| Spatial distribution in soil fertility is important to agriculture and precision farming. Reflectance measurements have the potential to detect crop growth and nutrient deficiencies. Field and controlled environment experiments were conducted to evaluate spatial variability in soil chemical properties influencing cotton growth, and the relationship between N nutritional status and leaf and canopy reflectance.; Spatial correlations in total soil N and C, extractable Ca, K, Mg, Na, P, sum of exchangeable bases (SEB), and pH were characterized by different semivariogram types and ranges. Spatial structures for total soil C, total soil N, K, Na, P and pH were characterized by exponential semivariograms with ranges of 357, 318, 255, 186, 534 and 609 m respectively. Spatial variation of Ca, Mg and SEB were described by spherical models with corresponding ranges of 165, 481, and 470 m. The nutrients' spatial pattern was detected by systematic sampling and random composite samples. Semivariograms for cotton lint yield had a range of 32 m within a mapped soil unit. Soil available N up to 0.30 m deep was positively correlated with plant height, and negatively correlated with clay content. Areas with greater clay content had greater total soil N and C. Faster and more complete cotton (Gossypium hirsutum L.) residue mineralization may take place in coarser soils.; Nitrogen deficiency increased leaf reflectance across the whole measured spectrum under greenhouse conditions and within certain ranges under field conditions. Reflectance near 700 nm was found to be related to leaf N concentration and lint yield. The correlation between the first derivative values of leaf reflectance at the red edge and leaf nitrogen concentration was not always significant. The greatest coefficient of determination (0.43) for the relationship between canopy reflectance and lint yield was found at 540 nm. The Normalize Difference Vegetation Index (NDVI) was positively correlated with soil available N, plant height and yield.; The combination of ground observations, spectral measurements and remote sensing, is a practical way to characterize cotton N nutritional status. These results provide information needed for the development of precision agriculture, and the use of variable rate N fertilization. |
