Estimating forage quantity and quality under different stress and senescent biomass conditions via spectral reflectance

Abstract

Assessing forage quantity and quality through remote sensing can facilitate grassland and pasture management. However, the high spatial and temporal variability of canopy conditions may limit the predictive accuracy of models based on reflectance measurements. The objective of this work was to develop this type of models, and to challenge their capacity to predict plant properties under a wide range of environmental conditions. We manipulated Paspalum dilatatum canopies through different stress treatments (flooding, drought, nutrient availability, and control) and by artificially varying the amount of senescent biomass. We measured canopy reflectance and constructed simple models, based on either normalized vegetation indices or a few selected wavebands, to estimate biomass and two variables related to forage quality: proportion of photosynthetic vegetation and biomass C:N ratio. General models satisfactorily predicted plant properties for the whole set of environmental conditions, but failed under specific conditions such as drought (for estimates of plant biomass), fertilization (for estimates of C:N ratio), and different levels of senescent tillers (for estimates of the proportion of photosynthetic vegetation). Where general models failed, specific models, based on different bands, achieved satisfactory accuracy. The general models performed better when based on a few selected bands than when based on two-band vegetation indices, having better accuracy (higher R 2 ) and parsimony (lower BIC). However specific models performed similarly for both approaches (similar R 2 and BIC). These results indicate that these plant properties can be predicted from reflectance information under a broad range of conditions, but not for some particular conditions, where ancillary data or more complex models are probably needed to increase predictive accuracy.

Modeling Nutrition Quality and Storage of Forage Using Climate Data and Normalized-Difference Vegetation Index in Alpine Grasslands

“…These findings indicate that it is hard to examine the exclusivity relationships between forage nutrition and climate variables based on simulated forage nutrition from remote-sensing data alone. Fourth, previous studies have mainly estimated the nutrient quality and pool of forage via exponential regression, stepwise linear regression, partial least-squares regression and the derivative transformation technique for hyperspectral data [25][26][27]. As an increasingly mature technology, big-data-mining technology (e.g., random-forest models, support-vector machines and recursive-regression trees) has higher precision and data processing ability than other methods [28].…”

The Relationship between Satellite-Derived Vegetation Indices and Live Weight Changes of Beef Cattle in Extensive Grazing Conditions

“…Satellite imagery with high a spatial and temporal resolution allows the calculation of VIs, which can be correlated to pasture quantity and quality [3,4]. The growth and weight of cows are of interest because they impact cows' reproductive efficiency, including their ability to conceive, gestate, and raise a calf to weaning [5,6].…”

Predicting Forage Quality of Grasslands Using UAV-Borne Imaging Spectroscopy

“…The most frequently applied statistical modelling method is the linear regression (simple linear, step-wise linear) with selected highly correlated spectral features, such as wavebands [20,27], normalised difference spectral indices (NDSIs) [18,24,28], spectral ratios (SRs) [15,29], and other well-known vegetation indices (e.g., NDVI, SAVI, NDRE) [17,23]. Predictive modelling (also known as machine learning) algorithms, such as partial least squares regression (PLSR) [16,27,[30][31][32], random forest regression (RFR) [24,33,34], and artificial neural network [20,21,35], were employed to estimate forage quality parameters using highly correlated spectral reflectance data. Predictive modelling algorithms frequently enhanced the predictive capability compared with the simpler linear regression models [24,34].…”