A silting tank in Anantapur: why the eroded field absorbs less water
Farmers in a drought-prone mandal of Anantapur district cultivate a 120 ha catchment draining into a minor irrigation tank. Mean annual rainfall is 620 mm, but most falls in two or three storms of 70–90 mm/h. The catchment slope averages 6% and soils are red sandy loams. Since 2015 farmers have removed about 3,000 m of contour bunds to enlarge fields. Tank capacity is 18 ha-m, and a 2024 survey shows it has lost 0.9 ha-m per year since 2015. Gullies are retreating upslope at 4 m/year. Infiltration tests give 55 mm/h on uncultivated sandy loam but only 12 mm/h on an eroded field where a clay-rich surface layer has formed. Downstream ayacut is 40 ha; open-well depths have fallen from 12 m to 18 m. The irrigation department proposes desilting the tank once every five years.
1. Explain the erosion processes operating in the catchment and resolve the infiltration paradox. Estimate the annual sediment yield per hectare.
Read the worked answer
The dominant processes are raindrop detachment, sheetwash, rill erosion and gully erosion. Removing 3,000 m of contour bunds lengthens the slope and concentrates runoff, so water that once spread across fields now reaches defined channels. The 70–90 mm/h storms exceed the 55 mm/h infiltration capacity of the sandy loam, and they greatly exceed the 12 mm/h capacity of the crusted field. Gullies retreating at 4 m/year show that the channel network is still expanding upslope.
The infiltration paradox is explained by surface sealing, not by clay content alone. On the eroded field, raindrop impact disperses aggregates; fine clay and silt wash into pores; and a thin depositional crust forms. The sandy loam has large pores and high conductivity when undisturbed, but the crust has small pores and low conductivity. The clay-rich layer also reduces the continuity of macropores.
Sediment yield: 0.9 ha-m = 9,000 m³/year. Over 120 ha this is 75 m³/ha/year. At a bulk density of about 1.3 t/m³, that is roughly 97.5 t/ha/year, a severe erosion rate consistent with active gullies.
2. Evaluate the proposal to desilt every five years against contour bunding and gully check dams. Recommend a treatment package with priorities.
Read the worked answer
Desilting treats the symptom. If 0.9 ha-m enters the tank each year, five-yearly desilting removes about 4.5 ha-m but the same sediment continues to arrive; the department would be paying repeatedly and losing the upstream soil.
Contour bunding costs about ₹18,000/ha. Treating all 120 ha would cost ₹21.6 lakh, which is too much to spend uniformly. Four gully check dams at ₹90,000 each would cost ₹3.6 lakh and would trap coarse sediment and stabilise the gully grade. The best package is a combination: first treat the 30 ha of critical gully heads and steep slopes with contour bunds and vegetative barriers (about ₹5.4 lakh), then build check dams across the active gullies. If the check dams trap about 70% of coarse sediment, tank siltation could fall from 0.9 to roughly 0.27 ha-m/year, but fine sediment will still pass. Desilting should be retained only as a one-time correction after the source is treated, not as a permanent five-year cycle.
3. Design a simple field monitoring plan that would show whether the conservation works.
Read the worked answer
Use five linked measurements. First, install about 20 erosion pins on 5% and 10% slopes and read them after each monsoon. Second, drive gully-head stakes with painted washers to measure retreat. Third, run a permanent transect across the tank margin and survey tank capacity with a dumpy level every two years. Fourth, use a double-ring infiltrometer on crusted and uncrusted plots to test whether surface sealing is changing. Fifth, keep a rain gauge and a small runoff plot.
The design must include a control: an untreated sub-catchment or plot. Without it, a fall in siltation could be attributed to conservation when it was caused by a dry year. Record tillage and bund condition after every season. The results should separate sheet erosion from gully sediment and show whether check dams are reducing delivery to the tank.