Prediction of soil salinity spatial distribution and its management implications for rice production in Magozi Irrigation Scheme, Iringa, Tanzania

M. Sc. Dissertation

Rice (Oryza sativa L.) plays an important role in improving household food security and national economies in Sub-Saharan Africa including Tanzania. There is an increase in annual per capita consumption of rice in Tanzania from 20.5 in 2001 to about 25 - 30 kg year -1 in 2011 coupled with an increase in population. Despite the increase in rice consumption, the current rice production in Tanzania is still as low as 2.3 t ha -1 while the potential rice yields are 4 to 10 t ha -1 in the country. Reasons for low rice production include poor agronomic practices and land degradation. Soil salinity which refers to the content of soluble salts in the soil is one of the main land degradation problems in many rice growing irrigation schemes in Tanzania. Managing soil salinity in irrigated agriculture is crucial for minimizing its negative effects and for ensuring the long-term sustainability of irrigated agriculture. To achieve this, adequate and accurate information on the magnitude and spatial distribution of soil salinity is required. The knowledge on the nature and properties of soils from pedological characterization studies is also vital in planning the best use and management of soils in crop production. Magozi Irrigation Scheme is one of the rice producing irrigation schemes with an area of 1300 ha in Iringa, Tanzania. The farmers of Magozi depend on rice production from this scheme as their main economic activity. Despite the importance of rice production in this irrigation scheme, the production yields are generally low where the average rice yields have been reported to be 3.05 t ha -1 while the potential yield in the area is 4.06 t ha -1 . There is no detailed study that has focused on addressing soil salinity problem in this irrigation scheme to understand the magnitude and its spatial distribution. This research assessed soil salinity and used GIS-based approach to predict spatial distribution of soil salinity. The study further recommends the soil, crop and irrigation management options that will contribute enhancement of sustainable rice production at Magozi Irrigation Scheme. In order to understand the nature and properties of soils in this irrigation scheme, the first specific objective was a study on pedological characterization whereby three (3) representative soil profiles namely MAG-P1, MAG-P2 and MAG-P3 were opened and characterized for their soil morphological, physical and chemical properties. The soils were then classified to the family level using USDA Soil Taxonomy and to the Tier 2 in the World Reference Base for Soil Resources (WRB). The second specific objective focused on a study to develop a linear regression model that can be used to predict electrical conductivity of the saturated paste extract (EC e ) from values of electrical conductivity measured in soil to water suspension (EC 1:2.5 ). The EC e is a globally used soil salinity index for assessing plant response to salinity. A total of 60 soil samples (45 samples for model training and 15 samples for model validation) were collected and analyzed for soil EC 1:2.5 , EC e and soil texture. A linear regression model relating EC e and EC 1:2.5 was developed and validated for use in the next study. Lastly, the study assessed soil salinity and used GIS-based approach to predict spatial distribution of soil salinity in Magozi Irrigation Scheme. A total of 81 geo-referenced soil samples at a depth of 0 - 30 cm collected from the scheme were analyzed for soil physical and chemical properties where EC e was used as the main soil salinity index. The soil salinity spatial distribution map of the scheme based on EC e was generated using Inverse Distance Weighting (IDW) interpolation method in Geographic Information System (GIS). The results on pedological characterization showed that the soils were moderately deep to very deep with vertic characteristics varying in degree of expression. Based on silt/clay ratios, the soils of Magozi Irrigation Scheme are relatively young with high degree of weathering potential. According to the USDA Soil Taxonomy, the soils were classified as Typic Haplusterts (MAG-P1), Vertic Endoaquepts (MAG-P3) and Vertic Epiaquepts (MAG-P3) while in WRB for Soil Resources they were classified as Haplic Vertisols, Eutric Vertic Cambisols and Eutric Vertic Stagnic Cambisols for MAG-P1, MAG-P2 and MAG-P3 respectively. The information from this study is crucial in planning the best use and management options of the soils of Magozi Irrigation Scheme. The linear regression model EC e = 3.4954*EC 1:2.5 (R 2 = 0.956) was developed from the study which focused on relating EC e with EC 1:2.5 to facilitate accurate soil salinity assessment in this area through predicting EC e from EC 1:2.5 values. The results on soil salinity assessment study indicated that soil salinity in terms of EC e ranged from non- saline (0.24 dS m -1 ) to extremely saline (33.3 dS m -1 ) with an EC e mean of 2.5 dS m -1 being slightly saline. The mean EC e value of 2.5 dS m -1 recorded in this area is high enough to cause 10 to 25 % crop yield reduction from the total yield. In terms of rice response to salinity and effects in its production, the mean EC e value of 2.5 dS m -1 is very close to 3 dS m -1 which is the EC e threshold for rice crop. The EC e showed positive significant correlation at p≤0.05 significance level with soil Cl - (r = 0.459), exchangeable Na (r = 0.341), ESP (r = 0.302) and SAR (r = 0.320). The soil salinity spatial distribution map indicated that out of 1300 ha of the cultivated land, about 622.21 ha (47.86%) were slightly saline to extremely saline soils. This research work found that soil salinity is a growing land degradation problem in Magozi Irrigation Scheme. This may be due to poor soil and irrigation management practices as well as poor drainage structures leading to waterlogging problems which promote salt accumulation in the soil. Therefore, suitable irrigation, crop and soil management practices must be adopted by the farmers to reduce soil salinity development for sustainable rice production in the scheme. It is recommended that the farmers should be encouraged to adopt efficient rice farming technologies such as the system of rice intensification (SRI), improve irrigation drainage channels and adopt growing of salt tolerant rice varieties such as SATO 1 as well as use of inorganic and organic fertilizers.