Authors :
Milan KC; Dr. Krishna Prasad Bhandari; Bikash Sherchan
Volume/Issue :
Volume 7 - 2022, Issue 9 - September
Google Scholar :
https://bit.ly/3IIfn9N
Scribd :
https://bit.ly/3S3kCoX
DOI :
https://doi.org/10.5281/zenodo.7212619
Abstract :
Rice is largely grown type of crop in Nepal
and has an important role in food security and nutrition.
Therefore proper and timely monitoring, forecasting,
and prediction are necessary for planning and
management purposes for governmental and nongovernmental bodies that have as been working for food
security. So the forecasting and estimation of crops and
monitoring of different phases of the crop are necessary
for food and crop management. For proper monitoring,
remote sensing imagery and NDVI series are capable of
the identification of crop growth and health as well as
crop yield prediction model development. This research
reflects a model for paddy monitoring using Sentinel 2
imagery of every 5 days interval by extracting NDVI
series. The land pooling area of seratar, and Nahala,
Bandipur Rural Municipality of Nepal has been selected
as a study area for understanding and analyzing the
NDVI value of different phonological stages of paddy
with different land management factors like water level,
level of damages, soil type, fertilizers used, date of
transplantation, amount of seed, sowing date, etc.
Google Earth Engine (GEE) is an ideal platform
that is used to study and extract the NDVI of multiple
sentinel images of study areas. Importantly, GEE
reduces time and space for data processing and analysis.
The hectic process of getting satellite imagery from a
different sources, the downloading process of large
images, their correction and classification collectively
known as remote sensing processes can be resolved with
this open source platform.
By using the time series NDVI value, we can assess
the correlation between the NDVI values and land
management factors. Along with the paddy monitoring,
the rice yield prediction and estimation model is
developed using the regression model. The model
classifies only those parameters which are highly
correlated to the NDVI values. And from that, we are,
able to develop the crop yield prediction model. Besides
that, we have prepared the rice distribution map, LULC
map, and agriculture map from the data that we
collected from the field survey. And there is also a socioeconomic analysis of crop yield model in the scenario of
world and Nepal. It reflects its importance in planning,
policy-making maintaining the demand-supply chain in
distribution of agriculture production etc.
Keywords :
Estimation, LULC, time series, Regression, NDVI, GEE.
Rice is largely grown type of crop in Nepal
and has an important role in food security and nutrition.
Therefore proper and timely monitoring, forecasting,
and prediction are necessary for planning and
management purposes for governmental and nongovernmental bodies that have as been working for food
security. So the forecasting and estimation of crops and
monitoring of different phases of the crop are necessary
for food and crop management. For proper monitoring,
remote sensing imagery and NDVI series are capable of
the identification of crop growth and health as well as
crop yield prediction model development. This research
reflects a model for paddy monitoring using Sentinel 2
imagery of every 5 days interval by extracting NDVI
series. The land pooling area of seratar, and Nahala,
Bandipur Rural Municipality of Nepal has been selected
as a study area for understanding and analyzing the
NDVI value of different phonological stages of paddy
with different land management factors like water level,
level of damages, soil type, fertilizers used, date of
transplantation, amount of seed, sowing date, etc.
Google Earth Engine (GEE) is an ideal platform
that is used to study and extract the NDVI of multiple
sentinel images of study areas. Importantly, GEE
reduces time and space for data processing and analysis.
The hectic process of getting satellite imagery from a
different sources, the downloading process of large
images, their correction and classification collectively
known as remote sensing processes can be resolved with
this open source platform.
By using the time series NDVI value, we can assess
the correlation between the NDVI values and land
management factors. Along with the paddy monitoring,
the rice yield prediction and estimation model is
developed using the regression model. The model
classifies only those parameters which are highly
correlated to the NDVI values. And from that, we are,
able to develop the crop yield prediction model. Besides
that, we have prepared the rice distribution map, LULC
map, and agriculture map from the data that we
collected from the field survey. And there is also a socioeconomic analysis of crop yield model in the scenario of
world and Nepal. It reflects its importance in planning,
policy-making maintaining the demand-supply chain in
distribution of agriculture production etc.
Keywords :
Estimation, LULC, time series, Regression, NDVI, GEE.
