Monitoring East African vegetation using AVHRR data

NOAA Advanced Very High Resolution Radiometer satellite data are applied to regional vegetation monitoring in East Africa. Normalized Difference Vegetation Index (NDVI) data for a one-year period from May 1983 are used to examine the phenology of a range of vegetation types. The integrated NDVI data for the same period are compared with an ecoclimatic zone map of the region and show marked similarities. Particular emphasis is placed on quantifying the phenology of the Acacia Commiphora bushlands. Considerable variation was found in the phenology of the bushlands as determined by the satellite NDVI, and is explained through the high spatial variability in the distribution of rainfall and the resulting green-up of the vegetation. The relationship between rainfall and NDVI is further examined for selected meteorological stations existing within the bushland. A preliminary estimate is made of the length of growing season using an NDVI thresholding technique     

典型资源型城市的植被覆盖变化及驱动力分析

干旱半干旱区植被变化研究较多,然而很少关注资源型城市社会经济对植被变化影响。基于2000~2020年鄂尔多斯市MOD13Q1数据、降雨和温度等气候数据、原煤产量等11个社会经济指标,结合GIS技术和线性回归法等统计学方法,对植被覆盖时空变化及其影响因素进行了研究。结果表明：①21年间鄂尔多斯的NDVI值介于0.233~0.395,呈波动性增长趋势,增长速率为0.059/10 a;下辖的8个区县的NDVI值也呈波动性增长趋势,但各地区存在差异。②鄂尔多斯植被呈东北高,西南低的分布特征,低植被区面积5.35万km²,占整个鄂尔多斯面积的61.58%,高植被区面积仅0.20万km²;植被改善区面积远远大于植被退化区面积,改善区占整个鄂尔多斯面积的52.19%,植被退化区仅占3.69%。③NDVI值与降雨量表现为极显著性正相关,相关系数为0.794（P<0.01）;NDVI变化与当月累计降雨量的相关系数较大,与1个月前温度的相关系数较大。④NDVI变化与11种社会经济指标均表现为极显著正相关,相关性为0.728~0.796（P<0.01）。鄂尔多斯植被恢复效果较好,降雨量和温度是影响植被生长的主要因素,NDVI变化对降雨量的响应无明显滞后性,对温度的响应存在一个月的滞后期,社会经济发展对植被覆盖的积极作用大于消极影响。     

Response of NDVI dynamics to precipitation in the Beijing–Tianjin sandstorm source region

This article analysed the spatio-temporal changes in vegetation cover in the Beijing–Tianjin sandstorm source region in China and related these changes to vegetation types based on the Advanced Very High Resolution Radiometer (AVHRR) normalized difference vegetation index (NDVI) data set from 1982 to 2006. The annual maximum NDVI and peak time were identified. The different periods (1–12 months) of accumulated precipitation before the peak time were then calculated at the grid scale for each year. On this basis, the NDVI–rainfall relationship and the temporal responses in this area were studied by calculating the correlation coefficient between the annual maximum NDVI and different periods (1–12 months) of accumulated precipitation before the occurrence of the annual maximum NDVI for each pixel. The results show an upward trend in regional vegetation, a significant recovery efficiency for grassland, and the evident degradation of cropland. Peak plant growth is significantly related to precipitation and is strongly positively correlated with precipitation in the previous period (1 month) regardless of vegetation type. The regions showing the strongest correlations between peak plant growth and 1 month cumulative rainfall are the western desert grassland, grassland to forest in the transitional hill regions, the mountains of Yanshan, and the Greater Hinggan Mountains.     

NDVI time series stochastic models for the forecast of vegetation dynamics over desertification hotspots

ABSTRACT

Land degradation in semi-arid natural environments is usually associated with climate vulnerability and anthropic pressure, leading to devastating social, economic and environmental impacts. In this sense, remotely sensed vegetation parameters, such as the Normalized Difference Vegetation Index (NDVI), are widely used in the monitoring and forecasting of vegetation patterns in regions at risk of desertification. Therefore, the objective of this study was to model NDVI time series at six desertification hotspots in the Brazilian semi-arid region and to verify the applicability of such models in forecasting vegetation dynamics. We used NDVI data obtained from the MOD13A2 product of the Moderate Resolution Imaging Spectroradiometer sensor, comprising 16-day composites time series of mean NDVI and NDVI variance for each hotspot during the 2000–2018 period. We also used rainfall measured by weather stations as an explanatory variable in some of the tested models. Firstly, we compared Holt-Winters with Box-Jenkins and Box-Jenkins-Tiao (BJT) models. In all hotspots the Box-Jenkins and BJT models performed slightly better than Holt-Winters models. Overall, model performance did not improve with the inclusion of rainfall as an exogenous explanatory variable. Mean NDVI series were modelled with a correlation of up to 0.94 and a minimum mean absolute percentage error of 5.1%. NDVI variance models performed slightly worse, with a correlation of up to 0.82 and a minimum mean absolute percentage error of 22.0%. After the selection of the best models, we combined mean NDVI and NDVI variance models in order to forecast mean-variance plots that represent vegetation state dynamics. The combined models performed better in representing dry and degraded vegetation states if compared to robust and heterogeneous vegetation during wet periods. The forecasts for one seasonal period ahead were satisfactory, indicating that such models could be used as tools for the monitoring of short-term vegetation states.     

On the relationship between monthly mean and maximum-value composite normalized vegetation indices

Abstract

A relationship between the maximum-value composite and monthly mean normalized difference vegetation index (NDVI) is derived statistically using data over the U.S. Great Plains during 1986. The monthly mean NDVI is obtained using a simple nine-day compositing technique based on the specifics of the scan patterns of the NOAA-9 Advanced Very High Resolution Radiometer (AVHRR). The results indicate that these two quantities are closely related over grassland and forest during the growing season. It is suggested that in such areas a monthly mean NDVI can be roughly approximated by 80 per cent of the monthly maximum NDVI, the latter being a standard satellite data product. The derived relationship was validated using data for the growing season of 1987.     

Synergism between NOAA-AVHRR and Meteosat data for studying vegetation development in semi-arid West Africa

Abstract

Rainfall estimates, based on cold cloud duration estimated from Meteosat data, are compared with vegetation development depicted by data of the normalized difference vegetation index (NDVI) from the National Oceanic and Atmospheric Administration's (NOAA) advanced very high resolution radiometer (AVHRR) for part of the Sahel. Decadal data from the 1985 and 1986 growing seasons are examined to determine the synergism of the datasets for rangeland monitoring. There is a general correspondence between the two datasets with a marked lag between rainfall and NDVI of between 10 and 20 days. This time lag is particularly noticeable at the beginning of the rainy season and in the more northern areas where rainfall is the limiting factor for growth. Principal component analysis was used to examine deviations from the general relationship between rainfall and the NDVI. Areas of low NDVI values for a given input of rainfall were identified: at a regional scale, they give an indication of areas of low production potential and possible degradation of ecosystems. This study demonstrates in a preliminary way the synergism of such datasets derived from satellite--borne sensors with coarse spatial resolution, which may provide new information for the improved management of the Sahelian grasslands.     

利用色调—亮度彩色分量的可见光植被指数

目的 无人机遥感具有高时效、高分辨率、低成本、操作简单等优势。但由于无人机通常只携带可见光传感器,无法计算由可见光-近红外波段组合所构造的植被指数。为解决这一问题,提出一种归一化色调亮度植被指数NHLVI （normalized hue and lightness vegetation index）。方法 通过分析HSL （hue-saturation-lightness）彩色空间模型,构建一种基于色调亮度的植被指数,将该植被指数以及其他常用的可见光植被指数,如归一化绿红差值指数NGRDI （normalized green-red difference index）、过绿指数ExG （excess green）、超绿超红差分指数ExGR （excess green minus excess red）等,分别与野外实测光谱数据和无人机多光谱数据的NDVI （normalized difference vegetation index）进行相关性比较;利用受试者工作特征曲线ROC （receiver operating characteristic curve）的特点确定阈值,并进行植被信息提取与分析。结果 NHLVI与NDVI相关性高（R²=0.776 8）,而其他可见光植被指数中,NGRDI与NDVI相关性较高（R²=0.687 4）;ROC曲线下面积大小作为评价不同植被指数区分植被与非植被的指标,NHLVI指数在ROC曲线下面积为0.777,小于NDVI （0.815）,但大于NGRDI （0.681）,区分植被与非植被能力较强。为进一步验证其精度,利用阈值法提取植被,NHLVI提取植被信息的总体精度为82.25%,高于NGRDI （79.75%）,尤其在植被稀疏区,NHLVI的提取结果优于NGRDI。结论 提出的归一化色调亮度植被指数,提取植被精度较高,适用于无人机可见光影像植被信息提取,为无人机可见光影像的应用提供了新方法。     

NDVI response to rainfall in different ecological zones in Jordan

The response of NDVI to rainfall was analyzed using NOAA/AVHRR satellite imagery acquired over a time period of ten growing seasons (1981 to 1992) and rainfall data from 16 weather stations in four ecological zones in Jordan. Results of linear regression analysis showed better response of NDVI to cumulative rainfall than to 10-day rainfall with best correlation in the Mediterranean zone. Significant relationships were found between seasonal rainfall and NDVI range (Δ_NDVI) with better correlations for logarithmic and power relationships than for linear relationship. A strong linear relationship occurred between the annual rainfall and end-of-season NDVI in the Mediterranean zone and weak or no correlation in other zones. The correlations were improved when the rainfall data were averaged, summed and correlated with the average NDVI. More agreement, however, was observed between the maximum NDVI image and rainfall than for the average NDVI image and rainfall. Results also showed that stratification of the data according to soil type and/or land cover would not necessarily improve the correlation. However, stratification of the data according to ecological zone demonstrated obvious differences between the NDVI-rainfall in the different zones.     

Assessment of rainfall and NDVI anomalies in Spain (1989–1999) using distributed lag models

In this study a link was established between anomalies in climatic and Advanced Very High Resolution Radiometer (AVHRR)/Normalized Difference Vegetation Index (NDVI) data in Spain for the period from 1989 to 1999 on a monthly and annual basis using multivariate distributed lag (DL) models and generalized least‐square (GLS) parameter estimation. In most areas significant time‐delayed correlation between anomalies of monthly rainfall and NDVI data was confined to an interval of 1 month. Locally higher lag orders of up to 3 months were found. By contrast, relationships between surface temperature and the NDVI were insignificant in the multivariate context at most locations. The multiple correlation coefficients of the DL models achieved 0.6 in the maximum. Regions characterized by the most significant NDVI–rainfall correlations include the southern forelands of the Pyrenees in Catal?na, rainfed agricultural areas in Extremadura, Andalusia, and the western parts of Castilla y Leon. Average ratios of rainfall to potential evapotranspiration (PET) in the sensitive areas ranged between 0.5 and 2, with annual rainfall amounts less than 700 mm. For each land‐cover class a linear discriminant analysis (LDA) was carried out to assess the environmental factors that might explain the differences in the NDVI–rainfall relationships. The highest discriminant coefficients and factor loadings were recorded for those factors that recurrently trigger water deficit in the sensitive regions, such as low total annual rainfall, large seasonal rainfall variability, high average PET and surface temperature. On the annual basis the lagged correlation of the NDVI and rainfall data was confined to natural vegetation (grassland and scrubland) areas in western Spain. This region suffered from a severe drought in the early 1990s, after which biomass production lagged several years behind improved rainfall conditions. The approach presented is useful for assessing the influence of climatic variables on the pattern of temporal anomalies in the NDVI or related vegetation parameters.     

Relaxation vegetation index in non-linear modelling of ground plant cover by satellite remote-sensing data

Abstract

In order to obtain a model equation for the calculation of percentage plant cover by multi-spectral radiances remotely-sensed by satellites, a regression procedure is used to connect space remote-sensing data to ground plant cover measurement. A traditional linear regression model using the normalized difference vegetation index (NDVI) is examined by remote-sensing data of the SPOT satellite and ground measurement of LCTA project for a test site at Hohenfels. Germany. A relaxation vegetation index (RVI) is proposed in a non-linear regression modelling to replace the NDVI in linear regression modelling to get a better calculation of percentage plant cover. The definition of the RVI is

where X _i is raw remote-sensing data in channel i. Using the RVI, the correlation coefficient between calculated and observed percentage plant cover for a test scene in 1989 reaches 0·9 while for the NDVI it is only 0·7; the coefficient of multiple determination R ² reaches 0·8 for the RVI while it is only 0·5 for the NDVI. Numerical testing shows that the ability of using the RVI to predict percentage plant cover by space remote-sensing data for the same scene or the scene in other years is much stronger than the NDVI.     

A new approach to predict normalized difference vegetation index using time-delay neural network in the arid and semi-arid grassland

ABSTRACT

Normalized difference vegetation index (NDVI) has been used to conduct important research on plant growth and vegetation productivity. In this paper, a new approach to predict NDVI based on precipitation in the grass-growing season for the arid and semi-arid grassland is proposed using time-delay neural network (TDNN). To intuitively know the ability of TDNN to learn the relationship between NDVI and precipitation and to predict NDVI, the performance of the TDNN model is compared with back propagation neural network (BPNN) trained with the same data. The results indicate that TDNN works well to predict precipitation. Moreover, the relationship between precipitation and NDVI can be predicted accurately by the proposed TDNN model. The results show the goodness-of-fit between the observed NDVI and predicted NDVI measured by the determination coefficient of R² being 0.999 from the TDNN model, with the mean absolute percentage error, mean absolute error, and root-mean-square error to be 0.23%, 0.20, and 0.19, respectively. The analysis shows that the proposed method can result in an accurate NDVI prediction. Thus, the algorithm is applied to predict the NDVI during the grass-growing season for the validation of the approach. This validation results suggest the potential application of this approach for reduction of overgrazing pressure and vegetation restoration in the arid and semi-arid grassland.     

Determinants of the interannual relationships between remote sensed photosynthetic activity and rainfall in tropical Africa

The response of photosynthetic activity to interannual rainfall variations in Africa South of the Sahara is examined using 20 years (1981-2000) of Normalised Difference Vegetation Index (NDVI) AVHRR data. Linear correlations and regressions were computed between annual NDVI and annual rainfall at a 0.5° latitude/longitude resolution, based on two gridded precipitation datasets (Climate Prediction Center Merged Analysis of Precipitation [CMAP] and Climatic Research Unit [CRU]). The spatial patterns were then examined to detect how they relate to the mean annual rainfall amounts, land-cover types as from the Global Land Cover 2000 data set, soil properties and soil types. Yearly means were computed starting from the beginning of the vegetative year (first month after the minimum of the NDVI mean regime), with a one-month lead for rainfall.One third of tropical Africa displays significant (95% c.l.) correlations between interannual NDVI variations and those of rainfall. At continental scale, soil types and soil properties are only minor factors in the overall distribution of the correlations. Mean annual rainfall amounts and land-cover types are much more discriminating. The largest correlations, mostly over 0.60, are distinctly found in semi-arid (200-600 mm annual rainfall) open grassland and cropland areas. The presence of one of these two determinants (semi-aridity, and favourable land-cover type, i.e. open grassland and cropland) in the absence of the other does not systematically result in a significant correlation between rainfall and NDVI. By contrast, NDVI variations are independent from those of rainfall in markedly arid environments and in most forest and woodland areas. This results from a low signal-to-noise ratio in the former, and the fact that precipitation is generally not a limiting factor in the latter.The marginal response of NDVI to a given increase/decrease in rainfall, as described by the slope of the regression, displays a similar pattern to that of the correlation, with maximum slopes in semi-arid regions, except that a weaker response is noted in more densely populated areas, suggesting an incidence of particular land-use and agricultural practises.One-year lag relationships between annual rainfall and NDVI in the next year were also considered. Ten percent of the grid-points show significant correlations, but the spatial patterns remain difficult to interpret.     

Vegetation dynamics of Zimbabwe investigated using NOAA-AVHRR NDVI from 1982 to 2006: a principal component analysis

The dominant modes of vegetation variability over Zimbabwe are investigated using principal component analysis (PCA) on National Oceanic and Atmospheric Administration Advanced Very High Resolution Radiometer (NOAA-AVHRR) normalized difference vegetation index (NDVI) monthly imagery from 1982 to 2006. Spectral analysis is also used to determine the periodicities of the component loadings. NDVI PCA-1 corresponds to the major vegetation types of Zimbabwe, and we demonstrated that grasslands and dry savannah have the strongest relationship with mean annual precipitation. Furthermore, the March–April loadings showed the highest correlation (r?=?0.73) with mean annual precipitation. NDVI PCA-1 sheds some light on the land reform challenge in Zimbabwe. NDVI PCA-2 is highly correlated (r?=?0.87) with the mean annual relative variability of the rainfall map indicating a southeast/north mode of anomalies associated with the convectional rainfall-bearing systems over Zimbabwe. NDVI PCA-2 is also highly correlated (r?=?0.86) with precipitation PCA-2. NDVI PCA-3 shows a southeast/west mode and is highly correlated (r?=?0.87) with precipitation PCA-3. A high correlation (r?=?0.66) is also noted between NDVI PCA-4 and the elevation map. Spectral analysis of the PCA loadings revealed several periodicities corresponding to those found in tropical sea surface temperatures (SSTs).     

Exploring the utility of NOAA AVHRR middle infrared reflectance to monitor the impacts of ENSO-induced drought stress on Sabah rainforests

The ever-wet tropics are under threat from ENSO events and there is a need for a monitoring system to analyse and describe their responses to such events. This letter explores the relative value of using NOAA AVHRR middle infrared (MIR) reflectance data and NDVI data for the monitoring of ENSO-related drought stress of a tropical forest ecosystem in Sabah, Malaysia. Relationships between rainfall and MIR reflectance were examined. Correlation coefficients are generally large and significant (at 0.1 level) while those between rainfall and NDVI were small and insignificant. This letter concludes that there is potential in using MIR reflectance for monitoring the effects of ENSO-induced drought stress on these forests and this has a bearing on how NOAA AVHRR data may be used to further our knowledge on the impacts of ENSO events on tropical forest environments.     

Mapping of boreal vegetation of a temperate mountain in China by multitemporal Landsat TM imagery

The Changbai Mountain Natural Reserve (2000 km 2 ), north-east China, is a very important ecosystem representing the temperate biosphere. The cover types were derived by using multitemporal Landsat TM imagery, which was modified with DEM data on the relationship between vegetation distribution and elevation. It was classified into 20 groups by supervised classification. By comparing the results of the classification of different band combinations, bands 4 and 5 of an image from 18 July 1997 and band 3 of an image from 22 October 1997 were used to make a false colour image for the final output, a vegetation map, which showed the best in terms of classification accuracy. The overall accuracy by individual images was less than 70%, while that of the multitemporal classification was higher than 80%. Further, on the basis of the relationship of vegetation distribution and elevation, the accuracy of multitemporal classification was raised from 85.8 to 89.5% by using DEM. Bands 4 and 5 showed a high ability for discriminating cover types. Images acquired in late spring and mid-summer were recognized better than other seasons for cover type identification. NDVI and band ratio of B4/B3 proved useful for cover type discrimination, but were not superior to the original spectral bands. Other band ratios like B5/B4 and B7/B5 were less important for improving classification accuracy. The changes of spectral reflectance and NDVI with season were also analysed with 10 images ranging from 1984 to 1997. Seperability of images in terms of classification accuracy was high in late spring and summer, and decreased towards winter. There were five vegetation zones on the mountain, from the base to the peak: deciduous forest zone, mixed forest zone, conifer forest zone, birch forest zone and tundra zone. Spruce-fir conifer dominated forest was the most dominant vegetation (33%), followed by mixed forest (26%), Korean pine forest (8%) and mountain birch forest (5%).     

Sahelian rangeland production in relation to rainfall estimates from Meteosat

Abstract

Abstract. Satellite data are routinely used to monitor the growing season over the Sahelian zone of Africa. This study seeks to relate the vegetation indices and the rainfall estimates, both derived from meteorological satellites, to help monitor and predict the production of rangelands and marginal agricultural areas. Plant water use was calculated from a simple model which lakes into account the timing and distribution of rainfall: over a three-year period, the response of the Normalised Difference Vegetation Index (NDVI) to this quantity was consistent and was spatially quantified for two calibration years. A predictive model for end-of-season accumulated NDVI was developed and validated for a test year.     

Analysis of trends in the Sahelian ‘rain-use efficiency’ using GIMMS NDVI, RFE and GPCP rainfall data

Rain-use efficiency (RUE; the ratio of vegetation productivity to annual precipitation) has been suggested as a measure for assessing land degradation in arid/semi-arid areas. In the absence of anthropogenic influence, RUE has been reported to be constant over time, and any observed change may therefore be attributed to non-rainfall impacts. This study presents an analysis of the decadal time-scale changes in the relationship between a proxy for vegetation productivity (ΣNDVI) and annual rainfall in the Sahel-Sudanian zone of Africa. The aim is to test the quality of data input and the usefulness of both the RUE approach and an alternative method for separating the effects on vegetation productivity of rainfall change and human impact. The analyses are based on earth observation of both rainfall (GPCP (Global Precipitation Climatology Project), 1982-2007 and RFE (Rainfall Estimate) (1996-2007)) and ΣNDVI (AVHRR GIMMS NDVI dataset, 1982-2007). It is shown that the increase in ΣNDVI has been substantial in the Sahel-Sudanian zone over the 1982-2007 period, whereas for the period 1996-2007 the pattern of ΣNDVI trends is more complex. Also, trend analysis of annual rainfall from GPCP data (2.5° resolution) and RFE data (0.1° resolution) suggests that rainfall has increased over both periods. Further it is shown that RUE values are highly correlated to rainfall, undermining the use of earth observation (EO)-based RUE (using ΣNDVI) as a means of separating rainfall impacts from other factors. An alternative method identify temporal trends in residuals of ΣNDVI, after regressing it against annual rainfall, is tested, yet is shown to be useful only where a high correlation between ΣNDVI and annual rainfall exists. For the areas in the Sahel-Sudanian zone for which this condition is fulfilled, trend analyses suggest very limited anthropogenic land degradation in the Sahel-Sudanian zone.     

Impact of climate variability on NDVI over the Indian region during 1981–2010

The normalized difference vegetation index (NDVI), derived from the Advanced Very High Resolution Radiometer (AVHRR) (1981–2000), and Moderate Resolution Imaging Spectroradiometer (MODIS) Aqua data (2000–2010) are analysed to examine their spatio-temporal variability over the Indian region. Climatic factors are well known to be associated with vegetation. Therefore, an attempt has also been made in this study to examine the impact of climate variability on NDVI over the Indian region. The average spatio-temporal patterns of NDVI suggest that the variability in NDVI is well associated with climatic factors such as rainfall and temperature. During hot weather, the all-India NDVI shows the lowest values; the values increase from the onset of the summer monsoon season (June–September) onwards over the Indian region. The NDVI attains its peak value in the month of October. The composite annual cycles of NDVI during drought and flood years also show similar features. During drought years, there is a decrease in all-India NDVI for all months. Opposite features are seen during flood years, with a substantial increase in all-India NDVI from the month of October onwards compared to normal years. This clearly indicates the impact of heavy summer monsoon rainfall over the country on NDVI during winter (October–December) and suggests that soil moisture gained by flood conditions helps the NDVI to increase. In contrast, drought conditions show an immediate effect on NDVI but the incremental changes are of smaller magnitude. Spatial patterns also show similar features, with negative anomalies in NDVI over large parts of the country during drought years and positive anomalies during flood years. There exist year-to-year variations in NDVI depending on the performance of the monsoon. NDVI is positively correlated with rainfall during the southwest (June–September) and northeast (October–December) monsoons over a large part of the country. Also, there exists strong lag correlation between summer monsoon rainfall of the current year and NDVI of the next year, indicating that an increase (decrease) in rainfall during the rainy seasons is favourable (unfavourable) for vegetation during the winter (January and February) and the pre-monsoon season (March–May) of the following year. Thus, the analysis shows significant impact of inter-annual variability of climate on the NDVI over the Indian region. Strong lag correlations between rainfall and NDVI indicate the potential for estimating NDVI over India by the regression method.     

Investigating spatial variation in the relationships between NDVI and environmental factors at multi-scales: a case study of Guizhou Karst Plateau,China

Knowing the spatial relationships between the normalized difference vegetation index (NDVI) and environmental variables is of great importance for monitoring rocky desertification. This article investigated the spatially non-stationary relationships between NDVI and environmental factors using geographically weighted regression (GWR) at multi-scales. The spatial scale-dependency of the relationships between NDVI and environmental factors was identified by scaling the bandwidth of the GWR model, and the appropriate bandwidth of the GWR model for each variable was determined. All GWR models represented significant improvements of model performance over their corresponding ordinary least squares (OLS) models. GWR models also successfully reduced the spatial autocorrelations of residuals. The spatial relationships between NDVI and environmental factors significantly varied over space, and clear spatial patterns of slope parameters and local coefficient of determination (R ²) were found from the results of the GWR models. The study revealed detailed site information on the different roles of related factors in different parts of the study area, and thus improved the model ability to explain the local situation of NDVI.