Use of remote sensing-derived variables in developing a forest fire danger forecasting system

Our aim was to develop a remote sensing-based forest fire danger forecasting system (FFDFS) and its implementation in forecasting 2011 fire season in the Canadian province of Alberta. The FFDFS used Moderate Resolution Imaging Spectroradiometer (MODIS)-derived 8-day composites of surface temperature, normalized multiband drought index, and normalized difference vegetation index as input variables. In order to eliminate the data gaps in the input variables, we propose a gap-filling technique by considering both of the spatial and temporal dimensions. These input variables were calculated during the i period and then integrated to forecast the fire danger conditions into four categories (i.e., very high, high, moderate, and low) during the i + 1 period. It was observed that 98.19 % of the fire fell under “very high” to “moderate” danger classes. The performance of this system was also demonstrated its ability to forecast the worst fires occurred in Slave Lake and Fort McMurray region during mid-May 2011. For example, 100 and 94.0 % of the fire spots fell under “very high” to “high” danger categories for Slave Lake and Fort McMurray regions, respectively.     

Identifying wildland fire ignition factors through sensitivity analysis of a neural network

Artificial neural networks (ANNs) show a significant ability to discover patterns in data that are too obscure to go through standard statistical methods. Data of natural phenomena usually exhibit significantly unpredictable non-linearity, but the robust behavior of a neural network makes it perfectly adaptable to environmental models such as a wildland fire danger rating system. These systems have been adopted by many developed countries that have invested in wildland fire prevention, and thus civil protection agencies are able to identify areas with high probabilities of fire ignition and resort to necessary actions. Since one of the drawbacks of ANNs is the interpretation of the final model in terms of the importance of variables, this article presents the results of sensitivity analysis performed in a back-propagation neural network (BPN) to distinguish the influence of each variable in a fire ignition risk scheme developed for Lesvos Island in Greece. Four different methods were utilized to evaluate the three fire danger indices developed within the above scheme; three of the methods are based on network’s weights after the training procedure (i.e., the percentage of influence—PI, the weight product—WP, and the partial derivatives—PD methods), and one is based on the logistic regression (LR) model between BPN inputs and observed outputs. Results showed that the occurrence of rainfall, the 10-h fuel moisture content, and the month of the year parameter are the most significant variables of the Fire Weather, Fire Hazard, and Fire Risk Indices, respectively. Relative humidity, elevation, and day of the week have a small contribution to fire ignitions in the study area. The PD method showed the best performance in ranking variables’ importance, while performance of the rest of the methods was influenced by the number of input parameters and the magnitude of their importance. The results can be used by local forest managers and other decision makers dealing with wildland fires to take the appropriate preventive measures by emphasizing on the important factors of fire occurrence.     

近20余年来西北地区植被变化特征分析

利用1982—2003年8 km分辨率的NDVI数据集,选中国西北地区森林、草原、灌溉农业、雨养农业区不同类型植被为研究区,分析了植被年、年际变化特征,并对植被覆盖空间变化进行动态研究.结果表明:森林、草原、灌溉农业区和以春小麦为主的雨养农业区NDVI年变化为单峰型曲线,以冬小麦为主的雨养农业区NDVI曲线呈双峰型;同一类型的植被NDVI受纬度或海拔高度的影响,绿峰出现时间存在1个月的位相差.22 a来森林植被NDVI多呈下降趋势,草原植被区为上升趋势;雨养农业区变化不大,灌溉植被区呈显著的上升趋势.西北东部雨养农业区植被波动频率和幅度最大,是受降水影响最敏感的地区;森林植被次之;有灌溉条件的绿洲植被,年际间波动最小.22 a间西北地区植被以增加趋势为主,增加面积约为20.5%,主要分布在新疆和河西走廊绿洲、黄河沿岸灌区以及青海草区,水分条件充足的绿洲是NDVI增加最显著的区域;NDVI减少地区面积为4.77%,主要分布在西北东部.     

A comparative assessment between linear and quadratic discriminant analyses (LDA-QDA) with frequency ratio and weights-of-evidence models for forest fire susceptibility mapping in China

Forest fire is known as an important natural hazard in many countries which causes financial damages and human losses; thus, it is necessary to investigate different aspects of this phenomenon. In this study, performance of four models of linear and quadratic discriminant analysis (LDA and QDA), frequency ratio (FR), and weights-of-evidence (WofE) was investigated to model forest fire susceptibility in the Yihuang area, China. For this purpose, firstly, a forest fire locations map was prepared implementing MODIS satellite images and field surveys. Then, it was classified into two groups including training (70%) and validation (30%) by a random algorithm. In addition, 13 forest fire effective factors were prepared and used such as slope degree, slope aspect, altitude, Topographic Wetness Index (TWI), plan curvature, land use, Normalized Difference Vegetation Index (NDVI), annual rainfall, distance from roads and rivers, wind effect, annual temperature, and soil texture. Using the training dataset and effective factors, LDA, QDA, FR, and WofE models were applied and forest fire susceptibility maps were prepared. Finally, area under the curve (AUC) of receiver operating characteristics (ROC) was implemented for investigating the performance of the models. The results depicted that WofE had the best performance (AUC = 82.2%), followed by FR (AUC = 80.9%), QDA (AUC = 78.3%), and LDA (AUC = 78%), respectively. The results of this study showed the high contribution of altitude, slope degree, and temperature. On the other hand, it was seen that slope aspect and soil had the lowest importance in forest fire susceptibility mapping. From the AUC results, it can be concluded that FR, WofE, LDA, and QDA had acceptable performance and could be used for forest fire susceptibility mapping at the regional scale.     

Assessing vegetation dynamics in the Three-North Shelter Forest region of China using AVHRR NDVI data

The Three-North Shelter Forest Programme (TNSFP) covers 551 Chinese counties and an area of 4,069,000 km² mostly in arid and semi-arid regions. In this paper, we discuss the temporal and spatial changes in value of the normalized-difference vegetation index (NDVI) in this region, and the relationships between NDVI and climatic factors (temperature and precipitation) based on NOAA Advanced Very High Resolution Radiometer Global Inventory Modeling and Mapping Studies NDVI data with 8-km resolution from 1982 to 2006. During the past 25 years, the vegetation cover has generally increased in eastern regions of China and the oasis in the north piedmont of Tianshan Mountains, but has decreased northwest of Xinjiang and in the Hulunbeier Plateau. The multi-year monthly average NDVI distribution map showed that NDVI increased from April to August, but in the western and northern plateau areas, the lower temperatures and high altitude created a shorter growing season (1 or 2 months). The vegetation of the study area has generally increased in the regions covered by the TNSFP. Linear regression analysis of the vegetation cover showed an increasing trend over large areas. The largest annual growth rate per pixel (the slope of the regression) was 0.009; the largest negative annual change was −0.004. The correlation between NDVI and precipitation was higher than that between NDVI and temperature, suggesting that precipitation is the most important factor that affects NDVI changes in the study area, especially for temperate desert vegetation in northwestern China.     

Application of meteorological and vegetation indices for evaluation of drought impact: a case study for Rajasthan, India

Drought is a serious climatic condition that affects nearly all climatic zones worldwide, with semi-arid regions being especially susceptible to drought conditions because of their low annual precipitation and sensitivity to climate changes. Drought indices such as the standardized precipitation index (SPI) using meteorological data and vegetation indices from satellite data were developed for quantifying drought conditions. Remote sensing of semi-arid vegetation can provide vegetation indices which can be used to link drought conditions when correlated with various meteorological data based drought indices. The present study was carried out for drought monitoring for three districts namely Bhilwara, Kota and Udaipur of Rajasthan state in India using SPI, normalized difference vegetation index (NDVI), water supply vegetation index (WSVI) and vegetation condition index (VCI) derived from the Advanced Very High resolution Radiometer (AVHRR). The SPI was computed at different time scales of 1, 2, 3, 6, 9 and 12 months using monthly rainfall data. The NDVI and WSVI were correlated to the SPI and it was observed that for the three stations, the correlation coefficient was high for different time scales. Bhilwara district having the best correlation for the 9-month time scale shows late response while Kota district having the best correlation for 1-month shows fast response. On the basis of the SPI analysis, it was found that the area was worst affected by drought in the year 2002. This was validated on the basis of NDVI, WSVI and VCI. The study clearly shows that integrated analysis of ground measured data and satellite data has a great potential in drought monitoring.     

最近18年来中国植被覆盖的动态变化

基于遥感和地理信息系统技术,利用NOAA-AVHRR数据对我国最近18年(1982～1999)来的植被覆盖的动态变化进行了分析.结果表明:我国植被覆盖的动态变化受气候波动的影响十分显著,并且这种变化的区域性差异明显.18年来,NDVI减小的地区主要分布在西北地区和青藏高原,而NDVI增加的地区主要发生在东部地区;20世纪80年代和90年代的NDVI变化趋势之间存在较大差异;90年代NDVI减小的区域明显地比80年代增加,特别是西北干旱地区NDVI的下降趋势明显.我国珠江三角洲和长江三角洲地区是18年来植被覆盖下降趋势最明显的地区,表明快速城市化的影响.     

Remote sensing-based assessment of vegetation damage by a strong typhoon (Meranti) in Xiamen Island,China

Remote sensing is a cost-effective tool for assessing vegetation damage by typhoon events at various scales. Taking Xiamen Island, southeastern China, as a study case, this paper aimed to assess and analyze the vegetation damage caused by Typhoon Meranti landfalling on September 15, 2016, using two high spatial resolution remote sensing images before and after the typhoon event. Seven severely damaged vegetation regions were selected based on the classification of vegetation types and visual interpretation of the images. Regression analysis was used to correct seasonal variation of the two high-solution images before and after typhoon. The vegetation area of the whole of Xiamen Island and the selected seven regions before and after typhoon were then calculated, respectively. Two spectral vegetation indicators, normalized difference vegetation index (NDVI) and fractional vegetation coverage (FVC), were also retrieved for the whole island and the seven regions. By comparing the difference in NDVI values before and after the typhoon of the two high spatial resolution images, we analyzed the most affected vegetation areas, as well as the most seriously damaged vegetation species. The typhoon has caused a decrease in vegetation area by 95.1 ha across the whole Xiamen Island. The mean NDVI and FVC decreased by 0.209 and 13 percentage points, respectively. While, in the seven selected severely damaged areas, the mean NDVI decreased by 0.356–0.444 and FVC decreased by 27–42 percentage points. The visual inspection showed that the tone of typhoon-damaged vegetation became darker, the patches of damaged vegetation became smaller and more fragmented, and the gap between vegetation canopies became larger. The most affected vegetation areas occurred in the southeastern hilly area, Jinshang and Hubin South Roads, as well as the Wuyuan Bay area. The most seriously damaged vegetation type is broad-leaved trees, especially the species, Acacia confusa, Delonix regia, Bauhinia variegata, Chorisia speciosa, Ficus benjamina and F. Concinna.     

The relationship between drought activity and vegetation cover in Northwest China from 1982 to 2013

We investigated drought activity and the relationship between drought and vegetation in Northwest China over the period 1982–2013 using the standardized precipitation evapotranspiration index (SPEI) and the normalized difference vegetation index (NDVI). The indexes were, respectively, calculated from ground-based meteorological data and from remotely sensed satellite data. The spatial and temporal distributions of drought (SPEI) and of vegetation cover (NDVI) were compared using annual trends, and the relationships between these trends were analyzed. The results are: (1) Overall, Northwest China had a drought trend during the study period, although some a few regions show a significant wetness trend; (2) the mean annual NDVI fluctuates, but overall shows an increasing trend, particularly in some mountainous areas that have at least adequate water and vegetation cover, while unused land becomes degraded; (3) most regions show a positive correlation between SPEI and NDVI, although the western parts of the Tarim basin, Qaidam basin, and some regions in the southeastern part of study area show a negative correlation; and (4) the various regions respond differently to global climate change, but in general regions with more vegetation cover show increased vegetation growth, while regions with less vegetation cover are becoming degraded and thus more vulnerable to the adverse effects of climate change.     

Fire danger assessment with remote sensing: a case study in Northern China

Every year several million hectares of forest and grassland in China are affected by wildfires. The majority of wildfires occur in the northern part of China, where grasslands and forests are ubiquitous. A critical step toward the protection of life, property, and natural resources from wildfires is the development of a fire danger rating system. This paper presents a new method to assess fire danger that capitalizes on the abundance of environmental data available via remote sensing and applies this new method to the northern part of China. Using an analytical hierarchy process, a fire danger index was developed based on five environmental factors that are known to affect fire frequency and severity, including land surface temperature, vegetation curing, equivalent water thickness, vegetation continuity degree, and fuel weight. Data for these five factors were derived from satellite imagery, instead of point data, allowing for predictions to be made over a large geographic area. Fire danger ratings were then mapped for the region based on the fire danger index. In addition, the accuracy of the fire danger index was evaluated by statistical analyses. The fire danger index was significantly correlated with air temperature and precipitation, suggesting that changes in these two environmental variables will affect the predictions of the index.     

黔桂喀斯特山地NDVI的垂直分布特征

黔桂喀斯特山地地形复杂,植被覆盖度垂直特征分异显著,以往研究多从气候因子响应方面探讨其垂直分布差异,而研究区人地矛盾尖锐,人类活动对植被分布有重要的影响。文章以黔桂喀斯特山地为例,利用2010年MODIS13Q1 NDVI数据表征植被覆盖度,结合高程、坡度和坡向等地形特征,不同土地利用类型的分布情况,叠置分析研究区的NDVI垂直分布特征。结果表明:黔桂喀斯特山地以林地、耕地和草地为主,不同土地利用类型随海拔、坡度和坡向的变化呈现不同的分布特征。研究区NDVI平均值为0.59,其中林地NDVI最大,达到0.63,草地为0.58,耕地最小为0.54。空间分布上,贵州境内NDVI值大部分为0.5~0.6,广西境内自西北向东南NDVI值由0.8逐渐降低至0.4,以0.6~0.7为主。NDVI在垂直梯度上分布特征显著,与植被垂直地带性分布以及不同地类的垂直分布特征有密切关系。海拔分布上,NDVI在海拔小于200 m区间最小,400~600 m的区间最大;北部贵州整体海拔较高,但植被覆盖度较低;南部广西海拔较低,但植被覆盖度较高。坡度分布上,在坡度小于35°范围,随坡度增大,耕地、水域、建设用地面积迅速减少,林草地面积逐步增加,使得NDVI随坡度增大逐渐增大。坡向分布上,NDVI不随坡向变化呈现明显变化,仅偏东坡向稍大于偏西坡向。研究表明应根据海拔和坡度等地形特征,并考虑土地利用情况,因地制宜进行生态建设。      

气温、降水量和人类活动对长江流域植被NDVI的影响

为了了解气温、降水量和人类活动对流域植被NDVI(normalized difference vegetation index)的影响,以长江流域为研究区,运用一元线性回归分析法和Theil-Sen Median趋势分析法研究了长江流域气温、降水量和植被NDVI变化特征,同时利用相关分析法和残差分析法探讨气温、降水量和人类活动对植被NDVI变化的影响.结果表明:1960—2015年长江流域年平均温度显著上升,而降水量的变化趋势并不显著;1982—2015年流域NDVI呈显著增加趋势;1982—2015年流域NDVI与气温的相关性较高,然而与降水量的相关性并不显著;人类活动使流域NDVI增加的区域主要分布于流域北部、东南和西南部分地区,而使NDVI下降的区域位于流域中西部区域和长三角地区.气温对长江流域植被NDVI变化的影响大于降水,气候变暖和人类活动对流域生态环境具有一定程度的影响.      

Monitoring Forest Fire Danger with Remote Sensing

Our paper presents a review of the use of remote sensing technologies for forest wildfire danger monitoring, with a particular emphasis on its applicability to fuel moisture monitoring. Remote sensing of fuel moisture was first done with NOAA-AVHRR NDVI images, but NDVI is more related to vegetation greenness rather than water stress. NOAA-AVHRR surface temperature images were also used, alone or in association with NDVI images. Both kinds of images have a limited image availability due to cloud cover. This is not the case for radar images, but their use in fuel moisture monitoring is still experimental, because of the noisy effects of several factors. Finally, the paper discusses the operational potentials and limitations of the use of each kind of satellite data for fire danger monitoring.     

The relationship between fire behaviour measures and community loss: an exploratory analysis for developing a bushfire severity scale

Current fire danger scales do not adequately reflect the potential destructive force of a bushfire in Australia and, therefore, do not provide fire prone communities with an adequate warning for the potential loss of human life and property. To determine options for developing a bushfire severity scale based on community impact and whether a link exists between the energy release rate (power) of a fire and community loss, this paper reviewed observations of 79 wildfires (from 1939 to 2009) across Victoria and other southern states of Australia. A methodology for estimating fire power based on fuel loading, fire size and progression rate is presented. McArthur??s existing fire danger indices (FDIs) as well as fuel- and slope-adjusted FDIs were calculated using fire weather data. Analysis of possible relationships between fire power, FDIs, rate of spread and Byram??s fireline intensity and community loss was performed using exposure as a covariate. Preliminary results showed that a stronger relationship exists between community loss and the power of the fire than between loss and FDI, although fuel-adjusted FDI was also a good predictor of loss. The database developed for this study and the relationships established are essential for undertaking future studies that require observations of past fire behaviour and losses and also to form the basis of developing a new severity scale.     

中国陆相侏罗、白垩系划分对比述评

对我国东北、西南和东南沿海地区目前常用的陆相侏罗、白垩系的划分、对比方案的精度及其与海相标准分阶的对比证据进行了扼要分析和述评 ,指出东北地区陆相白垩系序列可以划分出 13个组 ,其精度已赶上了海相白垩系 12个阶的标准 ,而且许多组已找到了将海陆相地层进行对比的联系化石分子 ,亦具备了陆相地层建阶的条件。同时认为我国中生代陆相地层划分对比表中以往很少表现出地层缺失 ,大多数地区似乎从下向上都存在一个完整的沉积序列 ,这是与陆相地层形成的实际不相符的 ,今后要进一步做工作。     

Long-term Trends in Vegetation Dynamics and Forest Fires in Brandenburg (Germany) Under a Changing Climate

The human influence on environmental processes has been described for many types of land use. One of the oldest tools to modify people’s environment is fire, which has dominated fire regimes in many regions over long time scales. This paper focuses on a German case study region, where 80–90% of the fires are human-caused. The objectives of this study are the application of the Regional Fire Model (Reg-FIRM), a process-based fire model that is incorporated into the LPJ Dynamic Global Vegetation Model, to temperate forests under historic climate conditions and to explore ranges of potential impacts of future climate change on fire and vegetation dynamics. Simulation experiments are designed to simulate historic fire pattern and to explore influences of vegetation on fire. Simulated fire pattern reproduced the observed average fire conditions reasonably well although with a smaller amplitude. This leads to underestimation of extreme fire years as well as an overestimation of low fire years. Vegetation composition influenced fire spread conditions in the temperate forest and had little impact on fire ignition potentials, except when only broad-leaved deciduous forests were assumed. Fire is likely to change under climate change conditions. Simulated experiments were conducted to explore the effects of climate change and rising CO₂ concentration given the potential natural vegetation as the best-case for Brandenburg. Three GCM scenarios predicting different future climatic changes were applied, and resulted in quantitatively different future fire patterns. Depending on future precipitation pattern and the influence of the CO₂ effect on canopy conductance and thus litter moisture, fire was predicted to either decrease or slightly increase in Brandenburg forests, but the burnt area would not exceed current, extreme fire years. Generally, fire changes had no implication for vegetation composition in Brandenburg, but reduced vegetation carbon gain after 2050. In the HadCM3 application, simulated increase in grass cover due to a large burnt area after 2075 accelerated fire spread conditions, thus still increasing the burnt area, while climatic fire danger and number of fires already began to decline. These interactions underline the importance to consider the full range of fire processes and interactions with vegetation dynamics in a simulation model.     

Topographic controls on vegetation index in a hilly landscape: a case study in the Jiaodong Peninsula, eastern China

This study examined topographic influence on spatial and temporal variability in the normalized difference vegetation index (NDVI) derived from the Satellite Pour l’Observation de la Terre-Vegetation at the regional and landscape scales in the Jiaodong Peninsula. The generalized additive models were used to quantify the spatial variation of NDVI attributable to local terrain and topographically related variables including altitude, exposure to incoming solar radiation, topographic wetness index, distance to the nearest stream and distance from the coast. NDVI distribution shows significant dependence on topography. The variables explained 38.3 % of variance in NDVI at the peninsula, and 30–45.3 % of variance in NDVI at the woodland, cropland, and grassland landscapes. At the Jiaodong Peninsula scale, NDVI is influenced primarily by distance from the coast. However, topographic wetness index has the most explanatory power for NDVI at the woodland, cropland, and grassland landscapes. Through a statistical nonparametric correlation analysis (Spearman’s r), the study indicates that spatial distribution of NDVI changes during the period 1998–2009 and future change trend of persistence determined by Hurst exponent is closely associated with topography and topography-based attribution. These results highlight the importance of topographic changes at landscape and regional scales as an important control factor on NDVI patterns.     

1981-2006年西北干旱区NDVI时空分布变化对水热条件的响应

气候是植被变化的重要驱动因子. 利用1981-2006年GIMMS归一化植被指数（NDVI）时间序列数据,结合68个气象站降水、气温数据和DEM地形数据等资料,研究分析了西北干旱区植被活动的年、季变化和空间差异. 结果显示：在1981-2006年的26 a,西北干旱区植被的覆盖率增加了4.5%,年平均NDVI增加了3.2%;植被的生长季延长,主要表现在生长季的推迟. 从总体来说,植被覆盖率、生长季和NDVI值在2000年以前显著增加,而在2000年以后都呈现减小的趋势;其中,减少明显的区域是在伊犁河谷、中天山及平原区,在河流上游山区或源头以及部分河流两岸呈现增加态势;在年际变化上,大部分区域的气温、降水与NDVI相关性不强. 而年平均气温在4.58 ℃以下低温区和年降水在180 mm以上的相对湿润区,气温和降水都呈现正相关;在季节变化上,NDVI值在春季和秋季与温度相关显著,而夏季与降水相关性强. 2000年以后,植被覆盖率和NDVI值开始出现降低趋势与气温持续升高、降水量增幅下降有关.     

Evaluating relationships between soil chemical properties and vegetation cover at different slope aspects in a reclaimed dump

Conducting research about the relationships between soil chemical properties and vegetation coverage at different slope aspects is especially important in reconstructed ecosystems of vulnerable ecological regions. This study was conducted in the first reclaimed dump within the Pingshuo mining area of Shanxi Province, China, to analyze patterns of soil chemical properties (soil organic matter (SOM), soil total nitrogen (STN), soil available phosphorus (SAP) and soil available potassium (SAp) and vegetation coverage (NDVI) and their correlations at different slope aspects. In the reclaimed dump, 26 quadrats were established along four slope aspects (i.e., shady, semi-shady, sunny and semi-sunny slopes). There was no significant difference in SOM or STN among different slope aspects, while SAP differed between shady slopes compared to semi-shady, sunny and semi-sunny slopes; SAP differed significantly between semi-shady and semi-sunny slopes. The NDVI of semi-sunny slopes differed significantly from that of the other three aspects. There was variation in the relationships between NDVI and soil chemical properties, depending on the slope aspects. The logarithm of SOM and NDVI was related linearly on shady and semi-shady slopes, while NDVI was inversely related to the natural logarithm of the logarithm of SOM on sunny and semi-sunny slopes. STN and NDVI had a first-order function relationship on shady and semi-shady slopes, yet a quadratic function relationship on sunny and semi-sunny slopes. The relationships between SAP and NDVI were inverse on all types of slopes. On shady and semi-shady slopes, NDVI had a quadratic relationship with the logarithm of SAp, but it was well fitted by using a cubic function on sunny and semi-sunny slopes. The sensitivity coefficients of soil chemical properties and NDVI were different, and soil chemical properties changed differently depending on changes in NDVI at different slope aspects.     

关岭-贞丰石漠化治理示范区植被覆盖变化及其对气候因子的响应

为揭示喀斯特石漠化治理示范区植被覆盖变化以及气候因子对植被覆盖变化的影响,利用2006—2015年Landsat 30 m/16 d分辨率影像数据,采用最大合成法、NDVI差值指数和相关、偏相关分析法,系统分析示范区归一化植被指数的时空变化特征及其与气候因子的关系。结果表明:(1)2006—2015年最大NDVI平均值为0.39,NDVI较高覆盖区域在示范区南北边界,而较低覆区域以花江南岸为主;(2)2006年以来示范区极低(-1.210)、低(-0.669)和中等(-0.729)植被覆盖度呈减少趋势,高(1.359)和极高(1.247)植被覆盖度增加,整体上呈显著增加趋势;(3)本月NDVI与本月、上月、上上月降雨量和气温的相关性均通过显著水平0.05检验,且本月NDVI与本月降雨量相关性高于本月气温(RNDVI降雨 =0.782),本月NDVI与上月气温相关性高于上月降雨量(RNDVI气温 =0.771);(4)在月尺度上,示范区植被生长对降雨量无滞后期,而对气温存在1个月的滞后期。