Seasonal dynamic pattern analysis on global FPAR derived from AVHRR GIMMS NDVI

Abstract

The purpose of this paper is to develop Advanced Very High Resolution Radiometer (AVHRR) Global Inventory Modelling and Mapping Studies (GIMMS) Normalised Difference Vegetation Index (NDVI; AVHRR GIMMS NDVI for short) based fraction of absorbed photosynthetically active radiation (FPAR) from 1982 to 2006 and focus on their seasonal and spatial patterns analysis. The available relationship between FPAR and NDVI was used to calculate FPAR values from 1982 to 2006 and validated by Moderate-resolution Imaging Spectroradiometer (MODIS) FPAR product. Then, the seasonal dynamic patterns were analysed, as well as the driving force of climatic factors. Results showed that there was an agreement between FPAR values from this study and those of the MODIS product in seasonal dynamic, and the spatial patterns of FPAR vary with vegetation type distribution and seasonal cycles. The time series of average FPAR revealed a strong seasonal variation, regular periodic variations from January 1982 to December 2006, and opposite patterns between the Northern and Southern Hemispheres. Evergreen vegetation FPAR values were close to 0.7. A clear single-peak curve was observed between 30°N and 80°N – an area covered by deciduous vegetation. In the Southern Hemisphere, the time series fluctuations of FPAR averaged by 0.7° latitude zones were not clear compared to those in the Northern Hemisphere. A significant positive correlation (P<0.01) was observed between the seasonal variation of temperature and precipitation and FPAR over most other global meteorological sites.     

Increasing the detail of European land use/cover data by combining heterogeneous data sets

ABSTRACT

Data on land use and land cover (LULC) are a vital input for policy-relevant research, such as modelling of the human population, socioeconomic activities, transportation, environment, and their interactions. In Europe, CORINE Land Cover has been the only data set covering the entire continent consistently, but with rather limited spatial detail. Other data sets have provided much better detail, but either have covered only a fraction of Europe (e.g. Urban Atlas) or have been thematically restricted (e.g. Copernicus High Resolution Layers). In this study, we processed and combined diverse LULC data to create a harmonised, ready-to-use map covering 41 countries. By doing so, we increased the spatial detail (from 25 to one hectare) and the thematic detail (by seven additional LULC classes) compared to the CORINE Land Cover. Importantly, we decomposed the class ‘Industrial and commercial units’ into ‘Production facilities’, ‘Commercial/service facilities’ and ‘Public facilities’ using machine learning to exploit a large database of points of interest. The overall accuracy of this thematic breakdown was 74%, despite the confusion between the production and commercial land uses, often attributable to noisy training data or mixed land uses. Lessons learnt from this exercise are discussed, and further research direction is proposed.     

Spatial prediction of landslide susceptibility by combining evidential belief function,logistic regression and logistic model tree

Abstract

In this study, we introduced novel hybrid of evidence believe function (EBF) with logistic regression (EBF-LR) and logistic model tree (EBF-LMT) for landslide susceptibility modelling. Fourteen conditioning factors were selected, including slope aspect, elevation, slope angle, profile curvature, plan curvature, topographic wetness index (TWI), stream sediment transport index (STI), stream power index (SPI), distance to rivers, distance to faults, distance to roads, lithology, normalized difference vegetation index (NDVI), and land use. The importance of factors was assessed using correlation attribute evaluation method. Finally, the performance of three models was evaluated using the area under the curve (AUC). The validation process indicated that the EBF-LMT model acquired the highest AUC for the training (84.7%) and validation (76.5%) datasets, followed by EBF-LR and EBF models. Our result also confirmed that combination of a decision tree-logistic regression-based algorithm with a bivariate statistical model lead to enhance the prediction power of individual landslide models.     

土地利用视角空气污染空间分异的地理分析

针对土地利用/覆盖（land-use and land-cover,LULC）方式是否影响城市空气污染空间分异特征形成的问题,利用遥感技术和景观生态学方法分别获取长株潭城市群核心城区LULC及其景观格局,绘制空气污染物浓度与气象影响因子空间分异图,引入地理探测器定量分析土地因子在融合气象要素前后对NO₂、PM₁₀、O₃、PM_2.5浓度空间分布差异的贡献强度。结果表明,建设用地面积比例越高,林地越低,NO₂、PM_2.5浓度越高,O₃越低。非建设用地区域,污染物浓度随着土地景观格局破碎度、多样性指数值增大而升高,建设用地区域反之。LULC和土地景观格局的复合因子贡献力（P0.03~0.28）高于两者任意单独因子贡献力（P:0.01~0.11）,融合气象要素后,LULC对空气污染物空间分异特征形成的因子贡献力（P:0.18~0.53）显著增强。     

时空过程对象的LULC时空演变分析算法

地理时空变化是地理学研究的重要内容之一,如何用计算机技术来表达空间数据的时空变化独具前瞻性。从揭示LULC时空演变过程和挖掘时空演变规律出发,讨论了基于地类图斑的时空演变过程类型与判定方法,并构建了一种基于地类图斑的时空变化分析算法。通过对抚仙湖流域近40年来LULC时空演变分析,验证了算法的可靠性与有效性。表明该方法可用于地表覆盖等地理要素的时空变化过程分析,能较好地揭示地理要素及其属性在时间轴上的改变过程。     

Tracking desertification on the Mongolian steppe through NDVI and field-survey data

Abstract

Changing environmental and socio-economic conditions make land degradation, a major concern in Central and East Asia. Globally satellite imagery, particularly Normalized Difference Vegetation Index (NDVI) data, has proved an effective tool for monitoring land cover change. This study examines 33 grassland water points using vegetation field studies and remote sensing techniques to track desertification on the Mongolian plateau. Findings established a significant correlation between same-year field observation (line transects) and NDVI data, enabling an historical land cover perspective to be developed from 1998 to 2006. Results show variable land cover patterns in Mongolia with a 16% decrease in plant density over the time period. Decline in cover identified by NDVI suggests degradation; however, continued annual fluctuation indicates desertification – irreversible land cover change – has not occurred. Further, in situ data documenting greater cover near water points implies livestock overgrazing is not causing degradation at water sources. In combination of the two research methods – remote sensing and field surveys – strengthen findings and provide an effective way to track desertification in dryland regions.     

1982-2013年青藏高原植被动态变化时序分析

针对不同的数据源及时间和空间尺度会使植被覆盖度及其与气象因子影响的结果有所差别这一情况,该文基于青藏高原1982-2012年GIMMS NDVI和2001-2013年MODIS NDVI遥感数据集,结合研究区内12个典型的气象站点数据,进行了青藏高原地区植被覆盖时空动态变化规律及其与气象因子响应的时序分析,并利用重合时间段的数据对比分析了两种传感器在青藏高原地区对植被动态变化监测方面的差异.结果表明:近30年来,青藏高原地区植被呈整体改善趋势,尤其是高海拔地区;不同阶段植被的变化趋势有所不同;两种传感器在反映植被动态变化趋势上差异显著,但两者与气候因子的响应规律相同.     

Soil characterization based on land cover heterogeneity over a tropical landscape: an integrated approach using earth observation data-sets

Soil is a vital part of the natural environment and is always responding to changes in environmental factors, along with the influences of anthropogenic factors and land use changes. The long-term change in soil properties will result in change in soil health and fertility, and hence the soil productivity. Hence, the main aim of this paper focuses on the analysis of land use/land cover (LULC) change pattern in spatial and temporal perspective and to present its impact on soil properties in the Merawu catchment over the period of 18?years. Post classification change detection was performed to quantify the decadal changes in historical LULC over the periods of 1991, 2001 and 2009. The pixel to pixel comparison method was used to detect the LULC of the area. The key LULC types were selected for investigation of soil properties. Soil samples were analysed in situ to measure the physicochemical soil properties. The results of this study show remarkable changes in LULC in the period of 18?years. The effect of land cover change on soil properties, soil compaction and soil strength was found to be significant at a level of <0.05.     

Investigating relationships between Landsat-7 ETM data and spatial segregation of LULC types under shifting agriculture in southern Cameroon

This paper investigates statistical relationships between land use/land cover (LULC), Landsat-7 ETM⁺ imagery and landscape mosaic structure in southern Cameroon where the conversion of tropical rain forest to shifting cultivation leads to dynamic processes, acting on the spatial aggregation of various LULC types. A Global Positioning System (GPS) was used in the field to identify a total of 171 shifting cultivation patches representing eight LULC types in two sub-areas. Because of the lack of a cloud-free image for the date of field sampling, the ETM⁺ imagery was acquired 2 months after field survey, during which it was assumed that no significant changes in LULC occurred (all dry season). Per pixel correlations were developed between spectral reflectance data, vegetation indices and LULC. As an exploratory study, several statistical methods (analysis of variance, means separations (Tukey HSD), principal component analysis (PCA), geo-statistical analysis, image classification and landscape metrics) were applied on point data and sensor images for evaluating the spatial variability within the landscape. Most variables explained 30–72% of LULC variation in the whole dataset. Those variables with high information content of LULC (infrared bands 4, 5, 7 and derived indices and PC1) also showed long ranges (6 km) spatial dependence as compared to those varying only within 1 km range. The results of these statistical analyses suggested the need to group some LULC types and the application of the Maximum Likelihood Classifier (MLC) for supervised classification provided a LULC map with the highest accuracy (81%) after consolidation of perennial LULC types, such as bush fallow, forest fallow and cocoa plantations. Landscape metrics computed from this map showed a high level of patch diversity and connectivity within the landscape and provided input data that can further be used to simulate predictive maps as substitute to cloud-covered sensor imageries. Landsat-7 ETM⁺ imagery proved to be useful in discriminating (with about 80% accuracy) the most dynamic LULC types such cropped plots and young fallow patches (shifting every season) and the extension front of the agricultural landscape.     

Generation of high spatial and temporal resolution NDVI and its application in crop biomass estimation

Abstract

While data like HJ-1 CCD images have advantageous spatial characteristics for describing crop properties, the temporal resolution of the data is rather low, which can be easily made worse by cloud contamination. In contrast, although Moderate Resolution Imaging Spectroradiometer (MODIS) can only achieve a spatial resolution of 250 m in its normalised difference vegetation index (NDVI) product, it has a high temporal resolution, covering the Earth up to multiple times per day. To combine the high spatial resolution and high temporal resolution of different data sources, a new method (Spatial and Temporal Adaptive Vegetation index Fusion Model [STAVFM]) for blending NDVI of different spatial and temporal resolutions to produce high spatial–temporal resolution NDVI datasets was developed based on Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM). STAVFM defines a time window according to the temporal variation of crops, takes crop phenophase into consideration and improves the temporal weighting algorithm. The result showed that the new method can combine the temporal information of MODIS NDVI and spatial difference information of HJ-1 CCD NDVI to generate an NDVI dataset with both high spatial and high temporal resolution. An application of the generated NDVI dataset in crop biomass estimation was provided. An average absolute error of 17.2% was achieved. The estimated winter wheat biomass correlated well with observed biomass (R ² of 0.876). We conclude that the new dataset will improve the application of crop biomass estimation by describing the crop biomass accumulation in detail. There is potential to apply the approach in many other studies, including crop production estimation, crop growth monitoring and agricultural ecosystem carbon cycle research, which will contribute to the implementation of Digital Earth by describing land surface processes in detail.     

Modelling the impacts of civil war on land use and land cover change within Kono District,Sierra Leone: a socio-geospatial approach

Capturing the scope and trajectory of changes in land use and land cover (LULC) is critical to urban and regional planning, natural resource sustainability and the overall information needs of policy makers. Studies on LULC change are generally conducted within peaceful environments and seldom incorporate areas that are politically volatile. Consequently, the role of civil conflict on LULC change remains elusive. Using a dense time stack of Landsat Thematic Mapper images and a hybrid classification approach, this study analysed LULC changes in Kono District between 1986–1991, 1991–2002 and 2002–2007 with the overarching goal of elucidating deviations from typical changes in LULC caused by Sierra Leone's civil war (1991–2002). Informed by social survey and secondary data, this study engaged the drivers that facilitated LULC changes during war and non-war periods in a series of spatial regression models in exploring the interface between civil conflict and LULC change.     

Landsat-8 vs. Sentinel-2: examining the added value of sentinel-2’s red-edge bands to land-use and land-cover mapping in Burkina Faso

The availability of freely available moderate-to-high spatial resolution (10–30 m) satellite imagery received a major boost with the recent launch of the Sentinel-2 sensor by the European Space Agency. Together with Landsat, these sensors provide the scientific community with a wide range of spatial, spectral, and temporal properties. This study compared and explored the synergistic use of Landsat-8 and Sentinel-2 data in mapping land use and land cover (LULC) in rural Burkina Faso. Specifically, contribution of the red-edge bands of Sentinel-2 in improving LULC mapping was examined. Three machine-learning algorithms – random forest, stochastic gradient boosting, and support vector machines – were employed to classify different data configurations. Classification of all Sentinel-2 bands as well as Sentinel-2 bands common to Landsat-8 produced an overall accuracy, that is 5% and 4% better than Landsat-8. The combination of Landsat-8 and Sentinel-2 red-edge bands resulted in a 4% accuracy improvement over that of Landsat-8. It was found that classification of the Sentinel-2 red-edge bands alone produced better and comparable results to Landsat-8 and the other Sentinel-2 bands, respectively. Results of this study demonstrate the added value of the Sentinel-2 red-edge bands and encourage multi-sensoral approaches to LULC mapping in West Africa.     

Spatio‐temporal prediction of land surface temperature using semantic kriging

Spatio‐temporal prediction and forecasting of land surface temperature (LST) are relevant. However, several factors limit their usage, such as missing pixels, line drops, and cloud cover in satellite images. Being measured close to the Earth's surface, LST is mainly influenced by the land use/land cover (LULC) distribution of the terrain. This article presents a spatio‐temporal interpolation method which semantically models LULC information for the analysis of LST. The proposed spatio‐temporal semantic kriging (ST‐SemK) approach is presented in two variants: non‐separable ST‐SemK (ST‐SemK_NSep) and separable ST‐SemK (ST‐SemK_Sep). Empirical studies have been carried out with derived Landsat 7 ETM+ satellite images of LST for two spatial regions: Kolkata, India and Dallas, Texas, U.S. It has been observed that semantically enhanced spatio‐temporal modeling by ST‐SemK yields more accurate prediction results than spatio‐temporal ordinary kriging and other existing methods.     

Effect of cover management factor in quantification of soil loss: case study of Sungai Akah subwatershed,Baram River basin Sarawak,Malaysia

The present study evaluates the effectiveness and suitability of cover management factors (C factor) generated through different techniques like land use/land cover-based arbitrary value (C_LULC), Normalised Different Vegetation Index-based methods C_NDVI1 and C_NDVI2 and Modified Soil Adjusted Vegetation Index 2-based method (C_MSAVI2). The C factors generated using these four methods were tested in the calculation and assessment of annual average soil loss from an upland forested subwatershed in the Baram river basin using the Revised Universal Soil Loss Equation (RUSLE). The four cover management factor maps generated by this analysis show some variation among the results. The LULC method uses a single arbitrary value for each LULC type mapped in the subwatershed. The other three methods show a range of C values within each mapped LULC type. The effects of these variations were tested in the RUSLE by keeping the factors such as rainfall erosivity (R), soil erodibility (K), slope-length and steepness (LS) constant. The maximum annual average soil loss is 1191 t. ha^?1. y^?1 based on the C_LULC. Soil losses estimated with other three methods are very different compared to those estimated with the C_LULC method. The highest calculated soil loss values were 1832, 1674 and 1608 t. ha^?1. y^?1 in the study area based, respectively, on C_NDVI1, C_NDVI2 and C_MSAVI2 C factors. These maximum values represent the worst pixel scenario values of soil loss in the region. The statistical analysis performed indicates different relationship between the parameters and suggests the acceptance of the methodology based on C_NDVI2 for the study area, instead of a single value method such as C_LULC. Among the other two methods, the C_MSAVI2 was found to be more consistent than the C_NDVI1 method, but both methods lead to over-prediction of annual soil loss rate and therefore need to be reconsidered before applied in the RUSLE.     

A comparative study on groundwater spring potential analysis based on statistical index,index of entropy and certainty factors models

The main aim of this study is to generate groundwater spring potential maps for the Ningtiaota area (China) using three statistical models namely statistical index (SI), index of entropy (IOE) and certainty factors (CF) models. Firstly, 66 spring locations were identified by field surveys, out of which, 46 (70%) spring locations were randomly selected for training the models and the rest 20 (30%) spring locations were used for validation. Secondly, 12 spring influencing factors, namely slope angle, slope aspect, altitude, profile curvature, plan curvature, sediment transport index, stream power index, topographic wetness index, distance to roads, distance to streams, lithology and normalized difference vegetation index (NDVI) were derived from the spatial database. Subsequently, using the mentioned factors and the three models, groundwater spring potential values were calculated and the results were plotted in ArcGIS 10.0. Finally, the area under the curve was used to validate groundwater spring potential maps. The results showed that the IOE model, with the highest success rate of 0.9126 and the highest prediction rate of 0.9051, showed the preferable performance in this study. The results of this study may be helpful for planners and engineers in groundwater resource management and other similar watersheds.     

Satellite Data Classification Using Open Source Support

In this study we explored the potential of open source data mining software support to classify freely available Landsat image. The study identified several major classes that can be distinguished using Landsat data of 30 m spatial resolution. Decision tree classification (DTC) using Waikato environment for knowledge analysis (WEKA), open source software is used to prepare land use land cover (LULC) map and the result is compared with supervised (maximum likelihood classifier – MLC) and unsupervised (Iterative self-organizing data analysis technique - ISODATA clustering) classification techniques. The accuracy assessment indicates highest accuracy of the map prepared using DTC with overall accuracy (OA) 92 % (kappa = 0.90) followed by MLC with OA 88 % (kappa = 0.84) and ISODATA OA 76 % (kappa = 0.69). Results indicate that data set with a good definition of training sites can produce LULC map having good overall accuracy using decision tree. The paper demonstrates utility of open source system for information extraction and importance of DTC algorithm.     

Comparing spatiotemporal patterns in Eurasian FPAR derived from two NDVI-based methods

Abstract

A long-term, consistent Fraction of Absorbed Photosynthetically Active Radiation (FPAR) product is necessary to study the spatial and temporal patterns of vegetation dynamics associated with climatic changes and human activities. In this study, Eurasia was selected as the study area. The relationship between FPAR and simple infrared/red ratio relationship (SR FPAR), and that between Moderate Resolution Imaging Spectroradiometer (MODIS) FPAR and a Normalised Difference Vegetation Index (NDVI) look-up table (LUT FPAR) were employed to estimate FPAR from 1982 to 2006 by different land cover types, focusing on the comparisons of spatiotemporal FPAR patterns between the two FPAR datasets. The results showed high agreement between MODIS standard FPAR and estimated FPAR in seasonal dynamics with peak values in July. The LUT FPAR was close to MODIS standard FPAR and larger than SR FPAR. The SR and LUT FPAR showed the same spatial distribution and inter-annual variation patterns and were primarily determined by land cover types. An overall increasing trend in FPAR was observed from 1982 to 2006, with reductions from 1991 to 1994 and 2000 to 2002. The inter-annual dynamics in evergreen broadleaf forests showed a decreasing trend over 25 years, while non-forest vegetation FPAR values had slow, stable growth in inter-annual variation.     

Driving forces of grassland vegetation changes in Chen Barag Banner,Inner Mongolia

ABSTRACT

Inner Mongolia is an important ecological zone of northern China and 67% of its land area is grassland. This ecologically fragile region has experienced significant vegetation degradation during the last decades. Although the spatial extents and rates of vegetation change have previously been characterized through various remote sensing and GIS studies, the underlying driving factors of vegetation changes are still not well understood. In this study, we first used time-series MODIS NDVI data from 2000 to 2016 to characterize the temporal trend of vegetation changes. These vegetation change trends were compared with climate and socioeconomic variables to determine the potential drivers. We used a set of statistical methods, including multiple linear regression (MLR), spatial correlation analysis, and partial least squares (PLS) regression analyzes, to quantify the spatial distribution of the driving forces and their relative importance to vegetation changes. Results show that the main driving factors and their impact magnitude (weight) are in the order of human activities (r = -0.785, p < 0.01, VIP = 1.37), precipitation (r = 0.541, p < 0.05, VIP = 0.89), temperature (r = -0.319, p > 0.05 VIP = 0.59). The area affected by human activities was 10.57%. Specific human activities, such as coal mining and grazing were negatively associated with vegetation cover, while eco-engineering projects had positive impacts. This study provided thorough quantification of driving forces of vegetation change and enhanced our understanding of their interactions. Our integrated geospatial-statistical approach is particularly important for sustainable development of ecosystem balance in Chen Barag Banner and other areas facing similar challenges.     

东亚土地覆盖对ENSO事件的响应特征

对1982—1993年气候年际变化的强信号——ENSO进行了确认及再分类。以美国地质调查局EROS中心提供的AVHRR 8km NDVI为数据源，应用地理信息系统技术，计算了1982—1993年每年夏季(5—9月)NDVI平均影像。在此基础上用数据断面分析法对ENSO年东亚地区土地覆盖的空间分布进行了分析，再用主成分分析法对同一时间序列NDVI平均影像进行了运算，发现其第7主成分影像所反映的土地覆盖分布与数据断面分析法所反映的结果是一致的。对此，进一步分析了第7主成分的特征向量与代表ENSO变化特征的南方涛动指数(SOI)之间的关系，进而，对ENSO驱动下的东亚地区土地覆盖年际变化的空间分布特征进行了总结。