Runoff estimation using satellite images in sparsely gauged basins: a case of Nzoia River Basin, Kenya
doi.org/10.2166/wpt.2022.088
Kivuti Mureithi, Job Rotich Kosgei, Gilbert Nyageikaro Nyandwaro, Ednah Jelagat Kemboi & Noah Kipyego Sum
This blog post was written by the author of a recent Water Practice & Technology paper and summarises the key features of the research and its implications.
Accurate runoff estimation is vital for many professions, and in this case particularly in civil engineering. Runoff data for a particular basin or study area of interest is most important in the design of hydraulic structures, flood control and management, irrigation management, among other engineering applications. The primary source of runoff is precipitation. Rainfall is the main form of precipitation, particularly in tropical regions. Consequently, rainfall-generated runoff estimation is generally accepted to be accurate for design. Therefore, the availability of accurate rainfall data at spatial and temporal scales is vital for runoff estimation.
The basin, a dynamic geographic unit area where runoff contributes to a common point, has been generally accepted as the basic area for hydrological planning and management. For effective runoff estimation, sufficient data that is well distributed is required. Failure to do so may result in catastrophic effects. Recurring flood disasters and failure of hydraulic structures are examples of these effects, which lead to mass loss of human life and property.
In developing countries, especially in Africa, insufficient runoff data has been observed to lead to the under-utilization of arable land, improper design of hydrologic structures, and poor flood warning systems. In Kenya, the same is reported for practically every basin. This easily becomes evident every time high rainfalls are recorded. During periods of high rainfall, flooding and failure of hydraulic structures is often the case. On the contrary, in periods of low rainfall, famine and socio-economic drought is usually the case. Most of these issues can either directly or indirectly be traced to insufficient runoff and rainfall data for planning and management of the catchment areas.
The same problem is protracted in the Nzoia basin. Despite rich natural resource endowment, a high incidence of poverty is reported for the basin. The Western Kenyan region in the basin is endowed with natural resources such as forests, rivers, and lakes, which should be adequate for poverty reduction. Poverty and vulnerability nonetheless still affect many in the region, with urban poverty level being the highest in the country at 80% (Odira et.al, n.d.). Flooding, disease, and degradation of natural resources [mainly land] are common problems facing the communities in the basin. Perennial flooding and poor management of the natural resources characterized in the region can be traced to poor rainfall-generated runoff data. The importance of the availability of adequate rainfall-generated runoff data at spatial and temporal scales cannot be underemphasized.
Historically, rainfall is measured using rain gauges. Rain gauges provide a reasonably accurate measurement at one point or field plot. Observational and instrumental errors are the major sources of errors in rain gauge measurement of rainfall. These may lead to up to a 30% difference between the measured data and the actual rainfall (World Meteorological Organization, 2006). With proper management, these errors could be managed, and rain gauge data is therefore accepted to be sufficiently accurate for runoff estimation over a basin.
Gauging stations in the Nzoia basin are sparsely distributed within the area, which means that they may currently be insufficient for runoff estimation. A sparse rain gauge network cannot reflect rainfall variability caused by topography and orography and will result in erroneous estimates of areal rainfall (AndrÉassian et.al, 2001)
Global rainfall databases from satellites such as the National Aeronautic and Space Administration (NASA) satellite are also available for runoff estimation. However, their spatial scales (ranging from 100x100km to 250x250 km and 250x375km, etc.) are too coarse and wide for proper hydrological management and analysis. It has also been observed that their temporal scales are usually insufficient.
By working out a calibration protocol for NASA satellite data with the rain gauge data, the challenges posed can be solved and the resulting data could be used for rainfall-runoff studies. This study, therefore, aimed to prove that satellite data such as NASA's GEOS-5 when adequately calibrated provide suitable estimates for runoff discharges in basins that are sparsely gauged.
The article can be read in full over on the Water Practice & Technology page.
References
World Meteorological Organization. (2006). Guide to meteorological instruments and methods of observation and information dissemination. Geneva: Secretariat of the WMO.
AndrÉassian, V., Perrin, C., Michel, C., Usart-Sanchez, I., & Lavabre, J. (2001). Impact of imperfect rainfall knowledge on the efficiency and the parameters of watershed models. Journal of Hydrology.
Odira, P. M., Nyadawa, D. M., Ndwallah, B. O., Juma, M. N., & Obiero, M. J. (n.d.). While the Western Kenyan region is endowed with natural resources such as forests, rivers, and lakes, which should be adequate for poverty reduction, poverty and vulnerability nonetheless afflict many in the region..