International Journal of Applied Earth Observation and Geoinformation

Transcription

1 International Journal of Applied Earth Observation and Geoinformation 18 (2012) Contents lists available at SciVerse ScienceDirect International Journal of Applied Earth Observation and Geoinformation jo u rn al hom epage: Integrating AVHRR and MODIS data to monitor NDVI changes and their relationships with climatic parameters in Northeast China Dehua Mao a,b, Zongming Wang a,, Ling Luo a, Chunying Ren a a Key Laboratory of Wetland Ecology and Environment, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun , China b Graduate University of Chinese Academy of Sciences, Beijing , China a r t i c l e i n f o Article history: Received 14 June 2011 Accepted 13 October 2011 Keywords: AVHRR GIMMS NDVI MODIS NDVI Temperature Precipitation Northeast China a b s t r a c t On the basis of AVHRR GIMMS NDVI and MODIS NDVI, we constructed monthly NDVI sequences covering Northeast China from 1982 to 2009 using a per-pixel unary linear regression model. The expanded NDVI passed the consistency check and were well used for analysis. The monthly NDVI trends were highly correlated with climatic changes. Spatially averaged NDVI in summer exhibited a downward trend with increased temperature and significantly decreased precipitation in the 28 years. NDVI trends were spatially heterogeneous, corresponding with the regional climatic features of different seasons. NDVI for the 95 meteorological stations exhibited significant correlations with monthly mean temperature and monthly precipitation during the study period. The NDVI temperature correlation was stronger than NDVI precipitation correlation in most stations and for all vegetation types. Different vegetation types showed various spatial responses to climatic changes Elsevier B.V. All rights reserved. 1. Introduction Normalized Difference Vegetation Index (NDVI) has been effectively used in vegetation dynamics monitoring and the study of vegetation responses to climatic changes at different scales during the past few years (e.g., Tucker et al., 2001; Zhou et al., 2001; Wang et al., 2003; Beck et al., 2006). Using remote sensing NDVI data to investigate vegetation changes and relationships between vegetation and climate had acquired abundant achievements. Fang et al. (2003) indicated that enhanced vegetation activities were observed in China from 1982 to 1999 based on AVHRR NDVI. Based on SPOT VGT dataset from 1998 to 2007, Qiu and Cao (2011) found that the vegetation across the country increased to varying degrees, and the health statues was improved as well. Many research results also showed that changes in vegetation are seriously influenced by temperature and precipitation. For example, Li et al. (2002) observed that a significant correlation exists between NDVI and ecoclimatic parameters, and that NDVI-growing degreedays correlation was stronger than NDVI-rainfall correlation. These studies presented reasonable and reliable conclusions regarding NDVI changes and the relationship between vegetation NDVI and climatic parameters. However, There have little researches focusing on long time sequence from 1980s to recent years, because the Corresponding author. address: zongmingwang@gmail.com (Z. Wang). NDVI data sequence is relatively short based on only one kind of data source in their analyses. Various NDVI datasets are available now: MODIS NDVI, AVHRR NDVI, SPOT VGT NDVI, and TM NDVI. The time sequence of AVHRR GIMMS NDVI is from 1982 to 2006, whereas that of MODIS NDVI is from 2000 to the present time. Single data source is constrained in time sequence analysis, and a long-term series analysis cannot be achieved. Developing a dataset that combines traditional NDVI AVHRR datasets used to monitor vegetation dynamics with new MODIS data with their narrower channels is necessary (Gitelson and Kaufman, 1998). The value from different remote sensors is discrepant because of differences in spectral bands, temporal compositing, spatial resolution, and other sensor characteristics (Morisette et al., 2004). Multiple sensors with different designs and resolutions must be used together to obtain a consistent and sufficiently long-term NDVI time-series data. The consistency and spectral differences in different data have been discussed by many studies (e.g., Tucker et al., 2005; Brown et al., 2006; Fensholt et al., 2009). Several approaches have been adopted in integrating data from multiple sensors (Steven et al., 2003). For example, simulated spectral response from multiple instruments and with simple linear equations was used to create conversion coefficients for transforming NDVI data from one sensor to another. Neural networks, as a tool for constructing NDVI time-series from AVHRR and MODIS, were used by Brown et al. (2008). Ding et al. (2010), using linear regression method, integrate GIMMS and SPOT VGT NDVI for monitoring grassland coverage in Tibetan Plateau of China from 1982 to The research that use per-pixel unary linear regression /$ see front matter 2011 Elsevier B.V. All rights reserved. doi: /j.jag

2 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) model to integrate different sensor data has not seen in the relevant report. Previous studies showed obvious responses of vegetation to climatic changes in northern mid- and high-latitudinal zones (e.g., Tucker et al., 2001; Gong and Shi, 2003; Slayback et al., 2003). Myneni et al. (1997) found an increase in plant growth associated with the lengthening of the active growing seasons in Ichii et al. (2002) observed that the NDVI increase in the northern mid- and high-latitudinal zones was related to the rise in temperature. Qi (1999) found that the yearly maximum NDVI was highly correlated with temperature at high latitudes of the northern hemisphere. Accordingly, Northeast China, located in the mid- and high-latitude zones, is a typical region sensitive to climatic changes. It is the eco-fencing of China and an important timber and commodity grain production base of the nation; thus, the study of vegetation change and relationship between NDVI and climatic parameters in this area is of significance. A number of articles about vegetation responses to climatic changes based on NDVI time series data in this region have been published (e.g., Zhang et al., 2001; Wang et al., 2009). However, those studies have short research duration based on one remote sensing data source. Are there any differences with those results during long time sequence? Thus far, long-term series analyses on NDVI and the relationships between NDVI and climatic parameters in Northeast China are needed. The purpose of this study is (i) to construct a long-term NDVI time-series covering by integrating AVHRR GIMMS NDVI and MODIS NDVI; (ii) present an analysis of the linear trends of vegetation, temperature, and precipitation, using the constructed NDVI, and investigate their relationship in growingseason and different seasons during ; (iii) make a comparison about the correlation of NDVI temperature and NDVI precipitation and correlations for different vegetation types. Meanwhile, authors also hope that results of this paper can supply an example for integrating different sources NDVI data to monitor long time-sequence NDVI change, and provide extended NDVI dataset as driving data for estimating long series net primary productivity of vegetation (NPP). 2. Materials and methods 2.1. Study area Northeast China, covering Heilongjiang, Jilin, and Liaoning Provinces, as well as the eastern parts of the Inner Mongolia Autonomous Region (IMAR) (i.e., Hulun Buir, Khingan, Tongliao, and Chifeng) as the border area in China, extends from E to E, and N to N (Fig. 1). Land area of Northeast China is about km 2. It is largely separated by the Amur, Argun, and Ussuri Rivers with Russia, with North Korea along the Yalu River, and Tumen River. The elevation of the entire region is below 1600 m. The study area is surrounded by middle and low mountains along three directions, including the Changbai Mountains in the southeast, the Greater Khingan Mountains in the northwest, and the Lesser Khingan Mountains in the northeast. Some plains are located in the central and southern parts and in the northeastern corner; Hulun Buir High Plain is located in the western tip; and hills and tablelands are situated between mountains and plains. The southern part of the study area is close to the Bohai Sea. The interior region is lined with some rivers, such as the Songhua, Mudan, Nen, and Liao Rivers. The major types of vegetation on hills and mountains are cold-temperate mixed broadleaved deciduous and needle-leaved forests, as well as cold-temperate coniferous forests. The dominant vegetation in the western parts of the region includes semi-arid shrubs, grassland, and temperate steppe. Most of the study area is characterized by a temperate Fig. 1. Location of the study area, Northeast China. monsoon continental climate, except areas located at >50 N latitude, which is dominated by the cold monsoon. Winter is long and cold, whereas summer is short. Air temperature spatially increases from north to south, with a mean annual value of 4 to 12 C. Precipitation varies greatly within and between years, with 70 80% of total precipitation occurring between mid-june and mid-august. It decreases from 1100 mm in the east to 250 mm in the west Data source The NDVI datasets used in this study are GIMMS NDVI and MODIS NDVI. The GIMMS NDVI dataset from the Global Inventory Monitoring and Modeling Systems group was derived from the NOAA/AVHRR land dataset at a spatial resolution of 8 km 8 km and 15-d interval. This dataset was corrected for calibration, sensor degradation, orbital drift, view geometry, cloud cover, volcanic aerosols, and other effects that are unrelated to vegetation change (Tucker et al., 2005). The monthly GIMMS NDVI dataset from 1982 to 2003 was obtained using the Maximum Value Composite (MVC) method. The spatial resolution of MODIS NDVI dataset obtained from NASA s Earth Observing System from 2000 to 2009 was 1 km 1 km. Monthly NDVI data were also obtained using the MVC method. MVC selects the highest observation for each pixel from a predefined compositing period to represent the current period. This approach is based on the logic that low-value observations are either erroneous or have less vegetation vigor for the period under consideration (Holben, 1986). The two datasets have the same time series from 2000 to We extracted the regional data that cover Northeast China using the ArcGIS-9.2 software. The data on vegetation type were digitized from the 1:1,000,000 vegetation map covering Northeast China mapped by the China Vegetation Editing Committee, Institute of Geography of the Chinese Academy of Sciences. Vegetation was classified into five types: forest, grassland, shrub, crop, and marsh. The distribution of different types of vegetation is shown in Fig. 2. The climatic datasets consisting of monthly mean temperature and monthly precipitation data were collected from the National Meteorological Center of China; the data include 95 meteorological stations during the period in Northeast China. The locations of the 95 meteorological stations are shown in Fig. 2. The

3 530 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) based on the monthly NDVI data of the two datasets from 2000 to The new NDVI was called expanded NDVI. To check the accuracy of the constructed long-term NDVI time series, several statistical and correlation analyses on the AVHRR GIMMS NDVI and the expanded NDVI were performed. The per-pixel unary linear regression model can acquire the most appropriate regression equation for each pixel. The elementary structure form of the model for different data is as follows (Ma, 2009): G i = a + bv i + ε i (1) n i=1 b = i V)(G i Ḡ) n i=1 (V i V) 2 a = Ḡ b V (2) Fig. 2. Vegetation map of Northeast China. spatial pattern of the annual mean temperature and precipitation of the study region is shown in Fig Methods The ERDAS 8.7 image processing and ArcGIS-9.2 software were used for data processing. The Albers Equal Area Conic Projection System and Beijing 1954 Coordinate System were used to integrate the different spatial data. To match the AVHRR GIMMS data, MODIS NDVI at 1 km 1 km resolution was resampled to 8 km 8 km resolution. The two datasets were obtained from different remote sensors and exhibited certain spectrum variance in response to vegetation. Thus, in the analysis of the long-term NDVI time-series based on AVHRR GIMMS NDVI and MODIS NDVI, checking the consistency of the two datasets was necessary (Fensholt et al., 2009). We constructed the monthly NDVI sequence, which covers , using the per-pixel unary linear regression model Parameters, a and b, are estimated by the least square method. ε i is random error. In this manuscript, G i represents GIMMS NDVI in ith month, meanwhile V i is MODIS NDVI (8 km 8 km) in ith month. Ḡ is mean of all monthly GIMMS data from 2000 to 2003 at corresponding pixel. V is mean of all monthly MODIS NDVI data (8 km 8 km) from 2000 to 2003 at corresponding pixel. In this study, the mean monthly NDVI, temperature, and precipitation of different months in the 28 years were calculated. To observe the correlations between monthly NDVI and climatic factors, we also discussed the trends of monthly NDVI and climate. Summer (June August) is the best season for vegetation growth, especially in Northeast China. Summer mean NDVI represents the best status of plant growth. Thus, we generated the spatially averaged time series of NDVI, mean temperature, and total precipitation over summers. The linear time trend was estimated by regressing it as a function of time over the study period. To further investigate the trends of yearly maximum NDVI, linear trends from 1982 to 2009 on a per-pixel basis were examined. Expression is shown in (3). Ā j means NDVI in j year: ( Slope = n n j=1 j n Ā j j) ( n j=1 j=1āj ( n n j=1 j2 n j) 2 j=1 To analyze the effects of regional climates on seasonal NDVI, correlations between NDVI and temperature and precipitation were determined for the four seasons. The analysis of the relationships between monthly NDVI and climatic parameters were performed based on stations. The process is described as follows. First, the NDVI time series data for each meteorological station were extracted from the mean of 3 3 ) (3) Fig. 3. Spatial pattern of air temperature and precipitation of Northeast China.

4 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) Fig. 4. Results of consistency check: (a) spatial mean data; (b) pixel value of stations; (c) station pixels statistics; (d) all pixels statistics. pixels around the location of the stations according to the geographical position using ArcGIS (Ding et al., 2007). Second, the correlation coefficient between NDVI and monthly mean temperature and monthly precipitation was calculated for each of the 95 stations. The time lag in the correlation between NDVI and temperature can be disregarded. Otherwise, a time lag of about 1 12 weeks exists in the correlation between NDVI and precipitation. To account for this interval and assess the real maximum correlation between NDVI and precipitation, the NDVI precipitation correlation coefficients were calculated using time lags of 0 3 months. The maximum value was then chosen as the NDVI precipitation correlation coefficient for each corresponding station (Nicholson et al., 1990; Li et al., 2002). Third, some spatial analyses and statistics were performed on the correlation coefficients. The vegetation type for each of the 95 stations was determined using the digital vegetation map of Northeast China and based on the station s latitude and longitude (with a precision of 0.01 ). Differences in NDVI temperature and NDVI precipitation correlations relative to vegetation type were also comprehensively investigated. 3. Results 3.1. Consistency check For the applications focusing on vegetation responses to climatic changes, the continuity and consistency of the two NDVI datasets are essential (Steven et al., 2003). The spectral and spatial resolution of MODIS NDVI is higher than that of AVHRR GIMMS NDVI. The spectral characteristics of AVHRR NDVI and its differences from MODIS NDVI are shown in Table 1. The precision evaluation of the expanded NDVI from 2000 to 2009 is indispensable. The contrasts of the GIMMS NDVI and the expanded NDVI from 2000 to 2003 in the monthly spatially averaged NDVI and NDVI extracted by climate stations are shown in Fig. 4a and b, respectively. Relative error was for averaged monthly data and for station value. The difference values for the stations and all pixels are shown in Fig. 4c and d, respectively. The difference value between ±0.05 in all pixels was 98.5%. The relative deviation of all pixels was within 10%. The correlation coefficient of GIMMS NDVI and expanded GIMMS NDVI was (p < 0.001) NDVI changes and correlation analyses Trends of monthly NDVI and climatic variables The magnitude of the monthly NDVI and its change over time are important indicators of the contribution of vegetation activity in different months to total annual plant growth (Piao et al., 2003). In Northeast China, the mean monthly NDVI reached maximum values in August. From December to April, the mean monthly NDVI was low (Fig. 5a). The mean NDVI value in August was In the 28 years, the trends of monthly NDVI showed positive values except in June, July, and October. Significant differences were observed from May to October. The maximum trend value was observed in September, whereas the minimum was observed in October. The NDVI trends and their patterns were coupled with those of climatic variables (Fig. 5b and c). Monthly temperature and precipitation showed a pattern similar to that of NDVI: a higher value was observed in summer and a lower value in winter. However, the Table 1 Geometrical and spectrum characteristic of GIMMS and MODIS dataset sensors. Sensor R/ m N IR/ m Spatial resolution/km NDVI compound days/d AVHRR MODIS

5 532 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) Fig. 5. Seasonal changes in monthly NDVI and climatic variables from 1982 to 2009 in Northeast China. (a) 28-Year averaged monthly NDVI and its trend; (b) 28-year averaged monthly mean temperature and its trend; (c) 28-year averaged monthly precipitation and its trend; (d) correlations between monthly NDVI and temperature and precipitation. highest mean temperature and precipitation was in July instead of August, the month with the largest mean NDVI. The monthly temperature trends were positive from January to October over the study period, indicating that the temperature in spring and plant growing season increased. The largest temperature rise occurred in February with an annual rate of 0.1 C, indicating that temperature in this month increased by 2.8 C relative to The monthly precipitation showed negative trends in June, July, and August. The largest precipitation decrease occurred in August, with an annual rate of 1.57 mm, indicating that precipitation during this month decreased by 44 mm from 1982 to The correlation between monthly climatic variables and NDVI is shown in Fig. 5d. A positive correlation between temperature and NDVI was observed for most of months. NDVI significantly correlated with temperature (5% level) only in spring. This confirms that increasing temperature improves vegetation growth in spring in the whole region. The relationship between precipitation and NDVI was complicated: significantly negative correlations were observed for August, September, and November, no significantly negative correlations occurred for January, February, and October, and positive correlations were discovered for other months. whereas precipitation had a significant downward trend (Fig. 6). In terms of coarse scale, no significant correlations between summer NDVI and temperature or precipitation were found over the 28 years. However, the combined influence on NDVI was evident from the variation curves. NDVI increased when temperature and precipitation rose, such as that observed in 1984 and NDVI decreased with reduced temperature and precipitation in When trends of temperature and precipitation differed, the NDVI of a given year exhibited a minimal variation Trends of spatially averaged summer NDVI and climatic parameters Summer is the best season for vegetation growth, especially in Northeast China because of relatively high temperature and strong precipitation. Considering seasonal feature of crops, we investigated the summer NDVI variation in the 28 years. The spatially averaged NDVI exhibited a downward trend but no significant change (p = 0.2). The temperature exhibited an upward trend, Fig. 6. Trends in the mean NDVI and mean temperature and total precipitation in summer from 1982 to 2009.

6 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) and temperature was observed in the Lesser Khingan and Changbai Mountains. A negative correlation between NDVI and precipitation was found mainly in the northern part of the Songnen Plain. In winter, significant negative correlation between NDVI and temperature and precipitation was keenly observed in northern part of the study area (Fig. 8g and h). Grasslands in Hulun Buir showed no vegetation because of snow cover Correlation analyses of monthly NDVI and climatic parameters based on meteorological stations Fig. 7. Spatial pattern of NDVI trends from 1982 to Linear trends of NDVI from 1982 to 2009 on a per-pixel basis Although the trends in monthly NDVI were complicated and exhibited no significantly downward trend in summer, we found a high degree of spatial heterogeneity on per-pixel analysis (Fig. 7). The pixels that showed moderate and remarkable increasing trends were mainly distributed in the three plains: the Sanjiang, Songnen, and Liaohe Plains. The increasing trend was higher than 0.003/year. The dominant vegetation type in these plains is crops. The pixels showing light increasing trends were mainly in the arid regions of the IMAR and around the plains. Grassland and marsh are main vegetation types in these areas. The areas that showed decreasing NDVI showed a changing trend below 0.003, observed mainly in forest dominated regions, including the Greater and Lesser Khingan Mountains, as well as the Changbai Mountains Spatial pattern of the correlations between seasonal NDVI and climatic variables To analyze effects of regional climatic changes on seasonal NDVI, correlations between NDVI and temperature and precipitation were explored for four seasons. In spring, NDVI was positively correlated with temperature in most areas (Fig. 8a), suggesting the possible effects of temperature in spring NDVI for most parts of Northeast China. A negative correlation between NDVI and temperature was found in the Songnen Plain and arid grasslands in the Hulun Buir Plateau. There was a weak positive correlation between NDVI and precipitation over most areas in spring (Fig. 8b), but a negative correlation was observed in the Lesser Khingan Mountains and Changbai Mountains. These findings are in agreement with those obtained by Piao et al. (2004). In summer, a negative correlation between NDVI and temperature was observed in arid areas of East IMAR and most areas of the Songnen Plain; positive correlations were observed between NDVI and temperature in other areas of the study region (Fig. 8c). A significantly positive correlation between NDVI and precipitation was found in the Hulun Buir grasslands, and negative correlation between NDVI and precipitation in forest areas (Fig. 8d). In autumn (Fig. 8e and f), a weak positive correlation between NDVI and temperature and precipitation was detected in most areas of the study region. A negative correlation between NDVI The calculated correlation coefficients between monthly NDVI and temperature (C NT ) indicated that for all meteorological stations in Northeast China, a highly significant correlation between NDVI and temperature was detected. For all stations, the correlation was highly significant at the 0.01 level. Among these stations, Baoguotu station showed minimum correlation, with a C NT of over the 28 year period. The maximum value was in Jian station. The average C NT for the 95 meteorological stations was The correlation coefficients between monthly NDVI and precipitation (C NP ) showed a significant correlation at the 0.01 level for all meteorological stations. Precipitation is the most important source of soil moisture; NDVI was closely linked with precipitation in temperate zones. In this study, most C NP values were greater than 0.7 (71 out of the 95 stations). The mean C NP for all stations was Anda station showed the maximum C NP at The minimum C NP observed in Kuandian station was Discussion 4.1. Integrated application of two remote sensing datasets Given the limitations presented by a single NDVI dataset in time sequence, the integrated application of two remote datasets is important. In this paper, the two datasets were obtained from different sensors. The difference value between MODIS NDVI and AVHRR NDVI in most pixels was about 0.1. In the regional scale, Xin et al. (2007) extended the regional averaged AVHRR NDVI in the Loess Plateau of China, using SPOT-VGT NDVI by unary linear regression equation to study annual variation trends. They also performed construction on pixel extent. Although they determined that accuracy in pixel integration could not be used in analysis, they provided a good idea for constructing NDVI time series and conducting consistency checks. With regard to pixel extent, Brown et al. (2008) constructed continuous NDVI time series from AVHRR and MODIS data using neural networks. They obtained accurate results, in which many influencing factors were taken into consideration. On the basis of these previous studies, this paper constructed NDVI time series using per-pixel unary linear regression equation and conducted a consistency check. Using this method, the pixels generated minimal exceptional values, which can be disregarded. According to the results, the constructed NDVI time series in this study passed the consistency check. Therefore, it can be used in trend and correlation analyses. The integrated application of the two different remote sensing datasets is an effective method for use in the correlation analysis of vegetation and climatic variables Comparison of NDVI change trends with other NDVI studies The monthly NDVI showed obvious variations during the growing season (April October). Monthly climatic changes and trends in the 28 years provided a clear illustration of the NDVI trends. Piao et al. (2003) obtained national-scale results for the monthly NDVI changes and trends from 1982 to They found that the maximum NDVI occurred in August; however, monthly NDVI showed

7 534 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) Fig. 8. Spatial pattern of correlation between NDVI and temperature and precipitation for spring (March May, a and b), summer (June October, c and d), autumn (September November, e and f), and winter (December February, g and h). positive values for all months. The difference in their results from ours is attributed to the longer study period and the relatively small scale of the present research. The unchanged increase in NDVI in August and September was attributed to the influence of time lag, the widely distributed river, and the relatively adequate precipitation for vegetation growth in Northeast China (Song and Ma, 2008). The increased temperature antedated the vegetation growing season (Piao et al., 2004), which yielded the interpretation of the positive NDVI trend in spring. Guo et al. (2008) found slowly decreasing NDVI trend by investigating NDVI changes in Northeast China from 1982 to Differently, results of this paper had obviously change trends pattern. More obviously special heterogeneity in accordance with terrain showed during Correlation between NDVI and climatic variables Negative correlation between NDVI and temperature in spring and winter was attributed to low temperature, which stemmed at least in part from the snow-covered and frozen ground. These areas had low vegetation under snow or no vegetation at all because of the low temperature. The region of the Lesser Khingan and Changbai Mountains showed negative correlation between NDVI and precipitation because spring is cold and snowy in these areas. Precipitation causes more clouds to appear and then reduces incident radiation. Limited incident radiation hinders photosynthesis, which is essential for vegetation growth (Song and Ma, 2008). In summer, the negative correlation between NDVI and

8 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) temperature in arid areas of East IMAR and most areas of the Songnen Plain can be explained by the limited vegetation growth caused by increased temperature and fewer precipitations. Increasing temperature enhanced the intensity of transpiration, and the decreasing precipitation reduced the available moisture for plants (Yang et al., 2009). Nevertheless, positive correlations between NDVI and temperature in other areas of the study region resulted from the increased temperature, which enhanced photosynthesis and respiration for plant growth. The significantly positive correlation between NDVI and precipitation in the Hulun Buir grasslands was caused by the precipitation which provided the required moisture for vegetation growth. Precipitation in summer benefited vegetation growth in arid areas. In forest dominated areas, where abundant summer rainfall occurs, a negative correlation between NDVI and precipitation was caused possibly by the increased precipitation accompanied by an increase in cloud cover. Thus, a reduction in incoming solar radiation also occurred. There had obvious seasonal fluctuation in Northeast China driven by climatic changes. Comparing NDVI changes in the four seasons, in summer, vegetation NDVI had the largest change and the correlation between NDVI and climatic factors was remarkable Effects of vegetation type on C NT and C NP To avoid the influence of integrated raster climatic data on the correlation of various elevations, station-based analyses were performed. The correlation analysis for monthly NDVI and climatic parameters for stations showed that C NT was larger than the C NP. For 92.6% of all stations (88 of the 95), C NT was higher than C NP. The average C NT and C NP were and 0.729, respectively. The effect of temperature on vegetation NDVI was stronger than that of precipitation in the entire region. This result was comparable with conclusions drawn by Wang et al. (2002), who used pixel statistics for Northeast China. Those stations where C NT was lower than C NP are distributed mainly in West Jilin Province and the arid areas of IMAR. As shown in Fig. 3, the main vegetation type in these areas is grassland. Less precipitation occurs in these areas, with a value below 300 mm. As the major source of soil moisture, precipitation plays an important role for the vegetation growth of this sub-region. The increased temperature accelerates evapotranspiration, necessitating more soil moisture for vegetation in arid and semi-arid areas (Song and Ma, 2008). Northeast China is characterized by diverse vegetation types. We classified vegetation into five types (Table 2). Forest comprises deciduous broad-leaved, mixed, and coniferous forests. Grassland comprises typical grassland, meadow steppe, and meadow. Marsh is composed of marshy grassland and marsh. Forest is the major vegetation type accounting for 38.14% of Northeast China. To investigate the influence of vegetation type on C NT and C NP, statistical analyses were performed for the five vegetation types (Fig. 9). Results showed that for all vegetation types, average C NT was higher than average C NP. C NT was significantly higher than C NP for forest and marsh. The difference between C NT and C NP for grassland was Fig. 9. Comparison of C NT and C NP for different vegetation types. Table 2 Vegetation type in Northeast China. Vegetation type Area (km 2 ) Percentage of total Forest 468, % Shrub 55, % Grassland 245, % Crop 359, % Marsh 99, % Total 1,228, % the smallest. Grassland is widely distributed over the western part of the research area in IMAR, where climate is dry. Temperature and precipitation are both important for the growth of grassland vegetation. Different vegetation types showed various NDVI climate correlations. Additional statistical results are shown in Table 3. These indicated that C NT values increased as one moves from grassland to crop, forest, wetland, and shrub, with mean C NT values of 0.826, 0.857, 0.892, 0.899, and 0.910, respectively. C NP values decreased as one moves from shrub to forest, wetland, grassland, and crop, with mean C NP values of 0.771, 0.738, 0.737, 0.727, and 0.721, respectively. Climate type is clearly an important factor determining vegetation type for each sub-region. At the same time, vegetation types play an important role in C NT and C NP. We can obtain the spatial pattern of correlation based on the different vegetation types. Broad leaves and low hills for the shrub attributed to C NT and C NP were superior. Shrub growth was highly correlated with climate change. The adequate moisture in marsh and its distribution, mainly spreading in the northern part of the study area where the temperature is low, had caused vegetation to lean toward a higher C NT. The C NP of crop mainly distributed in the three plains was the minimum, resulting from the fact that the growing period for annual crops is in summer when these regions have more precipitation and agricultural activities abound. Because of the warmer Table 3 NDVI temperature and NDVI precipitation correlation coefficient for vegetation type. Vegetation type NDVI temperature correlation coefficient NDVI precipitation correlation coefficient Location of stations Number of stations Mean Max. Min. St. deviation Mean Max. Min. St. deviation Forest Greater Khingan Mountains, Lesser 13 Khingan Mountains, and Changbai Mountains Shrub In the southern part of study area 3 Grassland In the western part of study area 12 Crop The Sanjiang Plain, the Songnen 59 Plain, and the Liaohe Plain Marsh In the northern part of study area 8

9 536 D. Mao et al. / International Journal of Applied Earth Observation and Geoinformation 18 (2012) temperature, permafrost degradation (Ding, 1998; Jin et al., 2000, 2007), and adequate precipitation in Northeast China, as well as the addition of well-developed root systems that can hold a substantial amount of moisture that can be gradually released over time (Li et al., 2002), precipitation in forest areas was not the limiting factor. Low C NP was observed mainly in the Changbai Mountains in Liaoning Province, where the major vegetation type is deciduous broad-leaved forest and deciduous shrub. Surface runoff in the hilly area, as well as the transpiration and respiration of deciduous broad-leaved and deciduous shrub, is strong. The required water supply for vegetation growth comes from rainfall, snow, and so on. 5. Conclusion This paper constructed the monthly NDVI sequences of Northeast China covering using the per-pixel unary linear regression model based on the monthly NDVI data of MODIS and AVHRR datasets. The data were comprehensively used in the analyses of the correlation between NDVI and climatic parameters. This paper confirmed the feasibility of long-term NDVI time seriesclimate research. In the 28 year period, monthly NDVI was closely correlated with climate variations. NDVI trends were spatially heterogeneous, corresponding with regional climatic characteristics for different seasons. Spatially averaged NDVI in summer exhibited a downward trend in , with increased temperature and significantly decreased precipitation. Yearly maximum NDVI trend reflected an obvious spatial pattern for different regions and different vegetation types. The NDVI of all meteorological stations had significant correlations with monthly mean temperature and monthly precipitation from 1982 to The correlation coefficient between NDVI and temperature was larger than that between NDVI and precipitation in most of stations and for all vegetation types. Different vegetation types showed various spatial responses to climatic changes. Acknowledgments This paper was supported by the Strategic Frontier Program of the Chinese Academy of Sciences (Grant No. XDA ), the Knowledge Innovation Program of the Chinese Academy of Sciences (No. KZCX2-YW-341) and the National Natural Science Foundation of China (No ). We thank the two anonymous reviewers for their comments that resulted in a greatly improved manuscript. References Beck, P.S.A., Atzberger, C., Høgda, K.A., Improved monitoring of vegetation dynamics at very high latitudes, a new method using MODIS NDVI. Remote Sensing of Environment 100, Brown, M.E., Pinzon, J.E., Didan, K., Morisette, J.T., Tucker, C.J., Evaluation of the consistency of long-term NDVI time series derived from AVHRR, SPOTvegetation, SeaWIFS, MODIS, and LandSAT ETM+ sensors. IEEE Transactions on Geosciences and Remote Sensing 44 (1-7), Brown, M.E., Lary, D.J., Vrieling, A., Stathakis, D., Mussa, H., Neural networks as a tool for constructing continuous NDVI time series from AVHRR and MODIS. International Journal of Remote Sensing 29 (24), Ding, Y.J., Recent degradation of permafrost in China and the response to climatic warming. In: Permaforst-Seventh International Conference (Proceedings), Yellowknife, Canada, p. 55. Ding, M.J., Zhang, Y.L., Liu, L.S., Zhang, W., Wang, Z.F., Bai, W.Q., The relationship between NDVI and precipitation on the Tibetan Plateau. Journal of Geophysical Sciences 17 (3), Ding, M.J., Zhang, Y.L., Liu, L.S., Wang, Z.F., Temporal and spatial distribution of grassland coverage change in Tibetan Plateau since Journal of Natural Resources 25 (12), (in Chinese). Fang, J.Y., Piao, S.L., He, J.S., Vegetation activity is strengthened during the Past 22 Years in China. Science in China Series C 33 (6), (in Chinese). Fensholt, R., Rasmussen, K., Nielsen, T.T., Mbow, C., Evaluation of earth observation based long term vegetation trends intercomparing NDVI time series trend analysis consistency of Sahel from AVHRR GIMMS, Terra MODIS and SPOT VGT data. Remote Sensing of Environment 113, Gitelson, A.A., Kaufman, Y.J., MODDIS NDVI optimization to fit the AVHRR data series-spectral considerations. Remote Sensing of Environment 66, Guo, Z.X., Wang, Z.M., Song, K.S., Li, F., Zhang, B., Liu, D.W., Changes of vegetation coverage in Northeast China from 1982 to Acta Botanica Boreali-Occidentalia Sinica 28 (1), (in Chinese). Gong, D.Y., Shi, P.J., Northern hemispheric NDVI variations associated with large-scale climate indices in spring. International Journal of Remote Sensing 24 (12), Holben, B.N., Characteristics of maximum-value composite images from temporal AVHRR data. International Journal of Remote Sensing 7, Ichii, K., Kawabata, A., Yamaguchi, Y., Global correlation analysis for NDVI and climatic variables and NDVI trends, International Journal of Remote Sensing 23 (18), Jin, H.J., Li, S.X., Cheng, G.D., Wang, S.L., Li, X., Permafrost and climatic change in China. Global and Planetary Change 26, Jin, H.J., Yu, Q.H., Lv, L.Z., Guo, D.X., He, R.X., Yu, S.P., Sun, G.Y., Li, Y.W., Degradation of permafrost in the Xing anling Mountains, Northeastern China. Permafrost and Periglacial Processes 18, Li, B., Tao, S., Dawson, R.W., Relations between AVHRR NDVI and eco-climatic parameters in China. International Journal of Remote Sensing 23 (5), Ma, M.G., Research of Monitoring and Simulation of Vegetation Dynamics in Northwest China, Based on Time Series Data. Academic Exchanges of Young Scholars in the Field of Geo-Information Science and Technology, Beijing, p. 10 (in Chinese). Morisette, J.T., Pinzon, J.E., Brown, M.E., Tucker, C.J., Justice, C.O., Initial validation of NDVI time series from AVHRR, vegetation and MODIS. In: Int. SPOT 4/5 VEGETATION Users Conf., Antwerp, Belgium. Myneni, R.B., Keeling, C.D., Tucker, C.J., Aarar, G., Nemani, R.R., Increased plant growth in the northern high latitudes from 1981 to Nature 386 (17), Nicholson, S.E., Davenport, M.L., Malo, A.R., A comparison of the vegetation responses to rainfall in the Sahel and east Africa, using normalized difference vegetation index from NOAA AVHRR. Climate Change 17, Piao, S.L., Fang, J.Y., Zhou, L.M., Guo, Q.H., Mark, H., Ji, W., Li, Y., Tao, S., Interannual variations of monthly and seasonal normalized difference vegetation index (NDVI) in China from 1982 to Journal of Geophysical Research 108 (14), Piao, S.L., Fang, J.Y., Ji, W., Guo, Q.H., Ke, J.H., Tao, S., Variation in a satellitebased vegetation index in relation to climate in China. Journal of Vegetation Science 15, Qi, Y., The effect of climate change on vegetation at high latitudes of the northern Hemisphere, a functional analysis. Acta Ecologica Sinica 19 (4), (in Chinese). Qiu, H.J., Cao, M.M., Spatial and temporal variation in vegetation cover in China based on SPOT vegetation data. Resources Science 33 (2), (in Chinese). Slayback, D.A., Pinzon, J., Los, S.O., Tucker, C.J., Northern hemisphere photosynthetic trends Global Change Biology 9, Song, Y., Ma, M.G., Variation of AVHRR NDVI and its relationship with climate in Chinese arid and cold regions. Journal of Remote Sensing 12 (3), (in Chinese). Steven, M.D., Malthus, T.J., Baret, F., Xu, H., Chopping, M.J., Intercalibration of vegetation indices from different sensor systems. Remote Sensing of Environment 88, Tucker, C.J., Slayback, D.A., Pinzon, J.E., Los, S.Q., Myneni, R.B., Taylor, M.G., Higher northern latitude normalized difference vegetation index and growing season trends from 1982 to International Journal of Biometeorology 45, Tucker, C.J., Pinzon, J.E., Brown, M.E., Slayback, D.A., Pak, E.W., Mahoney, R., Vermote, E.F., Saleous, N.E., An extended AVHRR 8-km NDVI dataset compatible with MODIS and SPOT vegetation NDVI data. International Journal of Remote Sensing 26 (20), Wang, J., Rich, P.M., Price, K.P., Temporal responses of NDVI to precipitation and temperature in the central Great Plains, USA. International Journal of Remote Sensing 11 (24), Wang, S.Q., Tian, H.Q., Liu, J.Y., Zhuang, D.F., Zhang, S.W., Hu, W.Y., Characterization of changes in land cover and carbon storage in Northeastern China, an analysis based on Landsat TM data. Science in China Series C 45 (Suppl.), Wang, Z.M., Guo, Z.X., Song, K.S., Luo, L., Zhang, B., Liu, D.W., Huang, N., Ren, C.Y., Responses of vegetation NDVI in Northeast China to climate change. Chinese Journl of Ecology 28 (6), (in Chinese). Xin, Z.B., Xu, J.X., Zheng, W., The influences of climate change and human activity on vegetation cover change in Loess Plateau. Science in China Series D 37 (11), (in Chinese). Yang, G.H., Bao, A.M., Chen, X., Liu, H.L., Huang, Y., Dai, S.Y., Study of the vegetation cover change and its driving factors over Xinjiang during Journal of Glaciology and Geocryology 31 (3), (in Chinese). Zhang, J., Ge, J.P., Guo, Q.X., The relation between the change of NDVI of the main vegetation types and the climatic factors in the Northeast of China. Acta Ecologica Sinica 21 (4), (in Chinese). Zhou, L., Tucker, C.J., Kaufmann, R.K., Slayback, D.A., Shabanov, N.V., Myneni, R.B., Variations in northern vegetation activity inferred from satellite data of vegetation index during 1981 to Journal of Geophysical Research 106 (17),

10 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具