Fraction of absorbed photosynthetically active radiation (FAPAR) is an essential indicator of vegetation productivity and carbon uptake. Though coarse resolution FAPAR products from MODIS and VIIRS sensors have supported global vegetation monitoring, their limitations in characterizing heterogeneity at ecosystem scales necessitate Landsat-derived FAPAR at the 30 m resolution. This study developed a practical approach integrating radiative transfer (RT) modeling and machine learning to estimate 30-m FAPAR from Landsat surface reflectance. A coupled land-atmosphere RT model and shuffled complex evolution (SCE) optimization algorithm was first implemented at globally distributed VIIRS pixels for realistic simulation of land surface reflectance corresponding Landsat spectral bands and FAPAR under various conditions. The simulated dataset encompassing diverse canopy structural and atmospheric states is utilized to train a random forest model relating Landsat surface reflectance bands to FAPAR. The trained RF model was applied to estimate long-term FAPAR from Landsat 5/7/8/9 surface reflectance data and demonstrated reliable performance in capturing FAPAR dynamics across vegetation types. Comprehensive validation of the Landsat FAPAR was conducted using global field measurements from three observation networks, including the VALERI (VAlidation of Land European Remote Sensing Instruments), ImagineS, and Copernicus GBOV (Ground Based Observations for Validation) projects, which involved shrubland, forest, grassland, and cropland, coarse resolution FAPAR products were also used for comparison to evaluate our estimates. Validation results demonstrate good agreement of the estimated Landsat FAPAR with ground measurements, achieving R2 of 0.876, RMSE of 0.108 and bias of 0.032. Consistent accuracy was attained across different Landsat sensors. The validation performance of the Landsat FAPAR over various land cover types was: shrubland (R2: 0.848; RMSE: 0.092), forest (R2: 0.876; RMSE: 0.101), grassland (R2: 0.739; RMSE: 0.118), and cropland (R2: 0.839; RMSE: 0.125). Comparison against GLASS/ MODIS products shows improved accuracy over existing FAPAR datasets, benefiting from the ability of Landsat to resolve within-pixel heterogeneity. This study synergized the strengths of radiative transfer model and machine learning algorithm while overcoming the limitations of RTM parameterization and generalization, providing an efficient and robust Landsat FAPAR estimation approach. The demonstrated approach creates opportunities to generate a long-term 30 m FAPAR record by leveraging the continuity of Landsat observations to advance vegetation productivity monitoring and carbon science.

Owing to advancements in satellite remote sensing technology, the acquisition of global land surface parameters, notably, the leaf area index (LAI), has become increasingly accessible. The Sentinel-2 (S2) satellite plays an important role in the monitoring of ecological environments and resource management. The prevalent use of the 20 m spatial resolution band in S2-based inversion models imposes significant limitations on the applicability of S2 data in applications requiring finer spatial resolution. Furthermore, although a substantial body of research on LAI retrieval using S2 data concentrates on agricultural landscapes, studies dedicated to forest ecosystems, although increasing, remain relatively less prevalent. This study aims to establish a viable methodology for retrieving 10 m resolution LAI data in forested regions. The empirical model of the soil adjusted vegetation index (SAVI), the backpack neural network based on simulated annealing (SA-BP) algorithm, and the variational heteroscedastic Gaussian process regression (VHGPR) model are established in this experiment based on the LAI data measured and the corresponding 10 m spatial resolution S2 satellite surface reflectance data in the Saihanba Forestry Center (SFC). The LAI retrieval performance of the three models is then validated using field data, and the error sources of the best performing VHGPR models (R2 of 0.8696 and RMSE of 0.5078) are further analyzed. Moreover, the VHGPR model stands out for its capacity to quantify the uncertainty in LAI estimation, presenting a notable advantage in assessing the significance of input data, eliminating redundant bands, and being well suited for uncertainty estimation. This feature is particularly valuable in generating accurate LAI products, especially in regions characterized by diverse forest compositions.

Abstract Vegetation growth is influenced by the microclimate driven by aspects, as evident in the asymmetric vegetation greenness on polar‐facing slopes (PFS) and equatorial‐facing slopes (EFS). However, it remains uncertain whether aspects influence vegetation phenology. To address this question, we defined the aspect‐induced phenological differences between PFS and EFS from 2019 to 2022 within each 3 × 3 km2 grid, using average phenological metrics extracted from Sentinel‐2 data. We found that the start of the growing season (SOS) occurs earlier on EFS in cold and humid regions, but in arid areas, PFS has an earlier SOS. The end of the growing season (EOS) consistently occurred later on EFS due to radiation limitations in autumn phenology. Employing the space‐for‐time approach, the observed distribution of phenological differences within the climate space could potentially indicate the phenological trends of different slope orientations in the future. Our study provides valuable insights into topographic regulation on vegetation phenology.

Deciduous forests spring phenology plays a major role in balancing the carbon cycle. The cloud cover affects images acquired from optical sensors and reduces their performance in monitoring phenology. Synthetic aperture radar (SAR) can regularly acquire images day and night independent of weather conditions, which offers more frequent observations of vegetation phenology compared to optical sensors. However, it remains unclear how SAR data-derived indices vary across different growth stages of forests. Here, we explored the relationship between the cross ratio (CR) index derived from Sentinel-1 data and the deciduous forest growth process. We proposed a deciduous forests spring phenology extraction method using CR and compared the extracted start of growing season (SOS) with those extracted using normalized difference vegetation index (NDVI) derived from Sentinel-2 optical satellite data and green chromatic coordinate (GCC) derived from ground PhenoCam data. We extracted the SOS of 41 PhenoCam sites over the Continental United States in 2018 using the dynamic threshold method. Our results showed that the variations of CR time series are closely related to the phenological processes of deciduous forests. The SOS extracted using CR data showed high consistency with those extracted using GCC (R2 = 0.46), with slightly lower accuracy compared with NDVI-derived results (R2 = 0.62). Our study illustrates the value and mechanism of deciduous forests spring phenology extraction using SAR data and provides a reference for using SAR data to improve forest phenology extraction in addition to using optical remote sensing data, especially in rainy and cloudy regions.

Vegetation, especially forest ecosystems, plays an important role in the global energy flow and material cycle. The vegetation index (VI) is an important index reflecting the dynamic change in vegetation and directly reflects the response of ecosystem to global climate change. The Greater Khingan Mountains Forest region is located in the northeast of China. It is the largest primeval forest region in China, which is well preserved and less affected by human activities. It is of great significance to study the driving mechanism of forest vegetation change for future ecological prediction and management. In this study, GIMMS NDVI data were used to explore the characteristics of nonlinear temporal and spatial variation of NDVI in the Greater Khingan Mountains and its relationship with climatic factors. Firstly, the EEMD method was used to analyze the characteristics of vegetation change in the study area from 1982 to 2015. Secondly, the relationship between vegetation change and climate was discussed by using precipitation and temperature data. The results showed that the following: (1) from 1982 to 2015, the interannual change in vegetation in the Greater Khingan Mountains presented a trend of slow fluctuation and gradual decrease (SLOPE = −0.1645/10,000, p < 0.01). (2) The spatial distribution of vegetation change had obvious geographical differences, and in the central region, the overall distribution characteristics had an obvious browning trend, and in the northwest and southeast, the distribution characteristics had a green trend. (3) The correlation analysis results of vegetation change and climate factors showed that NDVI change was significantly positively correlated with temperature and precipitation; additionally, NDVI change was more correlated with temperature with a range of 0.8–1 than precipitation. (4) The results of vegetation attribution analysis in four typical areas of the study area showed that the following: the coniferous forest area has good cold tolerance and drought tolerance, the correlation between vegetation change and climate factors (temperature, precipitation) was not the strongest, which was 0.537 and 0.828, respectively. The ecological transition area and the broad-leaved forest area, which was located at the edge of the study area, have relatively fragile ecosystems, showed a strong correlation with precipitation, and the correlation coefficients reached 0.670 and 0.632, respectively. The surface water resources provide favorable conditions for the growth of vegetation, it showed a weak correlation with precipitation, and the correlation coefficient was 0.5349.

Accurate estimation of photosynthetic phenology is of great importance for understanding the response of terrestrial biosphere to climate change. The near-infrared reflectance of vegetation (NIRv) has been increasingly used to estimate photosynthetic phenology. However, topography significantly affects illumination conditions and induces uncertainty in the retrieval of NIRv photosynthetic phenology over mountainous areas. We evalu-ated the illumination effects on three modalities of NIRv: (i) the original NIRv, (ii) the product of NIRv and solar incident radiation (NIRvP), and (iii) the topographically corrected NIRv (TCNIRv). We assessed the impact of sun geometry on the phenological metrics for the timing of the start and end of the season (i.e., SOS and EOS) derived from satellite NIRv time series as compared with those derived from in-situ gross primary production (GPP) measured at the Brasschaat (BE-Bra) and La & BULL;geren (CH-Lae) flux towers located, respectively, over a flat and a mountainous forest area in Europe. We observed a seasonally divergent performance of NIRv in photosynthetic phenology extraction: NIRv-derived SOS had a good consistency with GPP-derived estimates (109/114d for NIRv vs 108/115d for GPP at BE-Bra/CH-Lae), whilst it showed a positive time lag for the EOS (291/292d for NIRv vs 264/268d for GPP at BE-Bra/CH-Lae). The radiation constraint in NIRvP formulation corrected the bias in NIRv-based EOS estimates. Path length correction (PLC) also alleviates the illumination effects on the original NIRv, making TCNIRv comparable to NIRvP in estimating SOS and EOS. We conclude that illumination effects must be corrected from NIRv for the extraction of photosynthetic phenology, especially for the autumn phenology. Our study has significant implications for understanding the responses of phenology to climate change and the climate-carbon feedbacks, particularly over complex topography mountainous areas.

Leaf senescence plays a pivotal role in the regulation of carbon and nutrient cycles within terrestrial ecosystems. However, our understanding of leaf senescence velocity (LSV) remains comparatively limited when compared to the end of the growing season (EOS). In this study, we extracted the LSV in Tibet Plateau (TP) over the period 2001–2018 based on the satellite-derived normalized difference greenness index (NDGI), then we evaluated the influences of climate drivers on the spatio-temporal variations of LSV. Lastly, we explored the implications of LSV on vegetation growth by analyzing the correlation between LSV and the start of growing season (SOS) and the annual net primary production (NPP). Our findings revealed that the multi-year averaged LSV ranged from 30 to 70% mon-1 and displayed a discernible spatial gradient, declining from west to east, and the spatial pattern of LSV was mainly controlled by radiation. Trend tests and partial correlation analyses unveiled a temporal decrease in LSV within the central TP region, attributed to rising temperatures, while an increase was observed in the southwestern TP due to water deficits. We also found that LSV had a strong impact on current-year net NPP and following-year SOS. This suggests that LSV may play a significant role in regulating carbon exchange during the current year and influencing the onset of spring green-up in the subsequent year. By emphasizing the continuous nature of leaf senescence, our study provides fresh insights into the intricate interactions between vegetation growth and climate change, contributing to the existing body of knowledge in this field.

Accurate monitoring autumn photosynthetic phenology is essential for understanding carbon cycles. The broadband green-red vegetation index (GRVI) derived from broadband red and green reflectance has been increasingly used in this field. However, the performance of GRVI in large areas remains unclear. We evaluated the performance of the normalized vegetation index (NDVI), normalized difference greenness index (NDGI), the near-infrared reflectance of vegetation (NIRv), solar-induced chlorophyll fluorescence (SIF) and GRVI in tracking autumn photosynthetic phenology of alpine grassland at flux sites and the entire Tibetan Plateau (TP). The results revealed that GRVI (R2 = 0.42, RMSE = 7.68 d and Bias = 4.39 d) performed comparable with SIF (R2 = 0.45, RMSE = 5.79 d and Bias = 0.71 d) in extracting the end of the photosynthetically active season (EOS) with eddy covariance flux measurements as reference. On contrary, a systematically later EOS was estimated by NDVI (Bias = 20.35 d), NDGI (Bias = 13.62 d) and NIRv (Bias = 6.56 d). The application example on the TP collaborated these findings. The divergent performances between indicators were rooted in the photosynthesis downregulation in autumn which is jointly controlled by canopy structure and physiology. Due to the fact that NDVI, NDGI and NIRv primarily depict vegetation structure, while the use of SIF, which represents vegetation photosynthetic physiology, is influenced by the limited spatial resolution and temporal coverage. Our study highlights the unique advantage of GRVI over other existing satellite indicators for estimating autumn photosynthetic phenology with high resolution and long-time span. We suggest revisiting the dynamics of autumn photosynthetic phenology using GRVI, which has significant implications on carbon uptake studies.

Quantifying the contributions of air temperature and precipitation changes to drought events can inform decision-makers to mitigate the impact of droughts while existing studies focused mainly on long-term dryness trends. Based on the latest Coupled Model Intercomparison Project (CMIP6), we analyzed the changes in drought events and separated the contributions of air temperature and precipitation to the risk of future drought events. We found that drought frequency, duration, severity, and month will increase in the future (56.4%, 63.5%, 82.9%, and 58.2% of the global land area in SSP245, and 58.1%, 67.7%, 85.8%, and 60.5% of the global land area in SSP585, respectively). The intermediate scenario has a similar pattern to the most extreme scenario, but low emission was found to mitigate drought risk. Globally, we found that air temperature will have a greater impact than precipitation on intensifying drought. Increasing precipitation will mitigate drought risks in some middle and high northern latitudes, whilst the trend in increasing air temperature will counter the effects of precipitation and increase the impact of droughts. Our study improves the understanding of the dynamics of future devastating drought events and informs the decision-making of stakeholders.

Drought is an event of shortages in the water supply, whether atmospheric, surface water or ground water. Prolonged droughts have negative impacts on ecosystems, agriculture, society, and the economy. Although existing drought index products are widely utilized in drought monitoring, the coarse spatial resolution greatly limits their applications on regional or local scales. Machine learning driven by remote sensing observations offers an opportunity to monitor regional scale droughts. However, the limited time range of remote sensing observations such as vegetation index (VI) resulted in a substantial gap in generating high resolution drought index products before 2000. This study generated spatiotemporally continuous Standardized Precipitation Evapotranspiration Index (SPEI) data spanning from 1901–2018 in southwestern China by machine learning. It indicated that four Classification and Regression Tree (CART) approaches, decision trees (DT), random forest (RF), gradient boosted regression trees (GBRT) and extra trees (ET), can provide valid local drought information by downscaling the Estación Experimental de Aula Dei (EEAD) data. The in-situ SPEI dataset produced by the Penman–Monteith method was used as a benchmark to evaluate the temporal and spatial performance of the downscaled SPEI. In addition, the necessity of VI in SPEI downscaling was also assessed. The results showed that: (1) the ET-based product has the best performance (R2 = 0.889, MAE = 0.232, RMSE = 0.432); (2) the VI provides no significant improvement for SPEI re-construction; (3) topography exerts an obvious influence on the downscaling process, and (4) the downscaled SPEI shows more consistency with the in-situ SPEI compared with EEAD SPEI. The proposed method can be easily extended to other areas without in-situ data and enhance the ability of long-term drought monitoring.

Accurate estimation of photosynthetic phenology is of great importance for understanding the response of terrestrial biosphere to climate change. The near-infrared reflectance of vegetation (NIRv) has been increasingly used to estimate photosynthetic phenology. However, the influence of topography on NIRv induces uncertainty in photosynthetic phenology retrieval over mountainous areas. We evaluated the topographic effects on NIRv and employed topographically corrected NIRv (TCNIRv) to improve its phenology retrieval accuracy over a mountainous forest. We extracted the start/end of the photosynthetically active season (SOS/EOS) from the NIRv and TCNIRv time series, and compared them with those generated from in-situ gross primary production (GPP) measured at the Lägeren flux tower. We observed a seasonally divergent performance of NIRv in photosynthetic phenology extraction: NIRv-derived SOS had a good consistency with GPP-derived estimates (113d vs 114d from NIRv and GPP, respectively), whilst EOS obviously lagged GPP-derived ones (292d vs 268d from NIRv and GPP, respectively). The seasonal divergence of NIRv-derived photosynthetic phenology might be related to the contrasting velocities of the magnitude of topographic effects on NIRv around SOS and EOS. In contrast, TCNIRv reduced topographic effects in the original NIRv and it was comparable to GPP in estimating SOS/EOS. As such, we suggest that topographic effects should be eliminated when using NIRv for extraction of photosynthetic phenology, especially for the autumn phenology. Our study has significant implications for understanding the responses of phenology to climate change and the climate-carbon feedbacks over complex topography mountainous areas.

Spatiotemporally continuous monitoring of aboveground biomass (AGB), an important indicator of grassland productivity, is crucial for achieving sustainable grassland development. Most existing grassland AGB estimation methods are empirical, and their temporally and spatially specific nature hinders operational application at large scales. Grass is herbaceous, so its AGB can be represented as the product of leaf area index (LAI) and dry matter content (C-m), both are the inputs of PROSAIL model. We, therefore, proposed a novel physical-based method through PROSAIL model inversion. Results showed that the estimated AGB presented good consistency with fieldmeasured one, with R-2 = 0.87 and RMSE = 14.29 g/m(2). We then implemented our method on the Google Earth Engine platform and generated daily and monthly AGB products covering the Tibetan Plateau (TP) and spanning from 2000 to 2021. These products characterized the spatiotemporally continuous dynamics of AGB on the TP. For example, it captured the decrease in dry matter caused by grazing during grassland dormancy, which is impossible for other existing AGB retrieval methods. Our method provides a promising tool to generate spatiotemporally continuous grassland AGB, which would inform the decision making for the conservation and restoration of grassland.

叶面积指数LAI（Leaf Area Indexï¼‰æ˜¯è°ƒèŠ‚æ¤è¢«å† å±‚ç”Ÿç†è¿‡ç¨‹çš„æœ€é‡è¦çš„ç”Ÿç‰©ç‰©ç†å˜é‡ä¹‹ä¸€ï¼Œé«˜ç©ºé—´åˆ†è¾¨çŽ‡æ—¶é—´åºåˆ—LAIå¯¹äºŽæ¤è¢«ç”Ÿé•¿æ£€æµ‹ã€åœ°è¡¨è¿‡ç¨‹æ¨¡æ‹Ÿä¸ŽåŒºåŸŸå’Œå ¨çƒå˜åŒ–ç ”ç©¶è‡³å ³é‡è¦ï¼Œä½†æ˜¯ç”±äºŽæ•°æ®ç¼ºå¤±å’Œåæ¼”æ–¹æ³•é™åˆ¶ï¼Œç›®å‰è¿˜æ²¡æœ‰æ—¶ç©ºè¿žç»­çš„é«˜åˆ†è¾¨çŽ‡LAIæ•°æ®äº§å“ã€‚æœ¬ç ”ç©¶æå‡ºäº†ä¸€ç§ç”Ÿæˆæ—¶é—´è¿žç»­çš„é«˜ç©ºé—´åˆ†è¾¨çŽ‡LAIæ•°æ®çš„ç®—æ³•ï¼Œé¦–å ˆå¯¹MODIS LAI产品滤波平滑，生成时间序列LAIçš„ä¸ŠåŒ ç»œæ›²çº¿ï¼Œæ ¹æ®ä¸ŠåŒ ç»œæ›²çº¿æä¾›çš„å˜åŒ–ä¿¡æ¯æž„å»ºLAI动态模型。然后利用地面实测的LAI数据与Landsat反射率数据构建LAI反演的BP （Back Propagation）神经网络模型。将反演得到的高分辨率LAI数据作为LAI观测数据，利用集合卡尔曼滤波EnKF（Ensemble Kalman Filter）方法实时更新动态模型，生成时间连续的30 m空间分辨率LAI数据集。基于该算法生成了塞罕坝地区2000年—2018年长时间序列LAI数据集，利用Prophetæ·±åº¦å­¦ä¹ æ¨¡åž‹è¿›è¡Œæ¨¡æ‹Ÿå’Œé¢„æµ‹ï¼Œæ ¹æ®é¢„æµ‹å’ŒåŽŸå§‹LAI差异，利用支持向量机SVM（Support Vector Machine）方法检测植被干扰状况。结果表明：EnKF算法能够生成时空连续的高空间分辨率LAI数据，估算结果与地面测量值一致性较高，R2为0.9498，RMSE为0.1577，在区域尺度上与Landsat LAI参考值较为吻合，R2高于0.87，RMSE低于0.61。Prophet与SVMæ¨¡åž‹æ£€æµ‹åˆ°ç ”ç©¶åŒº2009年，2010年，2013年，2014年， 2015å¹´æ¤è¢«å—å¹²æ‰°è¾ƒä¸ºä¸¥é‡ï¼Œä¸»è¦ç”±äºŽå¹´é™æ°´é‡åå°‘å’Œæž—åŒºä½œä¸šç ä¼é€ æˆï¼Œæ£€æµ‹ç»“æžœä¸Žå½“åœ°é™æ°´é‡ä¸Žç ä¼æ•°æ®å»åˆã€‚æœ¬æ–‡æå‡ºçš„ç®—æ³•å¯ç”¨äºŽå¤§èŒƒå›´é«˜æ—¶ç©ºLAIæ•°æ®åæ¼”å’Œæ¤è¢«å˜åŒ–æ£€æµ‹ï¼Œå¯¹å¡žç½•åä¹ƒè‡³å ¨å›½æž—åŒºè§„åˆ’ç®¡ç†å ·æœ‰é‡è¦çš„å‚è€ƒä»·å€¼ã€‚

The leaf area index (LAI) is an important parameter for vegetation monitoring and land surface ecosystem research. Although a variety of LAI products have been generated, the moderate to coarse spatial resolution and low temporal resolution of these products are insufficient for regional-scale analysis. In this study, a modified ensemble Kalman filter model (MEnKF) was proposed to generate spatio-temporal complete 30 m LAI data. High-quality, filtered historical Moderate-resolution Imaging Spectroradiometer (MODIS) LAI data were used to obtain the LAI background, and an LAI temporal dynamic model was constructed based on it. An improved back-propagation (BP) neural network based on a simulated annealing algorithm (SA-BP) was constructed with paired Landsat surface reflectance data and field LAI data to generate a 30 m LAI. The MEnKF was used to estimate the spatio-temporal complete LAI beginning from the LAI peak value position where Landsat observations were available. The spatio-temporal 30 m LAI was estimated in farmland (Pshenichne), grassland (Zhangbei), and woodland (Genhe) sites. The results indicate that the MEnKF-estimated LAI is consistent with the field measurements for all sites (the coefficient of determination ( R 2 ) = 0.70; root mean squared error (RMSE) = 0.40) and is better than that of the conventional sequence data assimilation algorithm ( R 2 = 0.40; RMSE = 0.78). The regional LAI captures the vegetation growth pattern and is consistent with the Landsat LAI, with an R 2 larger than 0.65 and an RMSE less than 0.51. The proposed MEnKF algorithm, which effectively avoids error accumulation in the data assimilation scheme, is an efficient method for spatio-temporal complete 30 m LAI estimation.

卫星遥感技术的快速发展使得获取全球大范围叶面积指数成为可能,但基于现有的算法和数据估算高分辨率LAI的精度还需要提高.针对农作物、草地和林地等3种典型地表类型,选取地面观测数据较多的4个研究区,包括3个各地类用于建模验证的研究区与一个用于适用性验证的独立研究区,针对4个研究区,分别获取地面测量数据以及对应的30 m空间分辨率地表反射率数据.在3个主要研究区建立并比较了NDVI植被指数经验模型、BP神经网络模型和基于模拟退火算法的BP神经网络模型,利用地面实测数据对模型进行验证.结果表明:在研究所选的3个主要研究区,基于模拟退火算法的BP神经网络模型的估算精度比BP神经网络模型和NDVI经验模型的估算精度高,农田、草地和林地站点估算结果的决定系数分别为0.899、0.858和0.863,BP神经网络模型的估算结果决定系数分别为:0.763、0.710和0.742,NDVI经验模型的精度最差,其估算结果的决定系数分别为0.622、0.536和0.637.为了验证SA-BP神经网络的适用性,选取独立研究区进行验证,结果显示验证精度较高,R2为0.842,RMSE为0.6895,说明该模型外推能力较好.研究证明了基于模拟退火算法的BP神经网络模型提高了模型泛化能力,有效防止了BP神经网络模型滑入局部最小值,是提高高空间分辨率LAI估算精度的有效手段.

Land surface albedo analysis and prediction are of great significance for global energy budget research and global change forecasting. Research has been performed on time series albedo analysis but seldom attempt was performed on land surface albedo prediction. This article develops an effective method for land surface albedo prediction from Moderate-Resolution Imaging Spectroradiometer (MODIS) time series albedo data (MCD43A3). It consists of time series data decomposing and time series data forecasting. The ensemble empirical mode decomposition (EEMD) method decomposes the MODIS historical time series albedo data into several intrinsic mode functions (IMFs) and one residual series, then the nonlinear autoregressive neural network (NARnet) method is used to forecast each IMF component and residue. The predictions of all IMFs and residue are summed to obtain a final forecast for the albedo series. The proposed method was performed on monthly and daily albedo prediction both in snow-free and snowy areas. The results showed that the forecast albedo consists of the MODIS albedo data well, with R2 greater than 0.89 and RMSE less than 0.052 for snow-free areas. For snowy areas, the forecasting also performed well during snow cover periods, with R2 greater than 0.76 and RMSE less than 0.076. For irregular change periods of snow falling and melting, it is hard to get very high prediction accuracy due to the irregular land surface change. For this problem, more land surface information should be introduced, or adjusting the model over time is necessary.

Leaf area index (LAI) is one of the most important biophysical variables for regulating the physiological processes of vegetation canopy. Time series high-resolution LAI data is critical for vegetation growth monitoring, surface process simulation and global change research. However, there are no high-resolution LAI data products that are continuous in time and space. In this paper, we use MODIS LAI products and Landsat surface reflectance data to generate time series high-resolution LAI datasets from 2000 to 2018 in Saihanba based on the ensemble kalman filter, and uses time-series LAI data to monitor surface vegetation changes according to the Prophet model. Firstly, the multi-step Savitzky–Golay filtering algorithm is used to smooth the MODIS LAI data, and the upper envelope of time series LAI is generated. A dynamic model is constructed according to the trend of LAI upper envelope to provide the short-range forecast of LAI. Then the ground measured LAI data and the corresponding Landsat reflectance data are used to train a Back Propagation neural network. High-resolution LAI data from BP model is used to update the dynamic model in real time to generate high-resolution time series LAI data based on EnKF. Finally, the time series LAI data is used as the input of Prophet deep learning model to obtain the LAI time series prediction values of a certain year. Comparing the prediction results with the LAI of current year, the correlation coefficient and the root mean square error distribution maps can be obtained, a support vector machine method is used to classify the disturbed pixels and the normal pixels. The LAI time series estimation has a high accuracy of R²larger than 0.90, and RMSE less than 0.54. The disturbance monitoring results indicate that vegetation in 2009, 2010, 2013, 2014, 2015, 2017 is seriously disturbed, Variation of meteorological conditions and deforest contributes heavily to the disturbance.

Leaf area index (LAI) is an important parameter for describing vegetation change and growth trend. Due to various problems of satellite observation and retrieval algorithm, spatiotemporal complete LAI data is limited, which hindered the application of LAI in different areas. In this paper, a data fusion method, multiresolution tree (MRT) model, was used to develop LAI of different spatial resolution. Three LAI data sets of Landsat (30 m), MODIS (450 m) and GLASS (900 m) are used. MRT was performed in Ukraine area including farmland, forestland and grassland. Results indicate that MRT is an efficient algorithm to get spatial complete LAI estimation at different resolution. LAI of different spatial resolution consists well, R2 and RMSE are 0.8612 and 0.3986 when compare Landsat LAI with MODIS LAI, which are 0.7568 and 0.4736 when compare Landsat with Glass.

Continuous, long-term sequence, land surface albedo data have crucial significance for climate simulations and land surface process research. Sensors such as the Moderate-Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer (VIIRS) provide global albedo product data sets with a spatial resolution of 500 m over long time periods. There is demand for new high-resolution albedo data for regional applications. High-resolution observations are often unavailable due to cloud contamination, which makes it difficult to obtain time series albedo estimations. This paper proposes an “amalgamation albedo“ approach to generate daily land surface shortwave albedo with 30 m spatial resolution using Landsat data and the MODIS Bidirectional Reflectance Distribution Functions (BRDF)/Albedo product MCD43A3 (V006). Historical MODIS land surface albedo products were averaged to obtain an albedo estimation background, which was used to construct the albedo dynamic model . The Thematic Mapper (TM) albedo derived via direct estimation approach was then introduced to generate high spatial-temporal resolution albedo data based on the Ensemble Kalman Filter algorithm (EnKF). Estimation results were compared to field observations for cropland, deciduous broadleaf forest, evergreen needleleaf forest, grassland, and evergreen broadleaf forest domains. The results indicated that for all land cover types, the estimated albedos coincided with ground measurements at a root mean squared error (RMSE) of 0.0085–0.0152. The proposed algorithm was then applied to regional time series albedo estimation; the results indicated that it captured spatial and temporal variation patterns for each site. Taken together, our results suggest that the amalgamation albedo approach is a feasible solution to generate albedo data sets with high spatio-temporal resolution.