Applications of MODIS-visible bands index, greenery ratio to estimate CO2 budget of a rice paddy in Japan

Abstract

NDVI (Normalized Difference of Vegetation Index) was employed to estimate the carbon budget remotely with satellite data. However, NDVI has some difficulties in application to agricultural crops, boreal forest, and tundra ecosystems. We proposed a new vegetation index GR (greenery ratio) to detect the vegetation change remotely, and we applied it to estimate CO2 budget of Japanese rice paddy with MODIS satellite data. GR was ratio of green (G) to the trichromatic visible bands (R G B) of MODIS, and an empirical GR-CO2 budget model was developed as functions of MODIS-data and observed micrometeorology and fluxes at the rice paddy of Mase-site. The daily PAR (photosynthetically active radiation) was also estimated by MODIS. The parameterized model provided good performance to estimate daily magnitudes and seasonal trends of GPP (gross primary productivity), however, RE (ecosystem respiration) showed a little under-estimation, especially differed in late growing season. In contradiction to the observed CO2 budget, the estimated budgets were 2 greater of GPP and 5 less of RE and 9 greater of NEE (net ecosystem exchange). The large discrepancy in NEE was owing to the poor estimation of RE after drainage. Further study to improve RE estimation was needed.

Advantages of visible-band spectral remote sensing at both satellite and near-surface scales for monitoring the seasonal dynamics of GPP in a Japanese larch forest

“…First, the seasonal variations in MODIS VIS_indices showed strong linear correlations with those in GPP, indicating agreement of both seasonal and interannual variations. The results supported a strong correlation between photosynthetic parameters and GR reported by Harazono et al (2009). VIS_indices are sensitive to green leaf color, so they responded to photosynthetic pigments.…”

“…GR is defined as the ratio of green to the total digital numbers (R G B) of camera images (Table 1), and it can effectively detect phenological events (Ahrends et al, 2008;Maeda et al, 2008). Similarly, GR is used as the green reflectance ratio to total visible reflectance of MODIS data to estimate GPP (Harazono et al, 2009). According to the definition, GR≈1/3 indicates white or black targets such as snow, ice or soil.…”

“…However, inconsistencies between MODIS-derived NDVI or EVI results and vegetation phenology or flux-based GPP data derived from ground observations have been reported (Xiao et al, 2004b;Gitelson et al, 2006;Jenkins et al, 2007;Nagai et al, 2010). As an alternative, recent studies have suggested that newly developed visible-band (VIS) vegetation indices: the MODIS-derived greenery ratio (GR) data have been used for effective estimation of GPP in a rice paddy (Harazono et al, 2009), and the green-red vegetation index (GRVI) results derived from ground observations have been utilized to detect phenological changes in several types of ecosystems (Motohka et al, 2010).…”

Determination of rice paddy parameters in the global gross primary production capacity estimation algorithm using 6 years of JP-MSE flux observation data

“…However, when using other VIs, GPP occasionally does not show a clear linear relationship with the products of VIs and PAR. The rate of photosynthesis and the light intensity is often nonlinear for canopies Gu et al, 2002;Saito et al, 2005 . Several studies have used GPP calculation methods with a light -response curve Furumi et al, 2005;Harazono et al, 2009;Ide et al, 2010 using satellite sensor data. However, most of them did not include the concept of the "photosynthesis capacity".…”

Estimating daily gross primary production of maize based only on MODIS WDRVI and shortwave radiation data

“…The vast majority of remote-sensing based GPP models used vegetation index (VI) data as a proxy variable for fPAR, which was assumed to have close relationship with vegetation community structure. Examples of VIs are the normalized difference vegetation index, NDVI (Rouse, 1974;Ruimy et al, 1994;Running et al, 2000;Tucker, 1979), the enhanced vegetation index, EVI (Huete et al, 2002;Xiao et al, 2005;Yan et al, 2009), the greenery ratio, GR (Harazono et al, 2009), the chlorophyll index, Chl index (Gitelson et al, 2006, and the MODIS leaf area index (LAI) product (Yuan et al, 2010). Gamon et al (1992) presented photochemical reflectance index (PRI) and found that it was correlated with the LUE of a sunflower canopy under nitrogen-stressed condition, but not under water-stressed condition.…”