Validating the accuracy of land cover products using a reliable reference dataset is an important task. Recently, the amount of ground truth data provided by volunteers has increased. Although ground truth data can provide information that can produce reliable reference data, the information is only correct within the observable landscape. It is necessary to scale up reference data derived from ground truth data to match the spatial resolution of the global land cover product. We propose a scale-up method that confirms expanding land cover characteristics for a target position using the occurrence ratio of pixels that meet the criteria for the target position to the total pixels for the target scale. The results of applying the scale-up method to test sites showed that the occurrence ratio method was a better judge of expanding target land cover types than the average method.
Light use efficiency (LUE) is a key parameter in estimating gross primary production (GPP) based on global Earth-observation satellite data and model calculations. In current LUE-based GPP estimation models, the maximum LUE is treated as a constant for each biome type. However, the maximum LUE varies seasonally. In this study, seasonal maximum LUE values were estimated from the maximum incident LUE versus the incident photosynthetically active radiation (PAR) and the fraction of absorbed PAR. First, an algorithm to estimate maximum incident LUE was developed to estimate GPP capacity using a light response curve. One of the parameters required for the light response curve was estimated from the linear relationship of the chlorophyll index and the GPP capacity at a high PAR level of 2000 (µmolm<sup>-2</sup>s<sup>-1</sup>), and was referred to as“ the maximum GPP capacity at 2000". The relationship was determined for six plant functional types: needleleaf deciduous trees, broadleaf deciduous trees, needleleaf evergreen trees, broadleaf evergreen trees, C3 grass, and crops. The maximum LUE values estimated in this study displayed seasonal variation, especially those for deciduous broadleaf forest, but also those for evergreen needleleaf forest.
To estimate gross primary production (GPP), the process of photosynthesis was considered as two separate
phases: capacity and reduction. The reduction phase is influenced by environmental conditions such as soil
moisture and weather conditions such as vapor pressure differences. For a particular leaf, photosynthetic capacity
mainly depends on the amount of chlorophyll and the RuBisCO enzyme. The chlorophyll content can be
estimated by the color of the leaf, and leaf color can be detected by optical sensors. We used the chlorophyll
content of leaves to estimate the level of GPP.
A previously developed framework for GPP capacity estimation employs a chlorophyll index. The index is
based on the linear relationship between the chlorophyll content of a leaf and the maximum photosynthesis at
PAR =2000 (<i>μmolm </i><sup>-2</sup><i>s</i><sup>-1</sup>) on a light-response curve under low stress conditions.
As a first step, this study examined the global distribution of the index and found that regions with high
chlorophyll index values in winter corresponded to tropical rainforest areas. The seasonal changes in the chlorophyll
index differed from those shown by the normalized difference vegetation index. Next, the capacity of GPP
was estimated from the light-response curve using the index. Most regions exhibited a higher GPP capacity than
that estimated from Moderate Resolution Imaging Spectroradiometer (MODIS) observations, except in areas of
tropical rainforest, where the GPP capacity and the MODIS GPP estimates were almost identical.
Global land cover data sets are required for the study of global environmental changes such as global biogeochemical
cycles and climate change, and for the estimation of gross primary production. To determine land cover classification
condition, producers examine the phenological feature of each land cover class’s sample area with vegetation indices or
only reflectance. In this study, to detect the phenological feature of land surfaces, we use the universal pattern
decomposition method (UPDM) three coefficients and two indices; the modified vegetation index based on the UPDM
(MVIUPD) and the chlorophyll index (CI<sub>green</sub>). The UPDM three coefficients are corresponded to actual objects; water,
vegetation and soil. To detect the phenological feature of each land cover class simply, we use annual statistical values of
the UPDM coefficients and two indices. By visualizing three statistical values with combination of RGB, land areas with
similar phenological feature are able to detect globally. We produced the global land cover products by applying this
method with MODIS Aqua Surface Reflectance 8-Day L3 Global 500m data sets of 2007. The result was roughly similar
to the MOD12Q1 of the same year.