We work in groundwater dependent region. Agriculture has been the largest consumer of the groundwater. In order to manage the groundwater, we have to manage agricultural groundwater use. In order to manage agricultural groundwater use, we have to monitor the cropping pattern in the village. What is the most accurate and cost effective method to find out the village level cropping pattern (season wise acreage of each crop grown)?
Does remote sensing provide any solution?
Is physical survey the feasible solution? Or is there any other solution?
If you have conducted any such study, please share the publication.
Nikhil -
I don't know about remote sensing, but if you want to keep track of the production volume for each of the crops in a region, perhaps you might do this the way electric sales in different economic sectors, and some other energy establishment surveys, are conducted for official statistics as in the references given in the following presentation:
Invited presentation for mathematical statisticians at the US Energy Information Administration in 2017:
https://www.researchgate.net/publication/319914742_Quasi-Cutoff_Sampling_and_the_Classical_Ratio_Estimator_-_Application_to_Establishment_Surveys_for_Official_Statistics_at_the_US_Energy_Information_Administration_-_Historical_Development, using prediction.
The idea is to do relatively frequent sampling, and occasionally do a census. The approach is prediction-based (meaning regression is used - here, very simple regression). The less frequent census is used to acquire predictor data on the same crops as in the samples. In energy data noted here, there may be monthly sampling and an annual census. For the case of this question about crops, the census might be every five or perhaps eight years, and the sample for each of the other years, whatever works well enough considering the burden required. Perhaps you could stagger what crop is censused so that, say if there were 10 crops, you could census two each year, and sample the others. In the first years before you have censused everything, you could try using the crops you are censusing as predictor data for other crops, though that would not work nearly as well, and you might need much larger samples.
NOTE: The approach here works best if you only need total production, and just a very few of the largest farms have most or almost all of the production. Of course, some farms will be bigger for some crops than others.
Also, you might note the following:
https://www.researchgate.net/publication/262066356_Quasi-Cutoff_Sampling_and_Simple_Small_Area_Estimation_with_Nonresponse using prediction
The following project and its updates might be of interest:
https://www.researchgate.net/project/Cutoff-and-quasi-cutoff-sampling-with-prediction-for-Official-Statistics
If a ratio estimator is helpful, the following is with regard to one population, subpopulation, or group:
https://www.researchgate.net/publication/261947825_Projected_Variance_for_the_Model-based_Classical_Ratio_Estimator_Estimating_Sample_Size_Requirements
I hope something here is of some use to you.
Cheers - Jim
Land use Land cover mapping using ARC GIS or ERDAS software is the best method for monitoring of agricultural crops. As well as google earth is the best option for monitoring the crops on spatial manner.
LULC mapping using satellite images (Landsat, sentinel images). You can use ArcGIS, ERDAS, QGIS software for that. And also you can use Google earth data for accuracy assessments.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India