Modeling the sugarcane crop yield by using a composite model based on remote sensing data

The growing world population needs more food, with less water available for agriculture. This pinching situation can only improve if water is managed more effectively leading to increased crop yield per unit of water consumed. Crop yield is the ultimate indicator for describing agricultural response to water resources management. The need to monitor crop growth and assess the relationships between crop yield and hydrological processes is elementary for improving the productivity of water. Crop yield forecasts a few months before harvest can be of paramount importance for timely initiating food trade secure the national demand and timely organize food transport within countries. Managers and policy markers need information about agricultural products yield at different scales for devise suitable management strategies. These strategies include product prices and market in import or export. But always estimating crop yield due to the lack of sufficient ground information and existing problems was very difficult and costly. The most appropriate strategies are using satellite data and remote sensing. This study was conducted to evaluate the sugarcane crop yield using landsat 8 satellite data in 2013.
In this study a composite model was used to estimate the volume of sugarcane crop. This combined model includes Monteith model, Carnegie Institution Stanford model and the surface energy balance algorithm for land (SEBAL). Monteith’s model is used for the calculation of absorbed photosynthetically active radiation (APAR), the Carnegie Institution Stanford model is used for determining the light use efficiency, and the surface energy balance algorithm for land (SEBAL) is used to describe the spatio _ temporal variability in land wetness conditions. The new model requires a crop identification map and some standard meteorological measurements as inputs. In surface energy balance algorithm for land (SEBAL) method, all fluxes of the energy balance at the earth's surface including net radiation, soil heat flux, and sensible heat flux are calculated from satellite images and finally evaporative fraction is computed based on the energy balance at the earth's surface. The accumulation of biomass is according to the Monteith model proportional to accumulated APAR. Yield mapping was conducted with the implementation of this algorithm in 2013 and with the usage of 10 landsat 8 satellite images. The biomass development and crop yield computations have been executed in a GIS environment. The annual cycle has been split up into 10 discrete intervals to comply with the 10 Landsat 8 images. The time step varies, depending on cloudiness, and is 16 days on average. Each image is representative for a discrete time interval. All calculations are repeated for 10 different Intervals in an independent manner. The total biomass development of a crop lifetime is approximated by integrating the biomass growth over the cropping season using the cropping calendar.
The yield obtained from satellite images was compared and eraluated with the actual measured yield product in the to sugarcane field. Average sugarcane crop yield estimated to 56 tons per hectare. The yield estimation by the composite model revealed correlation and good distribution of farms actual yield. Then the effect of age and varieties on the model accuracy rate for sugarcane yield estimating was examined. It was found among the different varieties, calculated yield in the varieties cultivated fields CP57- 614 have correlation with the yield actual volues that is due to prematurity and better match to the last image. It was also observed with the aging of sugarcane cultivation until the fourth raton, yield dropped, the estimated yield decreased, solidarity and distribution becomes less that the sugarcane actual yield and correlation value is reduced to %51