Study area
Long-term experiments were conducted on predominant rainfed crops at 5 centres of All India Coordinated Research Project on Agrometeorology (AICRPAM), Indian Council of Agricultural Research, representing different soil and climatic conditions. Brief information on experimental locations and data availability for the study is given in Table 8. Among the selected study locations, Akola and Parbhani are located in Vidarbha and Marathwada regions respectively, in Maharashtra state. Cotton, sorghum, soybean, pigeon pea and other pulses are the main kharif crops, while wheat, sorghum and gram are major rabi crops grown in these regions. Soil moisture will be sufficient for short duration crops, while for long duration cops like pigeon pea, cotton etc., the crop suffers from moisture stress during physiological maturity39,40 which necessitating studies on soil water balance, crop-water relations. Ananthapuramu is located in the dry plains of south western of Andhra Pradesh and Bengaluru falls under eastern dry zone of Karnataka. Anantapuramu fails to receive rains from both SW and NE monsoon due to obstruction caused by Eastern Ghats of Tamil Nadu and Western Ghats of Karnataka41. The major crops grown in these two regions are pigeon pea, groundnut, sunflower, rice, cotton, maize, chillies, sesame, finger millet and sugarcane. Usually, these crops are sown immediately after onset of monsoon rainfall i.e. in the month of June-July. The crops may make full use of available soil moisture at early stages (July-September) to attain vegetative and reproductive stage, however, may suffer moisture deficit at late maturity stages due to decrease in rainfall, leading to yield loss. Kovilpatti is situated in the midst of the vertisol belt on the southern agro-climatic zone of Tamil Nadu. It is characterized by least rainfall receiving area in the state42, receives 646.8 mm annual rainfall (29 % CV) and in which only 60 mm of which falls during four months of monsoon season. 427.7 mm of rain falls during Northeast monsoon period (October - December). The geographical locations and selected crops of the study area are depicted in Fig. 7. The understanding of spatio-temporal heterogeneity in rainfall over the five selected locations helps to understand the interactions between crop and environment43. The impact of rainfall on soil moisture is related to the change of the soil moisture field, which is dominated by the water-diffusing mechanism, surface run-off or diffusion through the medium during the rainy period. It is also observed that the correlation time of the rainfall is much smaller than that of the soil moisture field. This obviously limits the impact of the rainfall on the soil moisture field variability44.
Table 8
Crops under study and dataset availability at the study locations. FC: Field capacity, PWP: Permanent Wilting Point.
Location
|
Crop
|
Study period
|
Crop data
|
Weekly Soil moisture data
|
Soil type
|
Growth stages
|
Date of sowing
|
Depth (cm)
|
Depth (cm)x number of depths
|
FC
(%)
|
PWP
(%)
|
Akola
|
Soybean
|
2008-18
|
NA
|
4
|
30
|
15 X 2
|
32.8
|
14.2
|
Vertisol
|
Ananthapuramu
|
Groundnut
|
2009, 2012-16
|
6
|
3
|
40
|
10 x 4
|
34.0
|
9.0
|
Alfisol
|
Bangalore
|
Finger millet
|
2014-19
|
3
|
1
|
30
|
15 X 2
|
22.0
|
8.0
|
Alfisol
|
Groundnut
|
2014-19
|
3
|
2
|
45
|
15 X 3
|
Pigeonpea
|
2014-19
|
3
|
3
|
60
|
15 X 4
|
Kovilpatti
|
Maize
|
2008-19
|
3
|
1
|
45
|
15 X 3
|
35.0
|
14.0
|
Vertisol
|
Parbhani
|
Soybean
|
2013-18
|
3
|
NA
|
60
|
15 X 4
|
33.3
|
13.3
|
Vertisol
|
Cotton
|
2014-18
|
3
|
4
|
60
|
15 X 4
|
Climatic and soil characteristics of the study area
All the locations fall under typical Indian monsoon climate, characterised by two peaks of rainfall during south-west monsoon (June to September) and north-east monsoon (October to December)45. Daily rainfall data of these locations during the study period was obtained from automatic weather stations (AWS) located in each centre. The mean monthly and seasonal rainfall along with coefficient of variation of these locations during the study period is furnished in Fig. 8.
The soils of the study locations were grouped into two broad categories; vertisols (Akola, Parbhani and Kovilpatti) and alfisols (Ananthpuramu and Bangalore)46. Vertisols of Akola and Parbhani are deep, calcareous, clayey and very dark greyish brown to dark brown in colour, with low electrical conductivity (EC ⩽ 2 dS m− 1), alkaline and have exchangeable sodium percentage (ESP) less than 5. The soils are characterised by presence of deep cracks during summer (low rainfall periods) due to the swelling and shrinkage property of the clay in the soil47 which helps in soaking of deeper layers of soil soon after receipt of rainfall. Whereas, vertisols of Kovilpatti are clay loam, alkaline (pH: 8.04) and had low EC (0.45 dS m− 1) and not as deep as black cotton soils of other two vertisols under study48. Alfisols of Bangalore and Ananthapuramu are sandy clay loam in texture, acidic (pH: 5.0-6.5), low in electrical conductivity (0.09 dS m− 1) even the clay content is high, the soil fails to supply enough moisture throughout the crop growth period due to poor infiltration of rainfall because of surface crusting49.
Estimation of soil moisture
Soil moisture is one of the several factors has major role in maintaining plant water relationships and hence influencing crop growth50. It constrains plant transpiration and photosynthesis, with consequent impacts on the water, energy and biogeochemical cycles. Soil moisture is usually defined as the water contained in the unsaturated soil zone51. Generally, it is measured gravimetrically using the following formula and can be expressed volumetrically.
Where, Wf = Fresh weight of the sample, Wd = Dry weight of the soil sample
The volumetric water content (θv) can be calculated by employing the following equation,
Where, ρw is the density of water (typically assumed to be 1 Mg m− 3), Mw is the mass of water content and Vs is the volume of the soil. The θv and θm are related by the soil bulk density (ρb), which is the dry weight of soil per unit volume of field soil. (ρb = Md/Vs).
In order to estimate the water availability at root zone, the information on root depth of the crop and multiple samples of at the rooting depth is required since, the rooting depth varies widely for different crops and crop stages. The rooting depth varies from 0.1–0.5 m for garden crops to several metres for perennial trees. The information of below soil surface is also critical since rooting depth maybe restricted by physical barriers like rock layers or the chemical properties viz., high pH, Boron and salinity. Hence, soil water measures should be taken at several depths, since water content usually varies with depth within the root zone52.
The root zone water content (Wrz) can be calculated as cumulative sum of θv at each depth multiplied by the depth of soil layer as follows.
Where, θv1, θv2 θv3 and θvn are volumetric water contents at soil depths representing the root zone; d1, d2, d3 and dn represent the thickness of each of soil layers sampled.
Calculation of PASM
The per cent available soil moisture is used as one of the impact indicators (triggers) in declaration of agricultural drought8. As per the drought manual published by Ministry of Agriculture and Farmers’ Welfare, Government of India. It (PASM) was calculated using observed moisture data or soil-water balance model following the ‘bucket approach’ and it was calculated using the following formula,
Where, SMW = Weekly calculated volumetric soil moisture (vol/vol) for the current week PWP = Permanent wilting point of the soil (vol/vol) FC = Field capacity of soil (vol/vol)
Development of relationship between crop yield and PASM
Consider a crop grown in a particular season, sown in different dates of sowing. In each date of sowing, weekly soil moisture observations were taken and PASM was calculated. These weekly observations were classified according to existing ranges as follow.
PASM (%)
|
Agricultural Drought Class
|
76–100
|
No drought
|
51–75
|
Mild drought
|
26–50
|
Moderate drought
|
< 25
|
Severe drought
|
Since the crop is grown under different dates of sowings, early sown crop may experience drought lately at reproductive or maturity stages and late sown crop may experience mid-season drought. Such data on long term experiments provide a base to understand the crop performance under various drought or moisture stress scenarios at different growth stages of the crop. Likewise, data from long term experiments were collected and grouped into 4 groups (0–25, 26–50, 50–75 and 75–100) according to existing classes of PASM for drought declaration (in each growth stage of the crop i.e. vegetative, reproductive, maturity and harvest).
For each crop, regression equations were developed between percent available soil moisture and yield. Best fitting (in terms of R2 value) equations were chosen to find the yield of crop at a particular PASM and PASM required for obtaining 50% of the optimal yield. Optimum yields are estimated by averaging crop yields at university and farmers’ field, half of that yield (i.e. 50%) is substituted to ‘y’ value of the developed regression equation to find ‘x’ at that level of ‘y’.
The hypothesis behind estimating PASM requirement for 50% of optimal yield is that, if a farmer gets at least 50% of yield, he will be able to meet out the expenses of cost of cultivation, thereby helping him to get rid of farm debt related financial crisis. Hence, the PASM-yield relationship establishment in individual crop lays a scientific base for timely and appropriate drought declaration thereby helping economically meaningful relief assistance.