Fig. 2: Variations in \({\gamma }_{{{{\rm{CGR}}}}}^{{{{\rm{T}}}}}\) with varying dryness conditions. | Nature Communications

Fig. 2: Variations in \({\gamma }_{{{{\rm{CGR}}}}}^{{{{\rm{T}}}}}\) with varying dryness conditions.

From: Recent decrease of the impact of tropical temperature on the carbon cycle linked to increased precipitation

Fig. 2

a Variations in long-term anomalies of detrended precipitation, the self-calibrating Palmer Drought Severity Index (scPDSI) and Standardized Precipitation and Evaporation Index (SPEI) calculated from a 12-month moving window (n = 721). Water conditions are derived based on the standardized deviation (σ) of long-term detrended precipitation that are extracted from both CRU and GPCC datasets, scPDSI and SPEI. The long time series are divided into four categories, denoting very wet (σ ≥ 1), wet (0 ≤ σ < 1), dry (−1 ≤ σ < 0) and very dry (σ < −1) conditions based on the anomalies of detrended precipitation, scPDSI and SPEI, which are shown in different background colors. b Average \({{{{\rm{\gamma }}}}}_{{{{\rm{CGR}}}}}^{{{{\rm{T}}}}}\) for each category (error bars denote 1 SD of \({{{{\rm{\gamma }}}}}_{{{{\rm{CGR}}}}}^{{{{\rm{T}}}}}\) in 500 bootstrap estimates). c Variations in the anomalies of bi-decadal precipitation (high frequency precipitation averaged per 20-y moving window). d Correlation coefficients between \({{{{\rm{\gamma }}}}}_{{{{\rm{CGR}}}}}^{{{{\rm{T}}}}}\) and bi-decadal precipitation. M1 and M2 denoted that \({{{{\rm{\gamma }}}}}_{{{{\rm{CGR}}}}}^{{{{\rm{T}}}}}\) are calculated based on Eqs. 1 and 2 (error bars denoted 1 SD of the correlation coefficient in 500 bootstrap estimates).

Back to article page