Smart Agriculture

How satellite data can help farmers to improve crop yields

Irrigation is essential for agriculture, but it can be challenging to manage effectively.
Traditional methods, such as relying on rainfall data or physical inspections, are often time-consuming and more importantly forecasting future requirements require deiled numerical models for weather and crop yield forecasts.
Satellite data offers a revolutionary solution for irrigation management, providingtimely insights into vegetation health and moisture content

Using Technology to Forecast Crop Health and Moisture Levels

The use of technology to improve the efficiency and sustainability of agricultural production.

NDVI is a measure of the amount of vegetation in a given area.
Soil moisture content is important for crop growth and development.
Farmers can use technology to forecast crop health and moisture levels.

This can help to improve crop yields, reduce water use, and protect the environment

NDVI and NDMI are two commonly used vegetation indices that can be used to assess vegetation health and water status.

NDVI is a ratio between the red and NIR bands.
NDMI is a ratio between the reflectance in the blue band and the SWIR band

Both indices can be used to identify areas of stress, such as drought or nutrient deficiency. They can also be used to track changes in vegetation over time.

NDVI is more sensitive to changes in vegetation greenness, while NDMI is more sensitive to changes in vegetation water content
NDVI is a versatile and powerful tool for improving agricultural practices and promoting sustainable water management.

Analysis at Nams Lanka Farm

The study area is located in Sri Lanka and is shown in below Figure 1

The study area is located in Sri Lanka and is shown in below Figure 2

The study analyzed small agricultural subplots shown in Figure 1 for NDVI forecasting for different vegetation types. Using LSTM models calibrated with past data, 2-3 NDVI 2-3 weeks a head can be predicted.
Models calibrated with different irrigation supply provides a mechanism for estimating irrigation requirement.

SCL (Scene Classification) data are used to filter out pixels that are covered by clouds or shadows
Missing data were interpolated based on the surrounding values.
Other satellite data, such as Landsat-8 and MODIS, will be used to compensate for cloud cover in Sentinel-2 data
Efficient methods for NDVI forecasting for larger areas based on clustering of similar agricultural fields will be developed