Abstract
This chapter presents the trend of drought as a stochastic natural disaster influenced by the climate variability for the upper Tana River basin in Kenya. Drought frequency, duration and intensity in the upper Tana River basin have been increasing over the years. To develop measures for mitigating impacts of drought, the influencing hydro-meteorological parameters and their interaction are necessary. Drought definitions, fundamental concepts of droughts, classification of droughts, types of drought indices, historical droughts and application of artificial neural networks in analyzing impacts of drought on water resources with special focus on a Kenyan river basin is presented. Gaps for more focused research are identified. Although drought forecasting is very vital in managing key sectors such as water, agriculture and hydro-power generation, drought forecasting techniques in Kenya are limited. There is need therefore to develop an effective drought forecasting tool for on-set detection, classification and drought forecasting. The forecasting is necessary for decision making on matters of drought preparedness and proper water resources planning and management.
Top2. Background
African countries are among the most vulnerable to impacts of drought and climate variability. The impacts adversely affect the well-being of human population. These impacts are compounded by numerous factors such as vast arid and semi-arid lands (ASAL) in the region, high levels of poverty, high population density, and recurrent of diseases. This is expected to multiply the demand for water, food and forage for livestock within the area in the next decades (Okoro et al., 2014). In East Africa, it has been projected that water availability will decline in future due to drought. In addition, there is a likelihood of increased desertification due to decline in seasonal amount of precipitation and alteration of its patterns at different magnitudes for different river basins (Wilby et al., 2006; Mwangi et al., 2013).
Key Terms in this Chapter
Drought: A condition on land characterised by scarcity of water that falls below normal average or defined threshold levels.
Unsupervised Learning: In unsupervised learning, network is able to learn and recognize patterns in data set whenever the data is introduced to the network. It is achieved through a competitive learning rule.
Remote Sensing: The science and art of obtaining data of points, objects, areas or phenomena through analysis of data acquired by a sensor, which is not in direct physical contact with the target of investigation.
Upper Tana River Basin: The upper Tana River basin is the area surrounded by a divide with an area of 17,420 km 2 and lies between latitudes 00 0 05' and 01 0 30' south and longitudes 36 0 20' and 37 0 60'.
Drought Indices: A function of water-related variables (such as precipitation, streamflow, reservoir volume, dam inflow and ground-water level) for quantifying drought.
Reinforced Learning: A learning approach in which the ANN learns by trial and error technique.
Satellite Based Drought Indices: The satellite based indices are those that use remote sensing data.
Global Warming: The critical changes in temperature and temperature dependent variables within the atmosphere and land, caused by high concentrations of hydrocarbon gases (carbon dioxide, methane and nitrogen) in the atmosphere.
Artificial Neural Networks: A model system for processing large and complex data which uses inputs to generate outputs in the process, it simulates the working principles of neuron of a human brain.
Supervised Learning: The ANN processes the inputs and compares its resulting outputs with the target. Errors are then propagated back through the system, causing the network to adjust the weights which controls the network.
Data Driven Drought Index: A tool that uses a single or a combination of hydro-meteorological variables as input parameters to assess drought intensity, duration, severity and magnitude.