Home > Geospatial Application Papers > Agriculture & Soil > Crop Production
Overview | Crop Production | Crop Pattern | Crop Yield | Irrigation | Soil Management | Precision Farming | Relevant Products | Relevant Links



Crop Monitoring for Selenge Basin using RS and GIS


Crop Parcel Mapping
Crop boundaries were precisely digitized from panchromatic and multispectral channels of Landsat E.TM image. During digitizing process new fields (small areas of vegetable and potato) were found and added from imagery.

For calculation of parcel area we used ARC/INFO 8.0 GIS software. Of course, the area received from imagery is very close to real size of the field.



An existing parcel maps at scale 1:100000 and 1:25000, received from LMA (Land Management Agency) have been used for inventory of crop parcels. The maps guided by local agronomists were old (1986-1989) and should be renewed. For all soums have been produced:
  • Parcel map by crop type
  • Registration map by users
  • Parcel boundary map
A ground survey had been in the period from 24th of August to 5th of September, 2000. The scope of these visits are the collection of ground observations for photo-interpretation and image classification result, and also parcel information by interviewing the agronomists that are responsible for statistics at soum level.



Comparison of Parcel Acreages
Parcel area calculation differences between statistical data and calculated by GIS environment are shown by charts. Totally 1857 amount of field was identified in image with total area 224076 ha of which 85976 ha sown area, 65050 ha fallow and 71223 ha non-cultivated land. On 2049 ha of field area we haven’t statistical data (by agronomist data).



The data recorded from agronomists and imagery is very close while the difference between official statistics and these data are quite big.

Conclusion
The usefulness of remotely sensed data particularly the Landsat ETM data in conjunction with relevant GIS data and field information has dispensed an important source of information in the assessment and monitoring of crop at the regional level.

As shifting cultivation continues to play a dominant role in pasture degradation from grazing area to dry vegetation and eroded area, forest conservation and agricultural development. Due to economic condition cultivated crop land is decreased, due to market economy and mining area is increased and it is influenced to the land to be come eroded.

Field sizes are large (150-350 ha) and the investigation of cultivated land, mainly wheat, with distinct spectral appearance possibly derived from high resolution data. Remotely sensed data should be confirmed by intensive ground survey.

Inventory of arable and abandoned land and determine the real parcel size is possible. The comparison area results (image data and official statistics) of parcels for 7 soums shows that the methodology of inventory and registration of crop land should be changed and renewed.

References
  1. “Observing and analyzing natural resources” Spot Image, Journal, CNES, France, 1998
  2. “Multispectral Imagery reference guide” Spectral Imagery Training Center, LOGICON Geodynamics, Inc. Fairfax, Virginia, USA, 1997
  3. “Report of Antjie” ISEAM Tacis expert, Ulaanbaatar, 2000
  4. “Report of Fabrizio” ISEAM Tacis expert, Ulaanbaatar, 2000
  5. “National Atlas of Mongolia” Academy of Science Mongolia, Ulaanbaatar, 1990
  6. “Land Cover Assessment and Monitoring” UNEP, Bangkok, 1995
  7. “Asian – Pacific Remote Sensing and GIS Journal” Volume 10, ESCAP, 1998
Page 2 of 2
| Previous |