Research Center for Ecology and Environment of Central Asia

Plant traits-the morphological, anatomical, physiological, biochemical and phenological characteristics of plants-determine how plants respond to environmental factors, affect other trophic levels, and influence ecosystem properties and their benefits and detriments to people. Plant trait data thus represent the basis for a vast area of research spanning from evolutionary biology, community and functional ecology, to biodiversity conservation, ecosystem and landscape management, restoration, biogeography and earth system modelling. Since its foundation in 2007, the TRY database of plant traits has grown continuously. It now provides unprecedented data coverage under an open access data policy and is the main plant trait database used by the research community worldwide. Increasingly, the TRY database also supports new frontiers of trait-based plant research, including the identification of data gaps and the subsequent mobilization or measurement of new data. To support this development, in this article we evaluate the extent of the trait data compiled in TRY and analyse emerging patterns of data coverage and representativeness. Best species coverage is achieved for categorical traits-almost complete coverage for 'plant growth form'. However, most traits relevant for ecology and vegetation modelling are characterized by continuous intraspecific variation and trait-environmental relationships. These traits have to be measured on individual plants in their respective environment. Despite unprecedented data coverage, we observe a humbling lack of completeness and representativeness of these continuous traits in many aspects. We, therefore, conclude that reducing data gaps and biases in the TRY database remains a key challenge and requires a coordinated approach to data mobilization and trait measurements. This can only be achieved in collaboration with other initiatives.

Phthalic acid esters (PAEs) are a class of lipophilic chemicals widely used as plasticizers and additives to improve various products’ mechanical extensibility and flexibility. At present, synthesized PAEs, which are considered to cause potential hazards to ecosystem functioning and public health, have been easily detected in the atmosphere, water, soil, and sediments; PAEs are also frequently discovered in plant and microorganism sources, suggesting the possibility that they might be biosynthesized in nature. In this review, we summarize that PAEs have not only been identified in the organic solvent extracts, root exudates, and essential oils of a large number of different plant species, but also isolated and purified from various algae, bacteria, and fungi. Dominant PAEs identified from natural sources generally include di-n-butyl phthalate, diethyl phthalate, dimethyl phthalate, di(2-ethylhexyl) phthalate, diisobutyl phthalate, diisooctyl phthalate, etc. Further studies reveal that PAEs can be biosynthesized by at least several algae. PAEs are reported to possess allelopathic, antimicrobial, insecticidal, and other biological activities, which might enhance the competitiveness of plants, algae, and microorganisms to better accommodate biotic and abiotic stress. These findings suggest that PAEs should not be treated solely as a “human-made pollutant” simply because they have been extensively synthesized and utilized; on the other hand, synthesized PAEs entering the ecosystem might disrupt the metabolic process of certain plant, algal, and microbial communities. Therefore, further studies are required to elucidate the relevant mechanisms and ecological consequences.

Central Asia has a land area of 5.6 × 10(6) km(2) and contains 80-90% of the world's temperate deserts. Yet it is one of the least characterized areas in the estimation of the global carbon (C) stock/balance. This study assessed the sizes and spatiotemporal patterns of C pools in Central Asia using both inventory (based on 353 biomass and 284 soil samples) and process-based modeling approaches. The results showed that the C stock in Central Asia was 31.34-34.16 Pg in the top 1-m soil with another 10.42-11.43 Pg stored in deep soil (1-3 m) of the temperate deserts. They amounted to 18-24% of the global C stock in deserts and dry shrublands. The C stock was comparable to that of the neighboring regions in Eurasia or major drylands around the world (e.g. Australia). However, 90% of Central Asia C pool was stored in soil, and the fraction was much higher than in other regions. Compared to hot deserts of the world, the temperate deserts in Central Asia had relatively high soil organic carbon density. The C stock in Central Asia is under threat from dramatic climate change. During a decadal drought between 1998 and 2008, which was possibly related to protracted La Niña episodes, the dryland lost approximately 0.46 Pg C from 1979 to 2011. The largest C losses were found in northern Kazakhstan, where annual precipitation declined at a rate of 90 mm decade(-1) . The regional C dynamics were mainly determined by changes in the vegetation C pool, and the SOC pool was stable due to the balance between reduced plant-derived C influx and inhibited respiration.

Climate models use quantitative methods to simulate the interactions of the important drivers of climate system, to reveal the corresponding physical mechanisms, and to project the future climate dynamics among atmosphere, oceans, land surface and ice, such as regional climate models and global climate models. A comprehensive assessment of these climate models is important to identify their different overall performances, such as the accuracy of the simulated temperature and precipitation against the observed field. However, until now, the comprehensive performances of these models have not been quantified by a comprehensive index except the existed single statistical index, such as correlation coefficient ( r ), absolute error (AE), and the root‐mean‐square error (RMSE). To address this issue, therefore, in this study, a new comprehensive index Distance between Indices of Simulation and Observation (DISO) is developed to describe the overall performances of different models against the observed field quantitatively. This new index DISO is a merge of different statistical metrics including r , AE, and RMSE according to the distance between the simulated model and observed field in a three‐dimension space coordinate system. From the relationship between AE, RMSE, and RMS difference (RMSD) (i.e., standard deviation [ SD ] of bias time series), the new index also has the information of RMSD which is the statistical index in Taylor diagram. An example is applied objectively to display the applications of DISO and Taylor diagram in identifying the overall performances of different simulated models. Overall, with the strong physical characteristic of the distance in three dimensional space and the strict mathematical proof, the new comprehensive index DISO can convey the performances among different models. It can be applied in the comparison between different model data and in tracking changes in their performances.

Abstract Taylor diagram has been frequently used to evaluate climate or hydrology models or data. A Taylor diagram summarizes three frequently used metrics including correlation coefficient (CC), standard deviation (STD), and centred root mean square error ( RMSE c ). Although these three metrics are relevant metrics for some applications, in some cases, additional indicators are needed, which calls for a new method. This study firstly addressed the short comments about the distance between indices of simulation and observation (DISO) described in a previous study. Secondly, the number of statistical metrics of DISO is extended from 3 to more than 3. The statistical metrics of the expanded DISO are more flexible than the Taylor diagram which uses the three fixed metrics. Thirdly, the current work compares the Taylor diagram with the expanded DISO in terms of their theoretical bases, revealing the advantage of DISO in terms of its flexibility in the selection of different types of metrics, and its suitability as an effective single metric to express a model's or dataset's overall quality. The power and flexibility of the expanded DISO are discussed.

The countries of Central Asia are collectively known as the five ''-stans'': Uzbekistan, Kyrgyzstan, Turkmenistan, Tajikistan and Kazakhstan. In recent times, the Central Asian region has been affected by the shrinkage of the Aral Sea, widespread desertification, soil salinization, biodiversity loss, frequent sand storms, and many other ecological disasters. This paper is a review article based upon the collection, identification and collation of previous studies of environmental changes and regional developments in Central Asia in the past 30 years. Most recent studies have reached a consensus that the temperature rise in Central Asia is occurring faster than the global average. This warming trend will not only result in a higher evaporation in the basin oases, but also to a significant retreat of glaciers in the mountainous areas. Water is the key to sustainable development in the arid and semi-arid regions in Central Asia. The uneven distribution, over consumption, and pollution of water resources in Central Asia have caused severe water supply problems, which have been affecting regional harmony and development for the past 30 years. The widespread and significant land use changes in the 1990s could be used to improve our understanding of natural variability and human interaction in the region. There has been a positive trend of trans-border cooperation among the Central Asian countries in recent years. International attention has grown and research projects have been initiated to provide water and ecosystem protection in Central Asia. However, the agreements that have been reached might not be able to deliver practical action in time to prevent severe ecological disasters. Water management should be based on hydrographic borders and ministries should be able to make timely decisions without political intervention. Fully integrated management of water resources, land use and industrial development is essential in Central Asia. The ecological crisis should provide sufficient motivation to reach a consensus on unified water management throughout the region.

Global climate change and human activities are expected to have far-reaching implications for the associations between ecosystem services (ESs), especially in arid regions. Here, Central Asia (CA) was taken as a case study to describe the complex relationship among key ESs under the combined effects of future climate change and socioeconomic development. We propose a new framework that integrates the future land-use simulation (FLUS) model and integrated valuation of ESs and trade-offs (InVEST) model. A four-model ensemble mean from the Coupled Model Intercomparison Project 6 (CMIP6) was chosen to project future (2021–2100) variations in water yield (WY), soil conservation (SC), carbon storage (CS) and habitat quality (HQ). Spearman correlation was adopted to analyze the trade-offs and synergies between multiple ESs. Results showed that cropland degradation (−4.11% to −19.93%) and urban (+33.14% to +127.96%) and forestland (+5.31% to +25.52%) expansion will be the main forms of future land-use change in CA. Compared with the reference period (1995–2015), four ESs will exhibit different decreasing trends across CA under the three scenarios. We observed that the risk of soil erosion will increase in the mountainous areas of Kyrgyzstan and Tajikistan; cropland degradation and urban expansion would lead to a sharp reduction of CS and HQ in the Amu Darya Basin, Syr Darya Basin and southern Turkmenistan, especially in SSP245 scenario. We found that the weak pairwise correlations between HQ, SC and CS will be strengthened (R = 0.22–0.58; p < 0.05) in Kyrgyzstan and Tajikistan, whereas the significant positive correlation (R = 0.47–0.60; p < 0.01) between WY and SC will be weakened. An important information/recommendation provided by this study is that different management strategies should be designed in accordance with the major interactions among water, soil, carbon and biodiversity services.

BACKGROUND: Although spiritual care is a basic element of holistic nursing, nurses' spiritual care knowledge and abilities are often unable to satisfy patients' spiritual care needs. Therefore, nurses are in urgent need of relevant training to enhance their abilities to provide patients with spiritual care. DESIGN: A nonrandomized controlled trial. OBJECTIVE: To establish a spiritual care training protocol and verify its effectiveness. METHODS: This study recruited 92 nurses at a cancer treatment hospital in a single province via voluntary sign-up. The nurses were divided into two groups-the study group (45 people) and the control (wait-listed) group (47 people)-using a coin-toss method. The study group received one spiritual care group training session every six months based on their routine nursing education; this training chiefly consisted of lectures by experts, group interventions, clinical practice, and case sharing. The control group participated in monthly nursing education sessions organized by the hospital for 12 continuous months. RESULTS: After 12 months of intervention, the nurses in the study group had significantly higher overall spiritual health and spiritual care competency scores as well as significantly higher scores on all individual dimensions compared with those in the control group (P < 0.01). CONCLUSIONS: A spiritual care training protocol for nurses based on the concept of mutual growth with patients enhances nurses' spiritual well-being and spiritual care competencies.

To investigate the performance of extreme gradient boosting (XGBoost) in remote sensing image classification tasks, XGBoost was first introduced and comparatively investigated for the spectral-spatial classification of hyperspectral imagery using the extended maximally stable extreme-region-guided morphological profiles (EMSER_MPs) proposed in this study. To overcome the potential issues of XGBoost, meta-XGBoost was proposed as an ensemble XGBoost method with classification and regression tree (CART), dropout-introduced multiple additive regression tree (DART), elastic net regression and parallel coordinate descent-based linear regression (linear) and random forest (RaF) boosters. Moreover, to evaluate the performance of the introduced XGBoost approach with different boosters, meta-XGBoost and EMSER_MPs, well-known and widely accepted classifiers, including support vector machine (SVM), bagging, adaptive boosting (AdaBoost), multi class AdaBoost (MultiBoost), extremely randomized decision trees (ExtraTrees), RaF, classification via random forest regression (CVRFR) and ensemble of nested dichotomies with extremely randomized decision tree (END-ERDT) methods, were considered in terms of the classification accuracy and computational efficiency. The experimental results based on two benchmark hyperspectral data sets confirm the superior performance of EMSER_MPs and EMSER_MPs with mean pixel values within region (EMSER_MPsM) compared to that for morphological profiles (MPs), morphological profile with partial reconstruction (MPPR), extended MPs (EMPs), extended MPPR (EMPPR), maximally stable extreme-region-guided morphological profiles (MSER_MPs) and MSER_MPs with mean pixel values within region (MSER_MPsM) features. The proposed meta-XGBoost algorithm is capable of obtaining better results than XGBoost with the CART, DART, linear and RaF boosters, and it could be an alternative to the other considered classifiers in terms of the classification of hyperspectral images using advanced spectral-spatial features, especially from generalized classification accuracy and model training efficiency perspectives.

Acute farmland expansion and rapid urbanization in Central Asia have accelerated land use/land cover changes, which have substantial effects on ecosystem services. However, the spatiotemporal variations in ecosystem service values (ESVs) in Central Asia are not well understood. Here, based on land use products with 300-m resolution for the years 1995, 2005 and 2015 and transfer methodology, we predicted land use and land cover (LULC) for 2025 and 2035 using CA-Markov, assessed changes in ESVs in response to LULC dynamics, and explored the elasticity of the response of ESV to LULC changes. We found significant expansions of cropland (+22.10%) and urban areas (+322.40%) and shrinking of water bodies (−38.43%) and bare land (−9.42%) during 1995–2035. The combined value of ecosystem services of water bodies, cropland, and grassland accounted for over 90% of the total ESVs. Our study showed that cropland ecosystem services value increased by 93.45 billion US$ from 1995 to 2035, which was mainly caused by the expansion of cropland area. However, the area of water bodies decreased sharply during 1995–2035, causing a loss of 64.38 billion US$. Biodiversity, food production and water regulation were major ecosystem service functions, accounting for 80.52% of the total ESVs. Our results demonstrated that effective land-use policies should be made to control farmland expansion and protect water bodies, grassland and forestland for more sustainable ecosystem services.

The analysis of various characteristics and trends of precipitation is an essential task to improve the utilization of water resources. Lake Issyk-Kul basin is an upper alpine catchment, which is more susceptible to the effects of climate variability, and identifying rainfall variations has vital importance for water resource planning and management in the lake basin. The well-known approaches linear regression, Şen’s slope, Spearman’s rho, and Mann-Kendall trend tests are applied frequently to try to identify trend variations, especially in rainfall, in most literature around the world. Recently, a newly developed method of Şen-innovative trend analysis (ITA) provides some advantages of visual-graphical illustrations and the identification of trends, which is one of the main focuses in this article. This study obtained the monthly precipitation data (between 1951 and 2012) from three meteorological stations (Balykchy, Cholpon-Ata, and Kyzyl-Suu) surrounding the Lake Issyk-Kul, and investigated the trends of precipitation variability by applying the ITA method. For comparison purposes, the traditional Mann–Kendall trend test also used the same time series. The main results of this study include the following. (1) According to the Mann-Kendall trend test, the precipitation of all months at the Balykchy station showed a positive trend (except in January (Zc = −0.784) and July (Zc = 0.079)). At the Cholpon-Ata and Kyzyl-Suu stations, monthly precipitation (with the same month of multiple years averaged) indicated a decreasing trend in January, June, August, and November. At the monthly scale, significant increasing trends (Zc &gt; Z0.10 = 1.645) were detected in February and October for three stations. (2) The ITA method indicated that the rising trends were seen in 16 out of 36 months at the three stations, while six months showed decreasing patterns for “high” monthly precipitation. According to the “low” monthly precipitations, 14 months had an increasing trend, and four months showed a decreasing trend. Through the application of the ITA method (January, March, and August at Balykchy; December at Cholpon-Ata; and July and December at Kyzyl-Suu), there were some significant increasing trends, but the Mann-Kendall test found no significant trends. The significant trend occupies 19.4% in the Mann-Kendall test and 36.1% in the ITA method, which indicates that the ITA method displays more positive significant trends than Mann–Kendall Zc. (3) Compared with the classical Mann-Kendall trend results, the ITA method has some advantages. This approach allows more detailed interpretations about trend detection, which has benefits for identifying hidden variation trends of precipitation and the graphical illustration of the trend variability of extreme events, such as “high” and “low” values of monthly precipitation. In contrast, these cannot be discovered by applying traditional methods.

The exponential growth of human activities has resulted in a substantial increase in land use practices that not only modify the characteristics of landscape patterns but also pose significant landscape ecological risk (LER), with the latter being pivotal for ecosystem conservation and sustainable social development. However, research on LER and driving factors of Irtysh River Basin (IRB) are limited. Objectively assessing the LER of the high latitudes within Central Asia (Irtysh River Basin) and quantitatively identifying the environmental factors driving its changes holds significant research value for ensuring the ecological security of human habitation amidst global change. In this study, the spatial autocorrelation analysis method and geographically weighted regression (GWR) and geographical detector (Geo-Detector) models were utilized to reveal the spatiotemporal changes in LER based on land use/land cover (LULC) changes in the IRB from 1992 to 2020. The findings indicate that (1) the temporal scale reveals a slight increasing trend in LER within the IRB. (2) The spatial distribution is characterized by a dominance of lower- and medium-risk regions, with evident positive spatial autocorrelation. (3) The spatial pattern of LER is influenced by various factors, with a significant impact from temperature in the geo-detector model. In addition, the spatial heterogeneity of the effects of major factors was further obtained using the GWR model. The findings presented herein can serve as scientific references for the development of sustainability and ecological safety management in global arid zones and high-latitude cold regions, thus promoting environmental protection in various countries, enhancing consensus on ecological protection and facilitating international cooperation on conservation.

Abstract Precipitation's temporal and spatial patterns under climate change significantly impact global terrestrial ecology and human social activities. Climate models are essential tools for assessing the impacts of climate change and formulating policies to address climate change. The evaluation results of historical climate model simulations can represent the reliability of their future simulations. This study evaluated the simulation capabilities of 41 historical All‐Forcing monthly precipitation simulations and three integrated models over global land in the Coupled Model Comparison Project Phase 6 (CMIP6). The results show that the simulation capability of global climate models (GCMs) in CMIP6 is highly variable overland around the world. This variability is manifested in two aspects: the spatial variability of the comprehensive simulation ability of each model in different geographical regions and climatic zones of the world and the significant difference in the simulation ability of different models in each region. These GCMs generally overestimate global monthly precipitation over land, with the exception of southeast Asia and tropical rainforest climate (Af), where all models underestimate monthly precipitation. Some GCMS can perform well regionally but poorly on the global scale. One example shows that EC‐Earth3's best capability at Cwc climatic zone, surpassing the integrated model, but failed to rank in the top 10 in 22 of the 29 climate zones. Our results highlight the need to select appropriate models for integration when conducting climate change studies at global and regional scales as a critical factor in studying climate change predictions.

A novel multiscale morphological compressed change vector analysis (M <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> C <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> VA) method is proposed to address the multiple-change detection problem (i.e., identifying different classes of changes) in bitemporal remote sensing images. The proposed approach contributes to extend the state-of-the-art spectrum-based compressed change vector analysis (C <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> VA) method by jointly analyzing the spectral-spatial change information. In greater details, reconstructed spectral change vector features are built according to a morphological analysis. Thus more geometrical details of change classes are preserved while exploiting the interaction of a pixel with its adjacent regions. Two multiscale ensemble strategies, i.e., data level and decision level fusion, are designed to integrate the change information represented at different scales of features or to combine the change detection results obtained by the detector at different scales, respectively. A detailed scale sensitivity analysis is carried out to investigate its impacts on the performance of the proposed method. The proposed method is designed in an unsupervised fashion without requiring any ground reference data. The proposed M <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> C <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> VA is tested on one simulated and three real bitemporal remote sensing images showing its properties in terms of different image size and spatial resolution. Experimental results confirm its effectiveness.

Abstract With the rapid development of big data, assessment of data quality or model performance has become a hot scientific question. However, most existing lots of metrics focus on specific aspects of the assessment, and comprehensive assessment is rare. Therefore, it is very necessary to develop new assessment system. To address this problem, a new assessment system is constructed which is named after Chen, Chen, Hu, and Zhou (CCHZ)‐distance between indices of simulation and observation (DISO) according to the contributions of Xi Chen, Deliang Chen, Zengyun Hu, and Qiming Zhou. CCHZ‐DISO system builds on the Euclidean Distance and flexible determination of statistical metrics and their numbers. Due to its simplicity and flexibility, CCHZ‐DISO can be readily and widely applied to any subject of science. Therefore, it follows the principle of the Chinese philosopher Lao Zi's Da Dao Zhi Jian which means that the most basic truth is very simple.

Under climate change and increasing water demands, groundwater depletion has become regional and global threats for water security, which is an indispensable target to achieving sustainable developments of human society and ecosystems, especially in arid and semiarid regions where groundwater is a major water source. In this study, groundwater depletion of 2003–2016 over Xinjiang in China, a typical arid region of Central Asia, is assessed using the gravity recovery and climate experiment (GRACE) satellite and the global land data assimilation system (GLDAS) datasets. In the transition of a warm-dry to a warm-wet climate in Xinjiang, increases in precipitation, soil moisture and snow water equivalent are detected, while GRACE-based groundwater storage anomalies (GWSA) exhibit significant decreasing trends with rates between-3.61 ± 0.85 mm/a of CSR-GWSA and −3.10 ± 0.91 mm/a of JPL-GWSA. Groundwater depletion is more severe in autumn and winter. The decreases in GRACE-based GWSA are in a good agreement with the groundwater statistics collected from local authorities. However, at the same time, groundwater abstraction in Xinjiang doubled, and the water supplies get more dependent on groundwater. The magnitude of groundwater depletion is about that of annual groundwater abstraction, suggesting that scientific exploitation of groundwater is the key to ensure the sustainability of freshwater withdrawals and supplies. Furthermore, GWSA changes can be well estimated by the partial least square regression (PLSR) method based on inputs of climate data. Therefore, GRACE observations provide a feasible approach for local policy makers to monitor and forecast groundwater changes to control groundwater depletion.

Abstract. The extensive loess deposits of the Eurasian mid-latitudes provide important terrestrial archives of Quaternary climatic change. As yet, however, loess records in Central Asia are poorly understood. Here we investigate the grain size and magnetic characteristics of loess from the Nilka (NLK) section in the Ili Basin of eastern Central Asia. Weak pedogenesis suggested by frequency-dependent magnetic susceptibility (χfd%) and magnetic susceptibility (MS) peaks in primary loess suggest that MS is more strongly influenced by allogenetic magnetic minerals than pedogenesis, and may therefore be used to indicate wind strength. This is supported by the close correlation between variations in MS and proportions of the sand-sized fraction. To further explore the temporal variability in dust transport patterns, we identified three grain size end-members (EM1, mode size 47.5 µm; EM2, 33.6 µm; EM3, 18.9 µm) which represent distinct aerodynamic environments. EM1 and EM2 are inferred to represent grain size fractions transported from proximal sources in short-term, near-surface suspension during dust outbreaks. EM3 appears to represent a continuous background dust fraction under non-dust storm conditions. Of the three end-members, EM1 is most likely the most sensitive recorder of wind strength. We compare our EM1 proportions with mean grain size from the Jingyuan section in the Chinese loess plateau, and assess these in the context of modern and Holocene climate data. Our research suggests that the Siberian High pressure system is the dominant influence on wind dynamics, resulting in loess deposition in the eastern Ili Basin. Six millennial-scale cooling (Heinrich) events can be identified in the NLK loess records. Our grain size data support the hypothesis that the Siberian High acts as teleconnection between the climatic systems of the North Atlantic and East Asia in the high northern latitudes, but not for the mid-latitude westerlies.

Vegetation phenology is a sensitive indicator that reflects the vegetation–atmosphere interactions and vegetation processes under global atmospheric changes. Fast-developing remote sensing technologies that monitor the land surface at high spatial and temporal resolutions have been widely used in vegetation phenology retrieval and analysis at a large scale. While researchers have developed many phenology retrieving methods based on remote sensing data, the relationships and differences among the phenology retrieving methods are unclear, and there is a lack of evaluation and comparison with the field phenology recoding data. In this study, we evaluated and compared eight phenology retrieving methods using Moderate Resolution Imaging Spectroradiometer (MODIS) and the USA National Phenology Network data from across North America. The studied phenology retrieving methods included six commonly used rule-based methods (i.e., amplitude threshold, the first-order derivative, the second-order derivative, the third-order derivative, the relative change curvature, and the curvature change rate) and two newly developed machine learning methods (i.e., neural network and random forest). At the large scale, the start of the season (SOS) values, derived by all methods, had similar spatial distributions; however, the retrieved values had large uncertainties in each pixel, and the end of the season (EOS) inverted values were largely different among methods. At the site scale, the SOS and EOS values extracted by the rule-based methods all had significant positive correlations with the field phenology observations. Among the rule-based methods, the amplitude threshold method performed the best. The machine learning methods outperformed the rule-based methods in terms of retrieving the SOS when assessed using the field observations. Our study highlighted that there were large differences among the methods in retrieving the vegetation phenology from satellite data and that researchers must be cautious in selecting an appropriate method for analyzing the satellite-retrieved phenology. Our results also demonstrated the importance of field phenology observations and the usefulness of the machine learning methods in understanding the satellite-based land surface phenology. These findings provide a valuable reference for the future development of global and regional phenology products.

In this study, an epidemic model was developed to simulate and predict the disease variations of Guangdong province which was focused on the period from Jan 27 to Feb 20, 2020. To explore the impacts of the input population and quarantine strategies on the disease variations at different scenarios, four time points were assumed as Feb 6, Feb 16, Feb 24 and Mar 5 2020. The major results suggest that our model can well capture the disease variations with high accuracy. The simulated peak value of the confirmed cases is 1002 at Feb 10, 2020 which is mostly close to the reported number of 1007 at Feb 9, 2020. The disease will become extinction with peak value of 1397 at May 11, 2020. Moreover, the increased numbers of the input population can mainly shorten the disease extinction days and the increased percentages of the exposed individuals of the input population increase the number of cumulative confirmed cases at a small percentage. Increasing the input population and decreasing the quarantine strategy together around the time point of the peak value of the confirmed cases, may lead to the second outbreak.

Building change detection from very high-spatial-resolution (VHR) remote sensing images has gained increasing popularity in a variety of applications, such as urban planning and damage assessment. Detecting fine-grained “from–to” changes (change transition from one land cover type to another) of buildings from the VHR images is still challenging as multitemporal representation is complicated. Recently, fully convolutional neural networks (FCNs) have been proven to be capable of feature extraction and semantic segmentation of VHR images, but its ability in change detection is untested and unknown. In this letter, we leverage the semantic segmentation of buildings as an auxiliary source of information for the fine-grained “from–to” change detection. A deep multitask learning framework for change detection (MTL-CD) is proposed for detecting building changes from the VHR images. MTL-CD adopts the encoder–decoder architecture and solves the main task of change detection and the auxiliary tasks of semantic segmentation simultaneously. Accordingly, the change detection loss function is constrained by the auxiliary semantic segmentation tasks and enables the back-propagation of the building footprints’ detection errors for the improvement of change detection. A building change detection data set named the Guangzhou data set is also developed for model evaluation, in which the bitemporal R–G–B images were collected by airplane (2009) and unmanned aerial vehicle (UAV, 2019) with different flight heights. Experiments on the Guangzhou data set demonstrate that the MTL-CD method effectively detects fine-grained “from–to” changes and outperforms the postclassification methods and the direct change detection methods.