Competing influences of anthropogenic warming, ENSO, and plant physiology on future terrestrial aridity

Abstract

The 2011-2016 Californian drought illustrates that drought-prone areas do not always experience relief once a favorable phase of El Niño-Southern Oscillation (ENSO) returns. In the 21^st century, such an expectation is unrealistic in regions where global warming induces an increase in terrestrial aridity larger than the aridity changes driven by ENSO variability. This premise is also flawed in areas where precipitation supply cannot offset the global warming-induced increased evaporative demand. Here, atmosphere-only experiments are analyzed to identify land regions in which aridity is currently sensitive to ENSO, and where projected future changes in mean aridity exceed the range caused by ENSO variability. Insights into the drivers of these aridity changes are obtained in simulations with incremental addition of three different factors to current climate: ocean warming, vegetation response to elevated CO₂ levels, and intensified CO₂ radiative forcing. The effect of ocean warming overwhelms the range of ENSO-driven temperature variability worldwide, increasing potential evapotranspiration (PET) in most ENSO-sensitive regions. Additionally, ~39% of the regions currently sensitive to ENSO receive less precipitation in the future, independent of the ENSO phase. Aridity increases consequently in 67-72% of the ENSO-sensitive area. When both radiative and physiological effects are considered, the area affected by aridity rises to 75-79% when using PET-derived measures of aridity, but declines to 41% when total soil moisture aridity indicator is employed. This reduction mainly occurs because plant stomatal resistance increases under enhanced CO₂ concentrations, which results in improved plant water use efficiency, and hence reduced evapotranspiration and soil desiccation. Imposing CO₂-invariant stomatal resistance may overestimate future drying in PET-derived indices.

Figure 1

a–b): Ensemble-mean surface T anomalies (°C) in response to El Niño (left column) and La Niña (right column) events based on historical (ΔH_h) and future (ΔF_f) anomalies, calculated relative to their own mean state. Regions that do not satisfy the two “ENSO-sensitive” criteria (see text) are masked. c–d–e): ΔF_h anomalies in response to ocean warming (+SST), ocean warming and plant physiological effects (+SST+Veg), and ocean warming, plant physiology, and the fast radiative effect (+SST+Veg+Rad). All anomalies are calculated relative to the same (amip) historical baseline. Stippling in c–e highlights all land grid-points where the sign of the variable changes between the two ENSO phases (Note: stippling is not present in the T figure, because the phase of ENSO only modulates the amplitude of the warming). f–g): Effects of +Veg only and +Rad only.

Figure 2

a–b): Ensemble-mean precipitation (mm/day) in response to El Niño (left column) and La Niña (right column) events based on historical (ΔH_h) and future (ΔF_f) anomalies, calculated relative to their own mean state. Regions that do not satisfy the two “ENSO-sensitive” criteria (see text) are masked. c–d–e): ΔF_h anomalies in response to ocean warming (+SST), ocean warming and plant physiological effects (+SST+Veg), and ocean warming, plant physiology, and the fast radiative effect (+SST+Veg+Rad). All anomalies are calculated relative to the same (amip) historical baseline. Red stippling in c–e highlights all land grid-points where the sign of the variable changes between the two ENSO phases. f–g): Effects of +Veg only and +Rad only.

Figure 3

a–f) Quantitative breakdown (in %) of the historical “ENSO-sensitive” land into 4 different categories. Results are for: a) P; b) PET and six aridity indices; c) CMI; d) P/(P+PET); e) PDSI; f) surface soil moisture; g) total soil moisture; and h) total runoff. Relative to the mean state during the historical period, the four regions distinguished here are either wetter during El Niño events, with positive anomalies for P and the six indices, or negative anomalies for PET (light green shading); drier during El Niño events, with negative anomalies for P and the six indices or positive anomalies for PET (yellow shading); and wetter or drier, irrespective of the ENSO phases (dark green and dark brown shading, respectively). The percentage of the land area in each of these four categories is calculated for the historical period (ΔH_h), and for the future climate (relative to the historical mean state; ΔF_h). In the latter case, results are given for the response to ocean warming only (+SST), as well as for the inclusion of plant physiological effects (+SST+Veg) and the fast radiative effect (+SST+Veg+Rad). Each individual error bar represents is ±1σ uncertainty estimated from the small (4-model) sampling distribution. The percentage of land between 60°S and 75°N qualifying as sensitive to ENSO (R₁ region) is indicated in the titles. Total runoff outputs are missing for HadGEM2-ES and surface soil moisture is only available for IPSL-CM5A-LR.

Figure 4

Same as Figure 3 but for: a) ET; b) GPP; c) LAI; d) EV_s; e) EV_v; f) TR; g) ET-derived CMI; and h) P/(P+ET). The color convention for panels a), and d–h) is the same as in Figure 3. For panels b) and c), results are partitioned between regions with a positive anomaly (i.e., more leaves, or more carbon uptake) during El Niño events (salmon shading); regions showing a negative anomaly (light-blue shading) during El Niño events; and regions with either positive or negative anomalies, irrespective of the ENSO phase (red and dark-blue shading, respectively). Note that LAI outputs are missing for HadGEM2-ES, and the quantitative breakdown of ET into EV_s, Ev_v and TR is only available for IPSL-CM5A-LR.

Figure 5

a–b): Ensemble-mean P/(P+PET) in response to El Niño (left column) and La Niña (right column) events based on historical (ΔH_h) and future (ΔF_f) anomalies, calculated relative to their own mean state. Regions that do not satisfy the two “ENSO-sensitive” criteria (see text) are masked. c–d–e): ΔF_h anomalies in response to ocean warming (+SST), ocean warming and plant physiological effects (+SST+Veg), and ocean warming, plant physiology, and the fast radiative effect (+SST+Veg+Rad). All anomalies are calculated relative to the same (amip) historical baseline. Red stippling in c-e highlights all land grid-points where the sign of the variable changes between the two ENSO phases. f–g): Effects of +Veg only and +Rad only.

Figure 6

a–b): Ensemble-mean total soil moisture in response to El Niño (left column) and La Niña (right column) events based on historical (ΔH_h) and future (ΔF_f) anomalies, calculated relative to their own mean state. Regions that do not satisfy the two “ENSO-sensitive” criteria (see text) are masked. c–d–e): ΔF_h anomalies in response to ocean warming (+SST), ocean warming and plant physiological effects (+SST+Veg), and ocean warming, plant physiology, and the fast radiative effect (+SST+Veg+Rad). All anomalies are calculated relative to the same (amip) historical baseline. Red stippling in c-e highlights all land grid-points where the sign of the variable changes between the two ENSO phases. f–g): Effects of +Veg only and +Rad only.