Temperature 35 C New Delhi, IN
Monday, April 19, 2021
No Image

Geospatial Technology Applications in Crop Insurance

Innovations and investments in agriculture sector are possible only when insurance mechanism exists. Crop insurance has become an indispensable risk reducing instrument in Agriculture which is exposed to multiple hazards leading to frequent crop losses.

 

C S Murthy and PVN Rao

 

Agricultural risk sharing through crop insurance has been in existence in many counties, in many forms and for many decades. India has a long history in the design, development and implementation of various crop insurance schemes by making improvements from time to time, to insulate the farming community against various cultivation risks.

 

Index based insurance schemes either weather or yield are  mostly adopted in agriculture and these schemes are faced with serious problems of basis risk and non-availability of historical data, making the scalability and sustainability of  agricultural insurance in the long run, a big challenge.  Therefore, the need for developing innovative crop insurance products   has been widely recognized by both developed and developing nations in recent years.

 

Pradhan Mantri Fasal BeemaYojana (PMFBY), being implemented in most parts of the country, from kharif 2016,  is a revolutionary step towards improving agriculture  insurance in the country.   Small field sizes, high variability in yields and frequent weather aberrations have made technology application inevitable for implementing crop insurance.   PMFBY guidelines (available at agricoop.gov.in) have indicated that use of technologies in crop insurance is mandatory. Remote Sensing, Mobile, Geospatial tools and drones have been identified to be potential technologies to support crop insurance system in the country.

 

Technology interventions can take place in different  segments of crop insurance value chain;  (1)  insurance rate making  by using satellite based agriculture risk maps, (2) monitoring and expansion of insurance coverage and timely settlement of claims based on cadastral layers, mobile apps and spatial analytics, (3) prevented/failed sowing risk assessment using satellite and weather indices, (4) crop mapping  and area discrepancy analysis using moderate-high resolution satellite data, (5) mid-season adversaries based on satellite indices and field information, (6) natural calamities assessments using satellite based flood and drought information, (7) localized risk with weather data and Mobile collected field data, (8) Crop Cutting Experiments for crop yield estimation with mobile based field data and satellite derived indices.

 

Crop classification using satellite images is one of the proven applications of remote sensing technology and is being practiced for many years in different countries. Space images can be used to identify major crop types permitting crop distribution analysis in the insurance units.  Such analysis is useful to resolve area discrepancy issues there by reducing moral hazard. Multi-temporal and moderate resolution satellite data during the crop season is required for this analysis. Microwave satellite data may be preferred for kharif crops when optical data is cloud covered.

 

Measurement of crop yield in the insurance units for indemnity fixation is a challenge. Availability of reliable, current and historical crop yield data, particularly in developing countries pose a serious challenge for objective assessment of crop loss and indemnity. In India, crop yield estimation in the insurance units is done by conducting Crop Cutting Experiments (CCE) in the field-plots selected through a sampling scheme. Subjectivity in the yield measurements has become a major concern and it is the most agreed-upon view that the crop yield estimation needs to be improved to enhance the strength of the crop insurance contracts. The technology interventions such as the use of  satellite data to improve crop yield estimation is largely recommended. Currently available satellite data sets, weather data sets, geospatial tools and techniques suggest the scope for improving the crop insurance loss assessment methodologies.

 

It is largely recognized that the mechanism of CCE needs improvements from different perspectives to make the yield estimates closer to the ground reality. Strengthening the CCE process is the immediate priority to make the crop insurance sustainable in the near future.

 

Strategies for improving crop yield estimation in the insurance units should  include three steps namely (1) enhance the  transparency and objectivity in the CCE process by way of using mobile apps,  (2) Smart sampling on the basis of yield proxies to improve the sampling design in terms of reduced sample size and improved distribution of the sample plots and (3) replace the CCE with alternate mechanism.

 

Mobile Apps  are being used extensively by most of the states for  recording CCE yield data, from 2016-17 kharif season.  Thus the process of CCE has become transparent in the country. Research on developing crop yield proxies is in progress at NRSC. A composite yield  index for jowar  and cotton crops was developed using rainfall, rainy days, soil, irrigation parameters and satellite based crop condition. While the initial observations are encouraging, the index is being validated and standardised for operational use. The yield proxies are useful to improve the sampling design of CCE.

 

Crop yield estimation methods using remote sensing inputs are of three categories – empirical, semi-empirical and simulation models. Remote sensing derived NDVI which represents crop vigour has been correlated with yield to investigate the possibility of developing crop yield index for crop insurance (Fig.1 and Fig.2). Considering the limitations of NDVI, some studies have recommended the use of bio-physical variables derived from satellite data to develop index based crop insurance schemes. Local weather conditions, crop management practices, soil, variety/hybrid, water related parameters, etc. are important yield determinants but their effect is not completely manifested in any single index.  Therefore, semi empirical techniques are being developed involving spectral indices, weather data and local crop growing conditions. Adopting crop simulation models calls for very intensive field data on different variables, calibrations etc limiting its scalability. Thus, alternate methods for crop yield estimation are still in development phase and hence replacement of CCE with alternate mechanism is yet to be realised.

 

NRSC (ISRO) has taken several initiatives in recent years to demonstrate the technology capabilities to meet the information requirements of crop insurance. These initiatives  include (a) pilot studies in different districts, (b) development and implementation of Mobile technology for field data collection for improving crop cutting experiments, yield estimation and crop  loss assessment, (c) training to the field level personnel of State Departments on mobile based field data collection, (d) collaborative studies with Agricultural Insurance Company of India  Limited to improve crop insurance with remote sensing and GIS technologies, (e) awareness-cum-training to the industry on technology utilisation in September 2016, (f) development of a Decision Support System for crop insurance for Odisha state and (g) conducting special studies to support the States.

 

Crop Insurance Decision Support System (CIDSS) is a web-enabled spatial Decision Support System with multi-source data, information products and services to address different components in the crop insurance value-chain,  being developed by NRSC for the Department of Agriculture, Government of Odisha. It is a single window system  for all technology interventions in crop insurance and immensely benefit the State Government for improving the crop insurance mechanism. Integration of different satellite based products  and other datasets, mobile collected field data and Crop Cutting Experiments data  using automated or semi-automated procedures  and development of decision rules for visualisation and computation to support the crop insurance requirements is the main thrust in the proposed system. Evidence based verification of CCE data quality with  decision tools is in progress. It is also envisaged to provide training to the State for effective use of Decision Support System on operational basis. It is a major initiative to facilitate the states with the use of technologies in crop insurance.

 

Resourcesat 2 – LISS III and LISS IV, Landsat 8 OLI, Sentinel MSI, Planet Scope data are suitable for deriving crop insurance related information.  With slightly coarser spatial resolution at 56m, Resourcesat 2 AWiFS has unique advantages of higher frequency of observation and larger swath making it a potential dataset for generating crop surveillance and performance information.  Synergistic use of all available data sets along with Mobile Apps based field data collection and data analytics together may  roughly cost Rs.5-10 per ha. for providing crop assessment and monitoring information products to support crop insurance.

 

There are enormous opportunities for utilisation of satellite remote sensing and GIS technologies to improve different segments of crop insurance value chain. Currently, technology utilisation is at low scale in the crop insurance. Many insurance companies and states have started the adoption of remote sensing and drones data along with Mobile app collected field observations. In the years to come, technology utilisation in crop insurance is expected to grow exponentially.

 

A large number of Organisations are associated with data collection in crop insurance leading to big pool of  multi-source data. Spatial analytics, data mining, data engineering, evidence based tools etc. could churn out useful information products to make the crop insurance mechanism more robust and transparent.  There is scope for developing new business models with the support of technologies to boost the agriculture insurance in the country. Considering the business volumes in crop insurance, technology infusion costs are very minimal indicating the economic viability of such business models.

 

Thus technologies are going to be game changers in the crop insurance and benefit all stake  holders.

 

Awareness/training requirements of stake holders for using technologies and the research agenda for the technology institutions for meeting the information requirements need to be simultaneously initiated to ensure sustained use of technologies in crop insurance.

 

Increasing crop production risks in agriculture coupled with the alarmingly lower growth and outreach of crop insurance contracts signify the huge potential for crop insurance in India and elsewhere in the world. Effective Implementation of crop insurance in developing countries like India is still a major challenge.

 

Currently, technology utilisation is at its infancy in the crop insurance and hence there is a need for a specific road map for integrating technologies with crop insurance. Some of the operational capabilities of the technology such as satellite based crop mapping, crop monitoring, natural calamities assessment and mobile based field data collection can be readily adopted and linked to crop insurance.

 

(The authors are working at National Remote Sensing Centre (ISRO).Hyderabad. Views expressed are personal.)