為了瞭解人們在我們所居住地球上的影響，估測區域尺度和全球尺度的主要生產量是很關鍵的部份。而美國太空總署（National Aeronautics and Space Administration, NASA）中的中解析度成像分光輻射度計(Moderate-resolution Imaging Spectroradiometer, MODIS)科學團隊提供一個一致性且方便的資料來源，以觀測全球生物圈和動態變化的過程。總初級生產量（Gross primary productivity, GPP）和淨初級生產量（Net primary productivity, NPP）是使用整合的方法去結合MODIS資料、氣象資料及生態模型作定期的測量。 本研究中，分析各種不同的地貌特徵對於GPP的影響，分析中包涵土地覆蓋類別、高度海拔、緯度和植被指數。說明GPP和一些現象的初步相關分析，也使用氣象資料針對GPP的時序變化作分析探討。從MOD17 GPP與氣候資料的成果可知， 2006到2009年太陽輻射的增加是影響GPP最主要的因素，其次是溫度和降雨，呈現出的結果也與方程式一致，但乾旱對GPP的影響卻不明顯。7個氣候分區中，GPP和PSN在春天和冬天時期才有較大的差異。GPP的年累積量是以SE區最大，SW區和CW區次之，EN區最小。 地貌特徵的分析中，影響GPP的主要因素是植被物種的分類，而緯度的劃分與太陽輻射相關，但緯度越低，GPP值不一定就越高。GPP與海拔高度無高度的相關性。土壤表面溫度在夏季時，各地溫差比較小，且GPP數值分佈範圍較大。最後以通量塔台和MOD17的資料觀察其時序變化和相關性，瞭解MODIS所估算的PSNnet和塔台NEE的現地觀測狀況;而Terra/Aqua-MODIS MOD17 的產品所估算之GPP 及PSNnet 可大致反應實地碳匯的變化狀況。Assessment of regional and global-scale primary productivity is crucial for the understanding of human impacts on our living earth. NASA’s MODIS science team has provided a consistent and convenient data source for monitoring the global biosphere and dynamic processes. Gross primary productivity (GPP) and net primary productivity (NPP) have been routinely measured by an integrated method which incorporates MODIS data, climatological data, and ecosystem modeling. In this study, GPP is analyzed with various terrain features, including land cover, terrain height, terrain latitude and vegetation index to address the variability and effects on GPP estimation. Preliminary correlation analysis between GPP and phenomenon was explained. We used Meteorological Information to discuss GPP’s variation of time series. The results show it that increase of solar radiation is the major factor to effect GPP from 2006 to 2009. The next is temperature and precipitation. The result is consistent with the equation. But the effects of drought for GPP is not clear. In the seven climate division, GPP and PSN have a greater difference in the spring and winter. The cumulative amount of GPP is the largest value in the zone of SE. The next is SW and CW. EN is the smallest. For the analysis of landscape features, the classification of vegetation species is the major factor to effect GPP. The division of latitude and the amount of solar radiation are related. When latitude is on the decrease, GPP does not always increase. GPP and the altitude are not very highly correlated. When land surface temperature is in the summer, the range of temperature variation is relatively small in everyplace. The distribution range of GPP value is large in the summer. In the end we examined diversification of time series and correlation with the data of flux tower and MOD17. We will understand situation of MODIS’s PSNnet and NEE of tower’s observation in field. And production of Terra/Aqua-MODIS MOD17 estimates GPP and PSNnet can reply more or less changing condition of carbon sequestration in field.
