Abstract
Grasslands are an important part of pre-Alpine and Alpine landscapes. Despite the economic value and the significant role of grasslands in carbon and nitrogen (N) cycling, spatially explicit information on grassland biomass and quality is rarely available. Remotely sensed data from unmanned aircraft systems (UAS) and satellites might be an option to overcome this gap. Our study aims to investigate the potential of low-cost UAS-based multispectral sensors for estimating above-ground biomass (dry matter, DM) and plant N concentration. In our analysis, we compared two different sensors (Parrot Sequoia, SEQ; MicaSense RedEdge-M, REM), three statistical models (Linear Model; Random Forests, RF; Gradient Boosting Machines, GBM) and six predictor sets (i.e. different combinations of raw reflectance, vegetation indices, and canopy height). Canopy height information can be derived from UAS sensors, but was not available in our study. Therefore, we tested the added value of this structural information with in-situ measured bulk canopy height data. A combined field sampling and flight campaign was conducted in April 2018 at different grassland sites in Southern Germany to obtain in-situ and the corresponding spectral data. The hyper-parameters of the two machine learning (ML) approaches (RF, GBM) were optimized and all model set-ups were run with a six-fold cross-validation. Linear models were characterized by very low statistical performance measures, thus were not suitable to estimate DM and plant N concentration using UAS data. The non-linear ML algorithms showed an acceptable regression performance for all sensor-predictor set combinations with average (avg) R2cv of 0.48, RMSEcv, avg of 53.0 g m2 and rRMSEcv, avg of 15.9 % for DM, and with R2cv, avg of 0.40, RMSEcv, avg of 0.48 wt.% and rRMSEcv, avg of 15.2 % for plant N concentration estimation. The optimal combination of sensors, ML algorithms and predictor sets notably improved the model performance. The best model performance for the estimation of DM (R2cv = 0.67, RMSEcv = 41.9 g m2, rRMSEcv = 12.6 %) was achieved with a RF model that utilizes all possible predictors and REM sensor data. The best model for plant N concentration was a combination of a RF model with all predictors and SEQ sensor data (R2cv = 0.47, RMSEcv = 0.45 wt.%, rRMSEcv = 14.2 %). DM models with the spectral input of REM performed significantly better than those with SEQ data, while for N concentration models it was the other way round. The choice of predictors was most influential on model performance, while the effect of the chosen ML algorithm was generally lower. The addition of canopy height to the spectral data in the predictor set significantly improved the DM models. In our study, calibrating ML algorithm improved the model performance substantially, which shows the importance of this step.
Highlights
Grasslands are import ecosystems covering about 40% of the global land area (White et al, 2000)
Parameter 385 proposals are less fluctuating for Random Forest (RF) than for Gradient Boosting Machines (GBM) as shown in the distance between consecutive parameter proposals (Supplementary Fig. SF2). 3.4 Model results Our results indicate that the machine learning (ML) algorithms are substantially better than the linear models in estimating DM and plant N concentration (Table 4, Table 5, Supplementary Table ST3)
This study aimed to develop, assess, and apply models to estimate DM and plant N concentration of pre-Alpine grasslands on the field-scale with unmanned aircraft system (UAS)-based multispectral data and canopy height information
Summary
Grasslands are import ecosystems covering about 40% of the global land area (excluding Antarctica and Greenland) (White et al, 2000). (Pre-)Alpine grassland ecosystems provide a variety of goods and services (Egarter Vigl et al, 2016) such as food and forage for livestock production, leading to a high economic value (Egarter Vigl et al, 2018; Gibson, 2009; White et al, 2000). Human intervention proofed to be an important driver to changing ecosystem functioning in managed (pre-)Alpine grasslands (Rossi et al, 2020; Schirpke et al, 2017; Spiegelberger et al, 2006; Walter et al, 2012). N uptake in relation to fertilization rates represents an important measure for optimizing grassland management on farm and regional scale, as decision-making is getting more and more complex due to legislation and climate change (e.g. drought 60 effects). Spatially explicit and accurate information on grassland biomass and quality at field and regional scale is lacking. Robust and reliable methods and applications for grassland monitoring are needed, which ideally scale well and are cost-effective
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
