Shibendu S. Ray

shibendu.ncfc@nic.in

Mahalanobis National Crop Forecast Centre

Ministry of Agriculture & Farmers’ Welfare, Government of India, New Delhi

Crop Forecasts in India and the Role of Weather Predictions

Crop Condition

of August, 2015

Conventional Crop Estimation in India

Area Statistics

• “Land Record States” or temporarily settled states: 17 Major States, 4 UTs: 86% ofreporting area, Timely Reporting Scheme (TRS), 20% villages are selected at random forcomplete area remuneration.

• States where area statistics are collected on the basis of sample surveys. Establishment ofan Agency for Reporting of Agricultural Statistics (EARAS). 9% of reporting area. Samplesurveys of 20% villages/ investigator zones.

• Hilly districts of Assam, rest of the states in NER, & other UTs. Area statistics based onimpressionistic approach. 5% of the reporting area.

Yield Estimates

• Crop Cutting Experiments (CCE) under scientifically designed General Crop EstimationSurveys (GCES).

• Around 950 thousand CCEs.• Stratified multi-stage random sampling: Tehsil / Taluk > Revenue Village> Survey Number /

Field> Experimental Plot (Specified size / shape)• 80-120 CCEs for a crop in a major district

Advance Estimates

• September (1st ), January (2nd ), March/April (3rd ), June/July (4th), January (Final)

S.N. Group Crop Type Crop Season Avg. Area

(M Ha)

Avg. Prod.

(M Ton)

Status (RS

Assessment)

1 Food grains Cereals Rice Kharif +

Rabi

39.1

4.3

85.3

13.6

1

1

2 Wheat Rabi 29.0 87.4 1

3 Coarse

Cereals

Maize Kharif + Rabi 7.2

1.3

15.1

5.3

3

2

4 Sorghum Kharif +

Rabi

2.8

4.2

3.1

3.3

3

1

5 Bajra Kharif 8.7 9.0 3

7 Pulses Tur (Arhar) Kharif 3.8 2.6 2

8 Gram Rabi 8.4 7.8 2

9 Oilseeds Groundnut Kharif 4.6 4.9 3

10 R & M Rabi 6.2 7.3 1

11 Soybean Kharif 10.0 11.9 3

12 Fibre Crops Cotton Kharif 11.0 29.5* 1

13 Jute Kharif 0.8 10.4* 1

14 Sugarcane Sugarcane Long Durn 4.7 324.4 1

Major Crops of India

* Million bales

NASA-ISRO-MoA

1969 JEP1978 CAPE1988FASAL Pilot

1997 FASAL2007 MNCFC2012

46 Years of use of Space Technology in Crop Forecasting

Coconut Root Wilt study in

Kerala

Experimental Studies on Crop Discrimination

Area & production

Estimates of major crops at

State level.

National Wheat, FASAL-Odisha

District-State-National forecasts

using multiple approaches for

multiple forecasts

Institutionalisation of Space

Technologies developed by

ISRO

5

Forecasting Agricultural out put using Space, Agrometeorologyand Land based observations

Crop Forecasting: Salient Features Crops (8)

Rice (Kharif & Rabi), Jute, Cotton, Sugarcane, Wheat, Rapeseed & Mustard,

Sorghum(Rabi), Rabi Pulses

Data

Multidate Microwave (RISAT-1 C band SAR) and Optical (Resourcesat-2 AWiFS & LISS

III) Data

Approach

Ground truth using Smartphone, Acreage using satellite data, Yield using weather and

remote sensing based models

States Covered

All those states, which together contribute >90% of the Crop’s area in the Country

Periodicity (All forecasts are issued pre-harvest)

Paddy, Wheat & Mustard – 3 Forecasts

Sugarcane, Cotton – 2 Forecasts

Jute, Sorghum & Pulses- 1 Forecast

Software

FASALSoft, developed by ISRO

Geospatial Technology for Field Data Collection

>12000 GT points, >1200 CCEs

State Agrl. Dept. Officials

Bhuvan Geoportal

Android based App. by NRSC

Smartphones/Tablets

Sampling plan based on RS data

Groundtruth Crop Cutting Expt

Agromet + Remote Sensing+ Field Observation

Simulation Models (Assimilation of RS data)

Empirical Models (Agromet + Remote Sensing)

Semi Physical Remote Sensing Models: Rice, Wheat

District Level Yield Models: IMD, SAUs

Meteorological Sub Division Level Yield Models: SAC, MNCFC

Crop Yield Forecasts

9

Yield Forecasts by IMD

1. Empirical: Models

• District Level

• Correlation Weighted Regression Models

• Variables: Rainfall, Temperature and Relative Humidity

2. Simulation

1. DSSAT-CERES

2. Calibrated for major varieties

3. One run per district and forecast at State/District level

4. Actual Data upto Forecast Date and rest average weather data

National Agricultural Drought Assessment & MonitoringSystem (NADAMS)

Operational Drought assessment during Kharif using Remote Sensing (Methodologydeveloped by ISRO)

Periodic District/Sub-District level drought assessment for 14 Agriculturally Dominantstates of India (5 at Sub District level)

Satellite based indices, Rainfall data, Ground information on Sowing progression andIrrigation Statistics are used for drought assessment

Drought Warning (Normal, Watch & Alert) is given in June July & August, while DroughtDeclaration (Mild, Moderate & Severe) in September & October

Drought Assessment: Inputs

1. Remote Sensing based indices:

• Normalized Difference Vegetation Index (NDVI)

• Normalized Difference Water Index (NDWI)

• Vegetation Condition Index (VCI)

2. Area Favourable for Crop Sowing (using Satellite

based Index and Soil Moisture Index)

3. District level Rainfall Deviation

4. Irrigation percentage

NDVI from NOAA

Data (1 km)

NDWI from MODIS

Data (500 m)

Rainfall Data from IMD

NDVI from

AWiFS Data (56 m)

District level Irrigation%Soil Moisture Index

28 August 2015

2012 2013 2014

Drought Assessment: September, 2015

Vegetation Condition

Numbers of Districts affected

Category 2012 2013 2014 2015 (upto Sept end)

Normal 316 369 323 247

Mild 43 34 85 149

Moderate 51 8 22 34

Usage by States-Extract from Andhra Pradesh Drought Memorandum

(Feb, 2015)

CHAMAN (Coordinated Horticulture Assessment & Managementusing geoiNformatics) Project: Launched in September, 2014

Horticultural Crops

Area assessment and production forecasting of major horticultural

crops

• Vegetables: Potato, Onion & Tomato

• Fruits: Mango, Citrus & Banana

• Spices: Chilli

Geospatial technologies will be used for generating horticultural

development plans for

i) Site Suitability

ii) Post-Harvest Infrastructure

iii) Crop Intensification

iv) Orchard Rejuvenation

v) Aqua-horticulture

• Exploring the role of Remote Sensingtechnology to supplement the yieldassessment through Crop CuttingExperiments (CCE)

• Launched in September, 2014 underNational Crop Insurance Programme(NCIP)

• Current season remote sensing databased CCE Planning

• Smartphone use for CCE Data Collection& CCE database on Bhuvan Portal

• Multi-level yield modeling using RemoteSensing and Crop Simulation Model

• KISAN Project Launched in Oct, 2015

• Proposed use of UAV/Drones

Selection of CCE sites

using NDVI map

Overlaying of CCE sites

in Google Map for better

identification in the field

Space Technology for Crop Insurance

Impact of Disaster on Agriculture

• Rice-Flooded Area Assessment, post-Phailin Cyclone in Odisha State,October, 2013

• Impact Assessment of Heavy Rainfall and Hailstorm in Northern Indiaduring Feb-Mar, 2015

All India Summer Monsoon: 1st Stage National Level; 2nd Stage 4 Homogenous zones

Met Sub-division wise 5 Day Rainfall (Categories)

NWP: District level 5 Day Forecasts (RF, Temp, RH, Cloud, Wind)

Nowcast: 3 Hour Extreme Weathers

ERFS (under FASAL project): Ensemble Technique (9models), Monthly & Seasonal Forecasts ofPrecipitation, Met Sub-Division level

IMD Weather Forecasts(In relation to Crop Assessments)

Weather Data Requirements for Agriculture

• Agro-meteorological parameters at higher spatial scale, Extended range forecasts (for crop simulation models)

Crop Forecasting

• Rainfall data at Taluk level, Extended Range forecasts, Derived products (MAI, SPI, PDSI, AAI, Soil Moisture), Dry & Wet Spells

Drought Assessment

• Higher accuracy of Short term & Medium term forecasts, withvalue additions at block level; Extreme Weather Events

Farmers’ Advisory

Summary

India has established successful operational programmes for Crop Forecasting and Drought Assessment

These activities are highly dependent upon weather data and forecasts

Though there has been tremendous improvement in weatherforecasting in India, what is further needed is higher accuracy athigher spatial scale, extended range forecasts and derived products.

22

URL: www.ncfc.gov.in

Email: ncfc@gov.in