Revisiting lake sediment budgets: How the calculation of lake lifetime is strongly data and method dependent

Abstract

Lake sedimentation has a fundamental impact on lake lifetime. In this paper, we show how sensitive calculation of the latter is to the quality of data available and assumptions made during analysis. Based on the collection of a large new dataset, we quantify the sediment masses (1) mobilized on the hillslopes draining towards Lake Tana (Ethiopia), (2) stored in the floodplains, (3) transported into the lake, (4) deposited in the lake and (5) delivered out from the lake so as to establish a sediment budget. In 2012–2013, suspended sediment concentration (SSC) and discharge measurements were made at 13 monitoring stations, including two lake outlets. Altogether, 4635 SSC samples were collected and sediment rating curves that account for land cover conditions and rainfall seasonality were established for the 11 river stations, and mean monthly SSC was calculated for the outlets. Effects of the floodplain on rivers' sediment yield (SY) were investigated using measurements at both sides of the floodplains. SY from ungauged rivers was assessed using a model that includes catchment area and rainfall, whereas bedload and direct sediment input from lake shores were estimated. As a result, the gross annual SY was c. 39.55 (± 0.15) Mt, dominantly from Gilgel Abay and Gumara Rivers. The 2.57 (± 0.17) Mt sediment deposited in floodplains indicate that the floodplains serve as an important sediment sink. Moreover, annually c. 1.09 Mt of sediment leaves the lake through the two outlets. Annual sediment deposition in the lake was c. 36.97 (± 0.22) Mt and organic matter accumulation was 2.15 Mt, with a mean sediment trapping efficiency of 97%. Furthermore, SSC and SY are generally higher at the beginning of the rainy season because soils in cultivated fields are bare and loose due to frequent ploughing and seedbed preparation. Later in the season, increased crop and vegetation cover lead to a decrease in sediment production. Based on the established sediment budget with average rainfall, the lifetime of Lake Tana was estimated as 764 to 1032 years, which is shorter than what was anticipated in earlier studies. The sedimentation rate of Lake Tana (11.7 ± 0.1 kg m−2 yr−1) is in line with the sedimentation rates of larger lakes in the world, like Lake Dongting and Lake Kivu. Copyright © 2017 John Wiley & Sons, Ltd.

Identifying erosion hotspots in Lake Tana Basin from a multisite Soil and Water Assessment Tool validation: Opportunity for land managers

“…Like the rainfall pattern in LTB, this severe rate of sediment delivery to main rivers also follows an increasing trend from north to south in clockwise direction, that is, SY in subbasins located in Megech catchment (57 Mg ha −1 yr −1 ) is lower than subbasins located in north‐east (65 Mg ha −1 yr −1 ), Gumara (73 Mg ha −1 yr −1 ), and Gilgel Abay (80 Mg ha −1 yr −1 ) catchments. This estimated SY in this study is within the range reported data that were observed in and around LTB (Ayele et al, ; Ebabu et al, ; Lemma et al, ; Melaku et al, ; Subhatu et al, ). In contrasts, relatively lower sediment is produced in the low‐lying floodplains of the basin.…”

“…A total of 3,855 suspended sediment concentration (SSC) and its corresponding streamflow data were measured in the monitoring stations in Megech, Gumara, Gelda, and Gilgel Abay rivers (2012–2016). The average SSC is 3.51 ± 6.25 g L −1 that ranges from 2.50 (±4.95) in Gelda to 4.43 (±9.63) g L −1 in Megech, which is quite within the range of values reported by Lemma et al () and Ebabu et al () in LTB. The SSC in Megech River is a bit higher than the natural condition due to intensive river bank excavation for dam construction.…”

Budgeting suspended sediment fluxes in tropical monsoonal watersheds with limited data: the Lake Tana basin

“…This value is a similar order of magnitude to the sediment delivered by other watersheds in the Ethiopian highlands (Abate et al, 2017;Dagnew et al, 2015;Tilahun et al, 2013aTilahun et al, , 2015Vanmaercke et al, 2010). Finally, the soil losses reported by Erkossa et al (2015) and Lemma et al (2017), with 2.7 t/ha/yr, are unrealistically low. Based on the average reported sediment concentration of 4.4 g/l in Lemma et al (2017), it can be easily shown that their sediment losses should agree with our calculations of around 30 t/ha/yr.…”

“…Finally, the soil losses reported by Erkossa et al (2015) and Lemma et al (2017), with 2.7 t/ha/yr, are unrealistically low. Based on the average reported sediment concentration of 4.4 g/l in Lemma et al (2017), it can be easily shown that their sediment losses should agree with our calculations of around 30 t/ha/yr.…”

Sediment yield at southwest Ethiopia's forest frontier

“…The findings on cropland are also higher than the median of the measured values in Africa for catchments of 2 ha (7.8 Mg ha −1 yr −1 ) or 8 ha (6.3 Mg ha −1 yr −1 ; Vanmaercke et al, ). In Ethiopia, all measured SSY for catchments up to 1,000 km 2 are situated in the range of 5–40 Mg ha −1 yr −1 (Vanmaercke et al, : Lemma et al, ). Despite the higher rainfall than on all other measurement sites, the 2‐year measurement results on cropland lie well within this range.…”