Inversion of winter wheat leaf area index from UAV multispectral images: classical vs. deep learning approaches.

Precise and timely leaf area index (LAI) estimation for winter wheat is crucial for precision agriculture. The emergence of high-resolution unmanned aerial vehicle (UAV) data and machine learning techniques offers a revolutionary approach for fine-scale estimation of wheat LAI at the low cost. While machine learning has proven valuable for LAI estimation, there are still model limitations and variations that impede accurate and efficient LAI inversion. This study explores the potential of classical machine learning models and deep learning model for estimating winter wheat LAI using multispectral images acquired by drones. Initially, the texture features and vegetation indices served as inputs for the partial least squares regression (PLSR) model and random forest (RF) model. Then, the ground-measured LAI data were combined to invert winter wheat LAI. In contrast, this study also employed a convolutional neural network (CNN) model that solely utilizes the cropped original image for LAI estimation. The results show that vegetation indices outperform the texture features in terms of correlation analysis with LAI and estimation accuracy. However, the highest accuracy is achieved by combining both vegetation indices and texture features to invert LAI in both conventional machine learning methods. Among the three models, the CNN approach yielded the highest LAI estimation accuracy (R 2 = 0.83), followed by the RF model (R 2 = 0.82), with the PLSR model exhibited the lowest accuracy (R 2 = 0.78). The spatial distribution and values of the estimated results for the RF and CNN models are similar, whereas the PLSR model differs significantly from the first two models. This study achieves rapid and accurate winter wheat LAI estimation using classical machine learning and deep learning methods. The findings can serve as a reference for real-time wheat growth monitoring and field management practices.

Altitude explains insignificant autumn phenological changes across regions with large topography relief in the Tibetan Plateau.

The start of the growing season (SGS) and the end of the growing season (EGS) are widely employed in global change studies to represent the spring and autumn phenology, respectively. Despite the Tibetan Plateau (TP) experiencing significant warming in recent decades, EGS has exhibited only slight changes. Previous studies have concentrated on exploring the environmental regulation of phenology, ignoring the distinctive influences of elevation. Therefore, a more in-depth investigation into the underlying mechanism is warranted. In this study, we investigate the variability of EGS among alpine vegetation regions at different elevations and conduct an analysis based on satellite data. Phenology data of alpine vegetation are extracted from SPOT NDVI dataset spanning from 1999 to 2018, using a piecewise-logistic-maximum-ratio method. We analyze the factors influencing EGS trends at different elevations. The results show that the overall insignificant variation in EGS is mainly attributed to altitude. With the altitude increasing, the annual mean EGS experiences a delay of 0.28 days/100 m below 3500 m, while it advances by 0.2 days/100 m above 3500 m. The opposing shift in elevation below and above 3500 m leads to this counteraction. Elevation emerges as the predominant factor influencing EGS trends, explaining the highest variations (38 %), followed by SGS (22 %) and precipitation (22 %). The elevation effect is most pronounced in areas with substantial topography fluctuations. Moreover, the elevation lapse rate of EGS (ELR_EGS) exhibits an opposite trend with growing season (GS) temperature and a similar trend with GS precipitation between the regions below and above 3500 m, ultimately linking to this counteraction. This study underscores elevation is a critical regulator of vegetation EGS responses to climatic changes over the TP, revealing significant spatial heterogeneities in these responses.

The confounding effect of snow cover on assessing spring phenology from space: A new look at trends on the Tibetan Plateau.

The Tibetan Plateau is the highest and largest plateau in the world, hosting unique alpine grassland and having a much higher snow cover than any other region at the same latitude, thus representing a "climate change hot-spot". Land surface phenology characterizes the timing of vegetation seasonality at the per-pixel level using remote sensing systems. The impact of seasonal snow cover variations on land surface phenology has drawn much attention; however, there is still no consensus on how the remote sensing estimated start of season (SOS) is biased by the presence of preseason snow cover. Here, we analyzed SOS assessments from time series of satellite derived vegetation indices and solar-induced chlorophyll fluorescence (SIF) during 2003-2016 for the Tibetan Plateau. We evaluated satellite-based SOS with field observations and gross primary production (GPP) from eddy covariance for both snow-free and snow covered sites. SOS derived from SIF was highly correlated with field data (R2 = 0.83) and also the normalized difference phenology index (NDPI) performed well for both snow free (R2 = 0.77) and snow covered sites (R2 = 0.73). On the contrary, normalized difference vegetation index (NDVI) correlates only weakly with field data (R2 = 0.35 for snow free and R2 = 0.15 for snow covered sites). We further found that an earlier end of the snow season caused an earlier estimate of SOS for the Tibetan Plateau from NDVI as compared to NDPI. Our research therefore adds new evidence to the ongoing debate supporting the view that the claimed advance in land surface SOS over the Tibetan Plateau is an artifact from snow cover changes. These findings improve our understanding of the impact of snow on land surface phenology in alpine ecosystems, which can further improve remote sensing based land surface phenology assessments in snow-influenced ecosystems.

Temperature-mediated responses of carbon fluxes to precipitation variabilities in an alpine meadow ecosystem on the Tibetan Plateau.

Effects of climate warming and changing precipitation on ecosystem carbon fluxes have been intensively studied. However, how they co-regulate carbon fluxes is still elusive for some understudied ecosystems. To fill the gap, we examined net ecosystem productivity (NEP), gross ecosystem productivity (GEP,) and ecosystem respiration (ER) responses to multilevel of temperature increments (control, warming 1, warming 2, warming 3, warming 4) in three contrasting hydrological growing seasons in a typical semiarid alpine meadow. We found that carbon fluxes responded to precipitation variations more strongly in low-level warming treatments than in high-level ones. The distinct responses were attributable to different soil water conditions and community composition under low-level and high-level warming during the three growing seasons. In addition, carbon fluxes were much more sensitive to decreased than to increased precipitation in low-level warming treatments, but not in high-level ones. At a regional scale, this negative asymmetry was further corroborated. This study reveals that future precipitation changes, particularly decreased precipitation would induce significant change in carbon fluxes, and the effect magnitude is regulated by climate warming size.

Increasing sensitivity of alpine grasslands to climate variability along an elevational gradient on the Qinghai-Tibet Plateau.

Monitoring and mapping the sensitivity of grassland ecosystems to climate change is crucial for developing sustainable local grassland management strategies. The sensitivity of alpine grasslands to climate change is considered to be high on the Qinghai-Tibet Plateau (QTP), yet little is known about its spatial pattern, and particularly the variations between different elevations. Here, based on the Normalized Difference Vegetation Index (NDVI) and three climate variables (air temperature, precipitation, and solar radiation), we modified a vegetation sensitivity index-approach to capture the relative sensitivity of alpine grassland productivity to climate variability on the QTP during 2000-2016. The results show that alpine grasslands on the southern QTP are more sensitive to climate variability overall, and that the climate factors driving alpine grassland dynamics are spatially heterogeneous. Alpine grasslands on the southern QTP are more sensitive to temperature variability, those on the northeastern QTP display strong responses to precipitation variability, and those on the central QTP are primarily influenced by a combination of radiation and temperature variability. The sensitivity of alpine grasslands to climate variability increases significantly along an elevational gradient, especially to temperature variability. This study underscores that alpine grasslands at higher elevations on the QTP are more sensitive to climate variability than those at lower elevations at the regional scale.

Nonlinear response of ecosystem respiration to multiple levels of temperature increases.

Global warming exerts profound impacts on terrestrial carbon cycles and feedback to climates. Ecosystem respiration (ER) is one of the main components of biosphere CO2 fluxes. However, knowledge regarding how ER responds to warming is still lacking. In this study, a manipulative experiment with five simulated temperature increases (+0℃ [Control], +2.1℃ [warming 1, W1], +2.7℃ [warming 2, W2], +3.2℃ [warming 3, W3], +3.9℃ [warming 4, W4]) was conducted to investigate ER responses to warming in an alpine meadow on the Tibetan Plateau. The results showed that ER was suppressed by warming both in dry and wet years. The responses of ER to warming all followed a nonlinear pattern. The nonlinear processes can be divided into three stages, the quick-response stage (W1), stable stage (W1-W3), and transition stage (W4). Compared with the nonlinear model, the linear model maximally overestimated the response ratios of ER to warming 2.2% and 3.2% in 2015 and 2016, respectively, and maximally underestimated the ratio 7.0% and 2.7%. The annual differences in ER responding to warming were mainly attributed to the distinct seasonal distribution of precipitation. Specially, we found that the abrupt shift response of ER to warming under W4 treatment in 2015, which might be regulated by the excitatory effect of precipitation after long-term drought in the mid-growing season. This study highlights the importance of the nonlinearity of warming effects on ER, which should be taken into the global-C-cycling models for better predicting future carbon-climate feedbacks.

Novel CsxWO3/TiO2 Microspheres as Enhanced Visible Light Photocatalysts for Dye Pollutant Treatments.

CsxWO3/TiO2 composites with different contents of CsxWO3 were successfully synthesized in this study by a facile hydrothermal process. CsxWO3/TiO2 composites were characterized by XRD, Raman, UV-visible diffuse reflectance spectra (DRS), photoluminescence spectra (PL) and SEM. TiO2 nanoparticles were distributed uniformly on the surface of the CsxWO3 microsphere in the prepared CsxWO3/TiO2 composites, and they formed heterojunctions with CsxWO3. The effect of CsxWO3 on the photoactivities of composites was investigated via DRS and PL. All CsxWO3/TiO2 catalysts showed enhanced photocatalytic activity for degrading rhodamine B under visible light irradiation. The 50% CsxWO3/TiO2 sample showed the best photocatalytic activity and its kinetic constant was 20 times larger than that of TiO2. The possible photocatalytic mechanism is also discussed from the trapping experiments of active species. The improved photocatalytic activity for the CsxWO3/TiO2 catalyst may be attributed to the synergetic effect between CsxWO3 microspheres and TiO2 nanoparticles. This novel photocatalyst can be used to degrade environmental pollutants in the future.