PhD Thesis
In the last three decades, Ethiopian agriculture solely depended on imported fertilizers,
mainly urea and di-ammonum phosphate (DAP), which supplied nitrogen (N) and
phosphorus (P). The use of such high-analysis fertilizers lacking adventitious S, coupled
with traditional farming and cropping systems that mine S from native soil, can lead to S
deficiency. Thus, recently there has been a perception that the production of such high
protein cereals like wheat, legumes and even oil crops can be limited by the deficiency of
sulphur (S) and possibly other nutrients as well. To explore the extent of S deficiency,
three sets of experiments were conducted in the 2012-14 cropping seasons. The major aim was to quantify S status of soils and plants vis-à-vis other soil fertility parameters in the Central Highlands (CHLs) of Ethiopia. Moreover, an investigation was made to explore the response of wheat to applied S in relation to N and P. The study was also aimed to assess the different S supply indices with the ultimate goal of setting the critical levels (CLs) in order to make rational fertilizer use. Finally, the study was also aimed at
determining the optimum S rate for wheat.
In the first experiment four treatment combinations, namely: an absolute control (without
any fertilizer) or CK = (N0P0S0), N alone or N = N1 = (N1P0S0), nitrogen plus sulphur or
NS = N1S1 = (N1P0S1), and nitrogen plus phosphorus plus sulphur or NPS = (N1P1S1)
were tested in Arsi, East Shewa and Oromia Liyuu zones covering diverse agro
ecological zones (AEZs) and soil types. Two levels of each nutrient were tested, namely:
sulphur (0 and 20 kg S/ha), phosphorus (0 and 20 kg P/ha), and nitrogen (0 and 69 kg
N/ha). The nutrient sources were gypsum, triple super phosphate (TSP) and urea.
The treatments were arranged in the randomized complete block (RCBD) design and
replicated three times. In the second experiment meant for evaluating the different indices of S supply, also four treatment combinations, namely: an absolute control (without any fertilizer) or CK = (N0P0S0), N alone or N = N1 = (N1P0S0), nitrogen plus sulphur or NS = N1S1 = (N1P0S1), and nitrogen plus phosphorus plus sulphur or NPS = (N1P1S1) were tested in the similar three locations covering diverse AEZs. Similarly, two levels of each nutrient were tested, namely: sulphur (0 and 20 kg S/ha), phosphorus (0 and 20 kg P/ha), and nitrogen (0 and 69 kg N/ha). The nutrient sources were gypsum, TSP and urea.
The treatments were arranged in the RCBD design and replicated three times. In this part of the experiments four indices of S deficiency (OC and SO4-S in native soils; and total S and N/S ratio in grains at harvest) were evaluated to establish their association with yields with the ultimate goal of setting their critical levels (CLs). A third experiment aiming at establishing optimum rates of S was set at six sites in Arsi, East Shewa and Oromia Liyuu zones. The nutrient elements tested were combined in nine treatments namely: an absolute control (without any fertilizer), CK = (N0P0S0); nitrogen alone or N = N1 = (N1P0S0); nitrogen and sulphur or N1S1 = (N1P0S1); nitrogen and sulphur or NS2 = (N1P0S2); nitrogen and sulphur or NS3 = (N1P0S3); nitrogen and phosphorus or N1P1 = (N1P1S0); nitrogen, phosphorus and sulphur or NPS1 = (N1P1S1); nitrogen, phosphorus and sulphur or NPS2 = (N1P1S2); and nitrogen, phosphorus and sulphur or NPS3 = (N1P1S3) were tested. The nutrient rates tested were four levels of S (S0 = 0 = CK, S1 = 5, S2 = 10 and S3 = 20 kg S/ha); two levels of P (P0 = 0 = CK and P1 = 20 kg P/ha); and two levels of N (N0 = 0 = CK, N1 = 69 kg N/ha) were applied as gypsum, TSP and urea. The wheat variety “Kekeba” was used as test crop in all trials.
Wheat responded significantly to N application at all sites. While in the case of sulphur,
about 50 % of the studied fields showed highly significant response and 22 % showed
marginal-response to S. Similar to N, all the study fields tested low in available P, but 56
% showed highly significant response, and 22 % marginal-response to applied P. Good relationship between soil-test values and crop response, especially for N and S was
observed. In addition, the study revealed that light textured soils in the peripheries of the
Rift-Valley and calcareous vertisols were more deficient in S than the rest of the soils.
The results of the second study indicated that, the N/S ratio and S content in wheat grain
were better correlated with S-uptake than the soil S indices i.e. OC and the SO4-S. Which
means that, plant analysis is more useful tool for diagnosing S supply from soils to plant
than the soil analysis. The critical levels were found to be 0.118 % for total S in grain,
14.7/1 for the N/S ratio in grains and 11.3 mg/kg for the SO4-S in soils. The results from
the sulphur rate trial indicated that four sites: G/Silingo, Keteba, N/Suba, and Bekejo had
highly significant response to S at all treatment levels. The other two sites: W/Gora and
B/Tokofa gave marginal response to S. Based on this, the three sites namely: Keteba,
Bekejo and N/Suba were categorized as having very low SO4-S status, while G/Silingo
was marginal. Whereas W/Gora and B/Tokofa were grouped under the category of sites
with adequate SO4-S in soils. Based on the extent of response to sulphur, optimum S rate for Keteba, Bekejo, N/Suba and Gora Silingo sites was about 20 kg S/ha, whereas at the W/Gora and B/Tokofa sites, where the SO4-S rated as adequate, yet wheat responded to low rates, applying S at a rate of 5-10 kg/ha is recommended.
The present study confirmed that S deficiency is becoming one of the major problems of
soil fertility in the Ethiopian crop production system. This low level of soil S status is due
to nutrient depletion without replenishment. Therefore, fertiliser recommendations should include the optimum amount of S for wheat and/or other crops. The alternative
management decisions may include improved management of natural nutrient re-cycling within an agro-ecosystem, and by the importation of nutrients into the system in the form of fertilizers mined (e.g. gypsum) or manufactured off-farm.
Alliance for Green Revolution in Africa (AGRA)