Multispectral Remote Sensingto Distinguish the Little Seed CanaryGrass Phalaris Minor) from Wheat CropUnder Field Conditions for EnvironmentalSustainability and Precision WeedManagement

摘要

To take full advantage of site-specific variable rate technology (VRT) systems, highly accurate digital mapping of weed infestations within fields via scouting, GPS, GIS and remote sensing technologies will be necessary. When combined, these tools can increase weed control efficiency and reduce herbicide use and residues, thereby avoiding excess applications that lead to increased costs, potential herbicide resistance in the field and runoff into the environment. Keeping this in view, a field experimentwas conducted at the Research Farm, Department of Agronomy, Punjab Agricultural University, Ludhiana, Punjab, for 2 years to study the multispectral remote sensing to distinguish the little seed canary grass (Phalaris minor) from wheat crop under fieldconditions for environmental sustainability and precision weed management. The experimental site during both the seasons were sandy loam in texture, with normal soil reaction and electrical conductivity, low in organic carbon and available nitrogen and medium in available phosphorus and potassium. The experiment consisted of five treatments, viz, T_1 control (weedy check); T_2, half of the recommended dose of herbicide for partial control of Phalaris minor, T_3, recommended dose of herbicide to obtain economic threshold level to control Phalaris minor, T_4, manual weeding (partial), done after a month of sowing of crop; and T5, weed free (manual). The treatments T_3 (recommend dose of herbicide) and T_5 (weed free, manual) being at par with one anotherrecorded highest plant height, dry matter accumulation and number of tillers per plant by wheat at all observational dates during both the years in experiments, whereas minimum plant height, dry matter accumulation and number of tillers per plant were recorded in control treatment (T_1). Reduction in dry matter production,number of tillers as well as effective number of tillers and ultimately yield of wheat are mainly attributed to the reduction in the number of effective tillers, lesser number of grains per spike and lesser 1000-grain weight. The weed-free treatments (T_3 and T_5) had lower red reflectance percentage as compared to other weed control treatments. The control treatment recorded the highest red reflectance. On the other hand, the two weed-free treatments T_3 and T_5 had higher IR reflectance percentage ascompared to other three weed control treatments, and the lowest IR reflectance was recorded under control treatment. Highest RR and NDVI values were obtained in treatments T_3 and T_5 where there was no competition between wheat and weeds, and control treatment had the lowest RR value amongst all the treatments during both years. Differences in RR between these three treatments are mainly due to dark green colour of wheat, more leaf area index (LAI) and more biomass of wheat as compared to Phalaris minor. The RR value increases in the early stages of crop growth which is maximum at maximum crop canopy cover and after that decreases as the leaves senesce. The highest RR values were obtained at 95 days after sowing almost in all the treatments. It is feasible to distinguish pure wheat from weeds just 34 days after sowing, but amongst different weed control treatments, i.e. pure Phalaris minor plot and less/partial Phalaris minor weeds 52 days after sowing amongst themselves, and they remain distinguished up to 107 days after sowing based on their NDVI values. After 52 DAS, the differences in the NDVI of different weed control treatments were very clear. So, from such type of information, we can discriminate/ define the areas which are heavily or partially infested with weeds so that timely weed control measures can be taken which can help the farmers in preventing yield losses due to weeds.