Rainwater harvesting in arid and semi-arid zones

In arid and semi-arid regions, the scarcity of water can be alleviated by rainwater harvesting, which is defined as a method of inducing, collecting, storing, and conserving local surface runoff for agriculture. Rainwater harvesting can be applied with different systems, and this dissertation deals with the system of micro-catchments. A microcatchment consists of a runoff area and a basin area in which a tree is planted. The purpose of this study was to develop a design procedure for micro-catchments, applicable to environmental and human conditions prevailing in developing countries. Underlying the design procedure is an analysis of the water balance of the system.

The design method is based on a prediction of actual transpiration by the numerical soil-water-balance model SWATRE, while the runoff component is predicted by a runoff model. The design- aims at sufficient soil water being available in an average rainfall year. Deep percolation losses occur in wet years, and water shortages in dry years. A tree suitable for these conditions is able to withstand dry periods and drought years. The practical problem selected for this study was the establishment of Neem windbreaks in Niger and Nigeria. Points to consider in the design are the seasonal distribution of rainfall, the soil hydraulic conditions, and the tree hydrological/ physiological characteristics.

The theory of four surface runoff models is presented. These models are compared in their capability and accuracy to predict runoff volumes for micro-catchment design and in their model concept, structure, parameters, and input data requirement. A kinematic-wave model with depression storage and a linear regression model are considered the most suitable for micro-catchment design. The theory of the soil-water-balance model is discussed, as is the calibration of this model with data from Sede Boqer in the Negev Desert. The application of the model for micro-catchment design is demonstrated for an extremely and zone and an and zone in the Negev Desert.

The extremely and zone is too dry for rainwater harvesting from micro-catchments, larger catchments being required there. In the and zone, the basin areas should be approximately 40 m ²for each tree, and the runoff areas 60 m ². The design approach is applied to five weather stations in Niger and northern Nigeria where data were available. Data from a Neem windbreak at Sadoré in Niger were used to calibrate the model. Data from Niamey, near Sadoré, were used to compare runoff prediction with two runoff models and to predict micro-catchment design. The combination of a runoff-depth model and the soil-water-balance model was used to predict microcatchment design at Sadoré, and Tahoua in Niger and at Sokoto and Katsina in Nigeria.

The conclusion of the design predictions is that the required runoff area per tree is about 40 m ²at Tahoua and about 20 m ²at Niamey, Sadoré, Sokoto, and Katsina. With such runoff areas, a good growth of trees could be achieved at degrees varying roughly from 40% of a certain target transpiration at Niamey, 50% at Sadoré, 80% at Sokoto, and 100% at Katsina. The overall conclusion is that, in and and semi-arid zones, runoff from small areas such as micro-catchments is an important potential source of water for the establishment, development, and growth of trees. A supply of runoff water can make the difference between death, survival, minimum development, and good growth of trees. Especially in dry years, the runoff water can considerably improve the environmental conditions in which the trees have to grow.

The data required to apply this approach and arrive at a preliminary design are discussed. Rainfall and evaporation records are needed to supply important weather data. Data on topography, soil profile, soil hydraulic functions, and tree hydrological characteristics can be measured or estimated in the field, or determined in a laboratory from samples. With a preliminary design, well-conceived field experiments can be set up. As more data become available from the field, the design can be adjusted and worked out in detail for a particular location.

The only potential alternative method of water supply to a windbreak would be trickle irrigation. But this would enhance the development of a shallow root system and would require a source of water, high capital investment, and irrigation management skills. All these requirements are difficult to realize for a windbreak. Instead, for this application, rainwater harvesting from micro-catchments is suitable, cheap, good, and efficient. Rainwater harvesting should be seen as complementing irrigated agriculture, rather than competing with it. Irrigated agriculture is practised on the best soils, where water is available to grow field crops. Rainwater harvesting is a good alternative on marginal lands where irrigation water is not available. Because of dry periods and drought years, rainwater harvesting works best for deep-rooting, drought-resistant trees.

The technology involved is not complicated and can easily be adapted to local conditions of climate, soil, and trees. In many of these areas, there is a lack of water, wood, food, and shade, while wind erosion is a major problem. Windbreaks and shelterbelts can serve both the local population and the environment. Once the trees have been planted and the runoff areas constructed, some annual maintenance is needed but no continuous care. This is important for nomads, who are not farmers. Windbreaks demarcate and protect farmland, while large-scale shelterbelts consisting of different types of trees and bushes also serve nomads who do not settle. Microcatchments also reduce soil erosion by water, because they control surface flow. In addition, deep percolation in wet years recharges the groundwater. This can help to redress an upset regional water balance and combat desertification.