The urban heat island (UHI) refers to the phenomenon of higher atmospheric 21 and surface temperatures occurring in urban areas than in the surrounding rural areas. Mitigation 22 of the UHI effects via the configuration of green spaces and sustainable design of urban 23 *Manuscript Click here to download Manuscript: ISPRS_Manuscript_final_R1_2013_12_20_final.docx Click here to view linked References 2 environments has become an issue of increasing concern under changing climate. In this paper, 24 the effects of the composition and configuration of green space on land surface temperatures 25 (LST) were explored using landscape metrics including percentage of landscape (PLAND), edge 26 density (ED) and patch density (PD). An oasis city of Aksu in Northwestern China was used as a 27 case study. The metrics were calculated by moving window method based on a green space map 28 derived from Landsat Thematic Mapper (TM) imagery, and LST data were retrieved from 29 Landsat TM thermal band. Normalized mutual information measure was employed to investigate 30 the relationship between LST and the spatial pattern of green space. The results showed that 31 while the PLAND is the most important variable that elicits LST dynamics, spatial configuration 32 of green space also has significant effect on LST. Though, the highest normalized mutual 33 information measure was with the PLAND (0.71), it was found that ED and PD combination is 34 the most deterministic factors of LST than the unique effects of a single variable or the joint 35 effects of PLAND and PD or PLAND and ED. Normalized mutual information measure 36 estimations between LST and PLAND and ED, PLAND and PD and ED and PD were 0.7679, 37 0.7650 and 0.7832, respectively. A combination of the three factors PLAND, PD and ED 38 explained much of the variance of LST with a normalized mutual information measure of 39 0.8694. Results from this study can expand our understanding of the relationship between LST 40 and street trees and vegetation, and provide insights for sustainable urban planning and 41 management under changing climate. 42 43 44 Keywords-urban heat island, urban green space, landscape metrics, configuration, normalized 45 mutual information measure. 46 Remarkable proliferations of studies focusing on the relationship between LST and green space 79 composition has been reported over the last two decades (

No abstract

No abstract

Early detection of water stress is critical for precision farming for improving crop productivity and fruit quality. To investigate varying rootstock and irrigation interactions in an open agricultural ecosystem, different irrigation treatments were implemented in a vineyard experimental site either: (i) nonirrigated (NIR); (ii) with full replacement of evapotranspiration (FIR); or (iii) intermediate irrigation (INT, 50% replacement of evapotranspiration). In the summers 2014 and 2015, we collected leaf reflectance factor spectra of the vineyard using field spectroscopy along with grapevine physiological parameters. To comprehensively analyze the field-collected hyperspectral data, various band combinations were used to calculate the normalized difference spectral index (NDSI) along with 26 various indices from the literature. Then, the relationship between the indices and plant physiological parameters were examined and the strongest relationships were determined. We found that newly-identified NDSIs always performed better than the indices from the literature, and stomatal conductance (G s ) was the plant physiological parameter that showed the highest correlation with NDSI(R 603 ,R 558 ) calculated using leaf reflectance factor spectra (R 2 = 0.720). Additionally, the best NDSI(R 685 ,R 415 ) for non-photochemical quenching (NPQ) was determined (R 2 = 0.681). G s resulted in being a proxy of water stress. Therefore, the partial least squares regression (PLSR) method was utilized to develop a predictive model for G s . Our results showed that the PLSR model was inferior to the NDSI in G s estimation (R 2 = 0.680). The variable importance in the projection (VIP) was then employed to investigate the most important wavelengths that were most effective in determining G s . The VIP analysis confirmed that the yellow band improves the prediction ability of hyperspectral reflectance factor data in G s estimation. The findings of this study demonstrate the potential of hyperspectral spectroscopy data in motoring plant stress response.

No abstract

Early detection of grapevine viral diseases is critical for early interventions in order to prevent the disease from spreading to the entire vineyard. Hyperspectral remote sensing can potentially detect and quantify viral diseases in a nondestructive manner. This study utilized hyperspectral imagery at the plant level to identify and classify grapevines inoculated with the newly discovered DNA virus grapevine vein-clearing virus (GVCV) at the early asymptomatic stages. An experiment was set up at a test site at South Farm Research Center, Columbia, MO, USA (38.92 N, −92.28 W), with two grapevine groups, namely healthy and GVCV-infected, while other conditions were controlled. Images of each vine were captured by a SPECIM IQ 400–1000 nm hyperspectral sensor (Oulu, Finland). Hyperspectral images were calibrated and preprocessed to retain only grapevine pixels. A statistical approach was employed to discriminate two reflectance spectra patterns between healthy and GVCV vines. Disease-centric vegetation indices (VIs) were established and explored in terms of their importance to the classification power. Pixel-wise (spectral features) classification was performed in parallel with image-wise (joint spatial–spectral features) classification within a framework involving deep learning architectures and traditional machine learning. The results showed that: (1) the discriminative wavelength regions included the 900–940 nm range in the near-infrared (NIR) region in vines 30 days after sowing (DAS) and the entire visual (VIS) region of 400–700 nm in vines 90 DAS; (2) the normalized pheophytization index (NPQI), fluorescence ratio index 1 (FRI1), plant senescence reflectance index (PSRI), anthocyanin index (AntGitelson), and water stress and canopy temperature (WSCT) measures were the most discriminative indices; (3) the support vector machine (SVM) was effective in VI-wise classification with smaller feature spaces, while the RF classifier performed better in pixel-wise and image-wise classification with larger feature spaces; and (4) the automated 3D convolutional neural network (3D-CNN) feature extractor provided promising results over the 2D convolutional neural network (2D-CNN) in learning features from hyperspectral data cubes with a limited number of samples.

No abstract

Leaf chlorophyll concentration (LCC) is an important indicator of plant health, vigor, physiological status, productivity, and nutrient deficiencies. Hyperspectral spectroscopy at leaf level has been widely used to estimate LCC accurately and non-destructively. This study utilized leaf-level hyperspectral data with derivative calculus and machine learning to estimate LCC of sorghum. We calculated fractional derivative (FD) orders starting from 0.2 to 2.0 with 0.2 order increments. Additionally, 43 common vegetation indices (VIs) were calculated from leaf spectral reflectance factor to make comparisons with reflectance-based data. Within the modeling pipeline, three feature selection methods were assessed: Pearson’s correlation coefficient (PCC), partial least squares based variable importance in the projection (VIP), and random forest-based mean decrease impurity (MDI). Finally, we used partial least squares regression (PLSR), random forest regression (RFR), support vector regression (SVR), and extreme learning regression (ELR) to estimate the LCC of sorghum. Results showed that: (1) increasing derivative order can show improved model performance until certain order for reflectance-based analysis; however, it is inconclusive to state that a particular order is optimal for estimating LCC of sorghum; (2) VI-based modeling outperformed derivative augmented reflectance factor-based modeling; (3) mean decrease impurity was found effective in selecting sensitive features from large feature space (reflectance-based analysis), whereas simple Pearson’s correlation coefficient worked better with smaller feature space (VI-based analysis); and (4) SVR outperformed all other models within reflectance-based analysis; alternatively, ELR with VIs from original reflectance yielded slightly better results compared to all other models.