An experiment to gauge an ungauged catchment: rapid data assessment and eco-hydrological modelling in a data-scarce rural catchment

Figures & data

Fig. 1 The Mod catchment, Jhabua, Madhya Pradesh, India. Main panel: Locations of the transects and sampling points, including the analytical aspect, ephemeral river network (blue) and topography (white contours) superimposed on a composite of bands 5, 3 and 2 of Landsat 7 scene 2000-04-01.Right panel: The geological (GSI 1976) and pedological (GSI 1988) data basis including the identified transects. The functional soil map is the result of a RS classification of the same Landsat 7 data; the five functional soil classes and three sub-classes as in Table 1.

Fig. 2 The spatial and process elements of the hydrological model WASA (after Güntner and Bronstert 2004).

Table 1 Overview of the resulting functional soil classes and the mean values of all analysed samples of a class at <0.6 m depth. Skel: mass-percentage of stones (%); Sand, Silt and Clay: mass-percent of the sieved sample, hence without the skeletal fraction (%); Δ: USDA texture class clay / loam / sand; CEC: cation exchange capacity in mol⁺ per gram of soil (mol⁺/g); C_org: organic carbon content in per cent of mass (%); z: depth (m); BD: bulk density (g/cm³); θ_S: saturated water content; θ_r: residual water content; van Genuchten parameters α (cm^-1) and n; and k_s: infiltration rate as saturated conductivity (m/s). Values in italics are estimates within the main classes highlighted by the shading.

Soil class	Cover	Skel	Sand	Silt	Clay	∆	CEC	Corg	z	pH	BD			α	n
1 Deep Clay Black	8.7%	0	20.1	23.9	56.0	c	750	0.6	2.5	8.34	1.7	0.383	0.073	0.944	1.21	1.7e-5
2 Loam Alkalic Dark	11.8%	16.5	39.77	30.99	29.24	l	457	0.49	1.2	8.05	1.8	0.305	0.0662	0.21	1.4	2.5e-5
3 Clay Loam Red Kaolinitic	1.6%	14	45.47	27.85	26.68	l	342	0.51	1.0	7.43	1.8	0.45	0.105	0.205	1.37	2.2e-5
4 Meagre Red	24.5%	35.1	39.28	38.17	22.58	l	330	0.37	0.2	7.69	1.9	0.49	0.043	0.205	1.37	3.2e-5
8 Skeletal Hill	6.1%	77.8	48.24	26.6	25.2	scl	109	-	0.2	6.69	1.8	0.49	0.043	0.205	1.37	2.4e-5
5 Meagre Loam Brown	18.2%	29.7	36.69	33.18	30.13	cl	309	0.47	1.2	8.5	1.8	0.48	0.050	0.18	1.32	1.4e-5
6 Skeletal Brown	18.9%	64.5	36.69	33.18	30.13	cl	209	-	0.2	8.5	1.8	0.48	0.050	0.18	1.32	1.4e-5
7 Acidic Red Feldspar	8.2%	23.2	39.24	36.6	24.2	l	216	0.48	1.0	6.8	1.7	0.48	0.050	-	-	4.3e-5

Table 2 Comparison of total monsoon precipitation to NDVI and agricultural productivity. Kharif and rabi crop productivity are given relative to the mean based on the statistical year books of Jhabua District.

Year	Precip.	Area NDVI > 0	Rel. productivity
			Kharif	Rabi
2000	250 mm	1.9%	0.5	0
2001	780 mm	22.8%	1	1

Fig. 3 WASA_crop performance simulated vs gauged discharge for the Mod catchment, 1992–1996. Upper panel: Time series of precipitation and discharge with compressed date axis for dry season. Middle panel: Time series of observed and modelled discharge. Lower panel: Seasonal lag cross-correlation of simulated against observed discharge. The tables present the observed runoff coefficient (C_obs), runoff coefficients from simulations (C), Spearman rank correlation (ρ), and rainfall accumulated (Σ rain) during the season which is reduced to the period of available discharge observation as indicated by the red bars.

Fig. 4 Qualitative validation of modelled soil moisture vs RS soil moisture proxy. Upper row: model results for topsoil moisture in sub-basins. Lower row: mid-IR reflectance of Landsat 7 band 5 at the indicated dates.

Table 3 Data basis of the cropping agent. (a) For each crop, suitability and requirements are listed. The agent ranks the crop depending on accordance with the given strategy and site. (b) and (c) additional soil and cultivation references for Table 3(a) which correspond to site specifications.

(a)
Crop		Season		Water				Soil						Production
No.	Crop name	Suitability for kharif	Suitability for rabi	Water requirement	Water stress resistance	Water- logging resistance	Minimum preci-pitation (mm/year)	Fertilization in rabi	Preferred soil fertility	Preferred soil type	Alternative soil type	Alternative soil type	Erosivity	Labour intensity of cropping	Production costs		Harvest gain
															Low	High	Low	High
1	Pigeonpea	1	2	1	1	0	300	1	1	2	3	1	1	1	930	1379	1400	1750
2	Millet	2	1	0	2	0	180	1	0	2	1	0	1	1	485	620	500	659
3	Sunflower	0	2	1	1	1	250	1	1	2	1	3	2	1	300	300	300	300
4	Sorghum	2	1	1	1	1	500	1	1	2	3	1	1	1	578	745	500	759
5	Chickpea	1	2	1	1	1	300	0	1	2	3	1	2	1	727	1188	1300	1500
6	Maize	2	1	2	0	0	500	2	2	2	3	1	3	1	445	579	510	630
7	Cotton	2	0	2	2	1	600	1	1	2	3	0	3	2	1893	2921	1800	2800
8	Groundnut	2	1	2	1	1	500	1	1	2	1	3	1	1	1197	1517	1400	2000
9	Sugarcane	2	0	2	2	1	1200	2	2	3	2	1	3	2	100	100	120	120
10	Soybean	2	1	2	0	1	500	2	1	2	1	3	2	1	678	979	1100	1650
11	Casawa	1	1	1	2	2	750	1	0	2	1	0	1	0	50	50	100	100
12	Wheat	1	2	1	1	1	250	1	1	2	3	1	2	1	424	640	600	900
13	Grass	2	2	0	1	1	100	0	0	2	1	3	0	0	9	10	10	11
		0: no;		0: low;						1: sand;			0: low;		Indian Rupees
		1: ok;		1: medium;						2: loam;			1: med;		per Quintal
		2: pref.		2: high.						3: clay.			2: high.
(b)
Soil no.	Major class	Minor class	CEC	Soil name
1	3	0	750	Deep Clay Black
2	2	3	457	Loam Alcalic Dark
3	2	3	342	Clay Loamy Red
4	2	0	330	Meagre Red
8	4	2	109	Skeletal Hill
5	2	3	309	Meagre Loamy Brown
6	2	4	209	Skeletal Brown
7	2	0	215	Acidic Red
S	4	0	59	Skeletal
1: sand; 2: loam; 3: clay; 4: skeletal; 0: no soil.
(c)
ID	Major class	Minor class	Land-use class name
1	4	0	Wasteland
2	3	4	Meagre grass
3	2	3	Dryland meagre agri.
4	2	3	Dryland agriculture
5	1	2	Floodplain

1: intensive; 2: normal; 3: meagre; 4: grass.

Fig. 5 Model results using different cropping agents. Upper panel: Simulated biomass production for kharif (top) and rabi (bottom) under the crop setting imposed by the agent. Middle panel: Series of modelled discharge without no-flow periods. Lower panel: Time series of mean soil moisture in the catchment.

Fig. A1 SWAP analysis calibration schemes.

Fig. A2 Overview about SWAP validation results after the respective calibration schemes A–C.

Fig. A3 Comparison of observed and modelled soil moisture dynamics at lysimeter 10_3 (Loess Chernozem Silt Loam) and 7_5 (Cambisol/Stagnosol Loam) as time series from 01-08-1996 with daily time step. Standard refers to a standard parameterization without calibration. Right panel: The box plots present the performance of all validation years for all lysimeters with respect to water balance (R²). The dashed (red) line marks the mean of all realizations within the specific calibration scheme.

Fig. A4 Cross-validation of yield simulation with SWAP using calibrations of the same crop to different soils.

GSI, 1976. Geology and Mineral Resources of the States of India — Part XI. Madhya Pradesh: Geological Survey of India.

 Google Scholar

GSI, 1988. District Resource Map — Jhabua District. Madhya Pradesh: Geological Survey of India.

 Google Scholar

Güntner, A. and Bronstert, A., 2004. Representation of landscape variability and lateral redistribution processes for large-scale hydrological modelling in semi-arid areas. Journal of Hydrology, 297 (1–4), 136–161. doi:10.1016/j.jhydrol.2004.04.008.

 Web of Science ®Google Scholar