Dynamics of ephemeral ponds and suitability for irrigation in the Vryburg District, South Africa

Abstract

South Africa is a semi - arid country and most parts of the country, including the study area, are dry due to water scarcity. This situation has adversely affected food security, social and economic development of Vryburg District. During the short rainy season that occurs from October to March, ephemeral ponds form in many places. This pond water is not beneficially utilized and is lost through evaporation and infiltration. The main objective of this study was, therefore, to determine the suitability of the pond water for irrigation. Goggle Earth was used to identify all ponds in the study area. This was followed by the use of the phase file to map the distribution of the ponds. It was found that the distribution of ponds depended on rainfall intensity, soil characteristics and the nature of the underlying rock. Five ponds were selected for study from 22 originally considered. The criteria used to select the 22 ponds were: proximity to major road, longevity and size of the pond(> 2ha). ASTER 30-m resolution digital elevation system model (DEM) data were used to extract slope length and height of each selected pond. The DEM was also used to demarcate the catchment area of each selected pond. Furthermore, remote sensing was used to display LULC of the individual pond in the sub catchment from 2004 - 2013 . The main land cover classes were woody plants, grass, bare area, built-up area and water. There was an increase in the area covered by woody plants. This was attributed to bush encroachment. Over-grazing was believed to be the reason for reduction of grass cover to create bare areas. Increase in the area covered by water was due to seasonal and daily variability in rainfall. Finally, there was an increase in the size of the built-up area which could be attributed to construction and migration of people to urban areas. In addition, the relationship between LULC and water quality was investigated. Water samples were collected from the 5 ponds and chemical and biological contaminants were analysed. All the chemical data were within the recommended range specified by DW AF and FAO (Na+40.5mg/l, K+ 3.16mg/l, N03- 0.45mg/l, Cd 0.03mg/L) except for cadmium. Escherichia coli counts were below the recommended value set by WHO (78 counts/I 00 ml). The results were combined with land cover change to run multi-linear regression equations to determine the effect of LULC on water quality of pond water. When the R 2value was O. 89, grass as well as bare area had no effect on nitrate concentration in the pond water. Similarly, (p<0.58) grass and bare soil had no effects on electrical conductivity in ephemeral pond water and the R2 value was 0.78. Furthermore, Na+ in water did not depend on grass or bare area with R2value of 0.92. In addition, grass and bare area did not have any impact on cadmium concentration in the pond water (p<0.85). In addition only 45% of the data could be accounted for by the equation. Lastly, grass and bare area had significant effects (p<0.006) on E. coli abundance (R2=1). Additionally, the Darcy's equation for infiltration rate, Penman's method for evaporation rate, and the depth of water column were used to model an equation on water balance in the ponds. This was tested in the field and was found that the water in ponds A, C and D could last below 42 days hence was not suitable for irrigation. The water in pond B lasted for 50 days and could be used to grow short seas - seasoned crops. The water in pond E lasted for 69 days. The water could be used to grow short - seasoned crops and some vegetables. However, the water from all the ponds could be suitable for irrigation when it is used in the middle of the rainy season. The water can also be used to supplement irrigation during dry spell in the cropping season. Climatic data such as rainfall, temperature, wind speed and evapo-transpiration were collected and standardized. The standardized Precipitation Index (SPI) was used to standardize the rainfall data. SPI revealed periods of above average (706 mm), average ( 415 mm) and below average rainfall (234 mm). The results were combined with water depth information and the data from water analysis to develop suitability indices for irrigation. Consequently five regimes were obtained. Ponds A, C and were not suitable for irrigation, C was suitable for irrigation only through soil, water and crop management. Pond D&E were suitable for irrigation due to the greater water depth and good water quality.