davis1985). Hence, those elements may decrease the quality of grassland biomass as fodder for the megafauna.
The soil nutrient availability is an important control on plant nutrient uptake and subsequent potential nutrient accumulation of the grassland biomass within the Serengeti National Park in Tanzania. As volcanic ash material, which is well-known for its high nutrient content but also high in toxic aluminium (Jones and Gislason, Reference Jones and Gislason2008), from Ngorongoro was spread in the whole area during past volcanic eruptions (Makongoro et al., Reference Makongoro, Vegi, Vuai and Msabi2022) there might be a nutrient and toxicant gradient from the Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti National Park in Tanzania. However, only few data exist on the nutrient availability of the soils in this area (Anderson and Talbot, Reference Anderson and Talbot1965; Jager, Reference Jager1982; Quigley et al., Reference Quigley, Donati and Anderson2017).
Hence, as existing data for the Serengeti National Park in Tanzania are mainly on nutrient content of grassland biomass within the Serengeti National Park in Tanzania we set up a study compiling first data on the nutrient availability of soil from the Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti National Park in Tanzania. Our hypothesis was that a soil nutrient and toxicant gradient exists from Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti National Park in Tanzania.
Methods
Study site, experimental design and sampling
We sampled a transect from the Ngorongoro crater (3.2106° S, 35.6020° E, S1) via the Ngorongoro Conservation area to the centre of the Serengeti (2.2162° S, 34.8538° E, S16) (Figure 1). Each sample was taken 10 m beside the track/road in a depth of 5 to 10 cm from three nearby locations (all samples were in a distance from each other of 5m; mixed sample) from an open grassland site using a shovel. All samples were dry at sampling. The sampling of the 5–10 cm depth may lead to data not identical with the 0–5 cm depth of the soil (Bal et al., Reference Bal, Louw, Struyf, Sara, Le Roux, Ayisi and Schoelynck2023), but decreases potential nutrient input from dust from for example the road nearby. It was not possible to take more samples, and it was not possible to take soil profile as sampling was restricted by national park regulations, and only very short time outside the car was allowed due to potential lion or cheetah attacks. Soil samples were air-dried and sieved (<2 mm) before further analyses.
Figure 1. Soil sampling transect from the Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti. The red circle around sampling points 1 and 2 outline the Ngorongoro crater.
Soil analyses
Available concentrations of Mg, Al, Si, P, K, Ca, Mn, and Fe were quantified using the Mehlich III method (Sims, Reference Sims1989; Schaller et al., Reference Schaller, Puppe, Kaczorek, Ellerbrock and Sommer2021). The Mehlich III extract is known to be able to extract most nutrients and Al (Sims, Reference Sims1989; Schaller et al., Reference Schaller, Puppe, Kaczorek, Ellerbrock and Sommer2021). The Mehlich III extract is defined as the biologically available (available to be taken up by plant roots) share of the analysed elements, in the manuscript labelled as ‘available’. This fraction includes the element concentrations dissolved in the pore water and the fraction adsorbed to organic and inorganic soil particles. Briefly, we extracted 1 g of freeze-dried soil using 10mL of Mehlich III solution (0.015M NH4F, 0.001M EDTA, 0.25M NH4NO3, 0.00325M HNO3, and 0.2M HAc) per gram of soil. The samples were shaken for 5 min at 50 rpm and centrifuged for 5 min at 10,000 × g. Afterwards, the supernatant was filtered using a 0.2 μm PET-membrane filter.
All element concentrations in the extracts were measured at an ICP-OES (Varian, Vista-Pro radial, Palo Alto, California, USA) for magnesium, aluminium, silicon, phosphorus, potassium, calcium, manganese and iron. Total organic (TOC) and total nitrogen (TN) contents of the soil samples were determined by elemental analysis (dry combustion) following DIN-ISO-10694 (1995) using a Leco RC.612 (Leco instruments GmbH, Germany).
Results
The content of total organic carbon in the soils was between 0.7 and 6.7 %, whereas total nitrogen content was between 0.08 and 0.6 %, both elements with highest concentration within Ngorongoro crater (Figure 2). The soil Mg availability was between 0.17 and 1.05 mg g−1 with highest values both in the middle of the Ngorongoro Conservation area (sampling point 4) and near the centre of the Serengeti (sampling points 13 and 14) (Figure 2). The availability of potentially toxic Al in the soils was between 0.06 and 2.2 mg g−1 with highest values near the eastern border to Ngorongoro Conservation area (sampling points 10 and 11) (Figure 2). A comparable pattern to Al was found for Si with values between 0.65 and 2 mg g−1 with highest values again at sampling points 10 and 11 (Figure 2). A pattern different to all other analysed elements was found for P with values between 0.04 and 0.78 mg g−1 with highest values sampling point 2 within the Ngorongoro crater and lowest values around the border of Serengeti National Park to Ngorongoro Conservation area (sampling points 9 to 14) (Figure 2). For K values between 0.18 and 5.2 mg g−1 were found with high values near the border of Serengeti National Park to Ngorongoro Conservation area (sampling points 9 to 11) as well as in the centre of Serengeti National Park and low values everywhere else (Figure 2). A comparable pattern to K was found for Ca with values between 1.2 and 17.4 mg g−1 with high values near the border of Serengeti National Park to Ngorongoro Conservation area (sampling points 9 to 10) and lowest value at sampling point 6 (Figure 2). For Mn values between 0.045 and 0.33 mg g−1 were found with high values at sampling points 4 and 8 and lowest values near the border of Serengeti National Park to Ngorongoro Conservation area (sampling points 10 to 11) (Figure 2). Fe showed a different pattern with values between 0.02 and 0.33 mg g−1 with high value at sampling point 7 and the lowest in the centre of the Serengeti National Park (Figure 2). Only for Ca, there might be a reduction in availability due to livestock (sampling points 3–7).
Figure 2. Availability of total organic carbon (TOC) and total nitrogen (TN) as well as Mehlich III extraction for magnesium, aluminium, silicon, phosphorus, potassium, calcium, manganese and iron of soil along the gradient from Ngorongoro crater to the centre of the Serengeti.
We found significant linear correlations between Al and Si (R 0.84, p < 0.001), between K and Ca (R 0.65, p = 0.007), between Si and Ca (R 0.67, p = 0.005), and between TOC and TN (R 0.98, p < 0.001).
Discussion
Soil nutrient availability data for Serengeti and surrounding areas are only sparse, except for data on P and N (Anderson et al., Reference Anderson, Ritchie and McNaughton2007). Our data clearly show no clear gradient in nutrient and toxicant availability in the soils along the transect from the Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti. This unclear relation to potential volcanic ash deposition may be potentially due to non-systematic volcanic ash input from past volcanic eruptions of Ngorongoro (Anderson and Talbot, Reference Anderson and Talbot1965). However, the pattern was not as expected (decreasing nutrient availability from Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti), but the pattern was more complex with large differences in nutrient availability. We found higher values for some nutrients for sampling points 9 and 10 near the border of Serengeti National Park. This difference from the other unclear pattern may be explained by the fact that both sites are potentially affected by water flow and sediment accumulation with high tendency to nutrient accumulation (Schaller et al., Reference Schaller, Schoelynck, Murray-Hudson, Frings, van Pelt, Hegewald, Mosimane, Gondwe, Wolski and Meire2016), as both sites are within a shallow broad river bed or near this river bed (Supporting Material Table S1). The potential reduction in Ca availability due to livestock may be explained by the high demand of for example cattle for Ca potentially reducing Ca cycling by feeding plant biomass and with this limiting Ca recycling by litter decay as even in the faeces Ca is reduced (Hansard et al., Reference Hansard, Crowder and Lyke1957). The values of TOC and TN at site 1 (taken near freshwater within the Ngorongoro crater exceed the values of carbon and nitrogen values from other studies on African savanna (Bal et al., Reference Bal, Louw, Struyf, Sara, Le Roux, Ayisi and Schoelynck2023) potentially being explained by the tendency of aquatic affected sediments to exhibit high values of carbon and nitrogen (Schaller et al., Reference Schaller, Schoelynck, Murray-Hudson, Frings, van Pelt, Hegewald, Mosimane, Gondwe, Wolski and Meire2016). The correlations between Si and Ca, Si and Al, K and Ca may be explained by mineralogy, whereas the correlation of TOC and TN is related to nutrient content of organic matter in the soils. The differences in soil nutrient availability are at least partially explaining differences in the nutrient content of plant biomass in this region (McNaughton, Reference McNaughton1990; Seagle and McNaughton, Reference Seagle and McNaughton1992) with this plant biomass being the fodder for the herds of the megafauna. Hence, the differences in soil nutrient availability may be the reason for especially the herds of megafauna with offspring going to the southeast border of Serengeti to the Ngorongoro Conservation area to get enough Ca and K in the fodder (McNaughton, Reference McNaughton1990), as soil K and Ca availability is high in this area. Furthermore, Si might also be an important supplement for large ruminants offspring as shown before (Ojha et al., Reference Ojha, Malik, Mani, Singh and Singh2025) or by Si increasing milk production of cows (Radkowski et al., Reference Radkowski, Sosin-Bzducha and Radkowska2017), with Si availability being also increased at the southeast border of Serengeti to the Ngorongoro Conservation area. However, it may also be that soil Si availability as a factor in the Si content of grassland biomass alters the feeding strategies of large herbivores. For example, it is known that Si in the plant biomass deter herbivores from feeding in this area (McNaughton et al., Reference McNaughton, Tarrants, McNaughton and Davis1985). The differences in soil nutrient availability being a control for the nutrient content of the grassland biomass within this area may at least partially explain the differences in grassland biomass nutrient content and with this may explain why the herds of large herbivores migrate (McNaughton, Reference McNaughton1990).
The nutrient content of the grassland biomass reported in other studies (McNaughton, Reference McNaughton1990) with high K and Ca concentrations in plant biomass is in line with our data on nutrient availability of the soils from Ngorongoro crater via the Ngorongoro Conservation area to the centre of the Serengeti, with K and Ca availability in soil being also high at the border of Serengeti to the Ngorongoro Conservation area where plant biomass is also high in those nutrients (Figures 1 and 2). However, as our and other published datasets (Eckmeier et al., Reference Eckmeier, Kübler, Meya and Mathai Rucina2020) is limited, we call for collection and analysis of more samples and potentially soil profile analysis in this area.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0266467425000070
Author contributions
J.S. designed the experiment. J.S. and E.S did the sampling. M.S. and E.S. did the measurements. All authors wrote the manuscript, discussed the results and commented on the manuscript.
Competing interests
The authors declare no comparing interests. Experimental research and field studies on plants comply with relevant institutional, national and international guidelines and legislation.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Open access
