Legacies of La Niña: North American monsoon can rescue trees from winter drought

Abstract

While we often assume tree growth–climate relationships are time‐invariant, impacts of climate phenomena such as the El Niño Southern Oscillation (ENSO) and the North American Monsoon (NAM) may challenge this assumption. To test this assumption, we grouped ring widths (1900‐present) in three southwestern US conifers into La Niña periods (LNP) and other years (OY). The 4 years following each La Niña year are included in LNP, and despite 1–2 year growth declines, compensatory adjustments in tree growth responses result in essentially equal mean growth in LNP and OY, as average growth exceeds OY means 2–4 years after La Niña events. We found this arises because growth responses in the two periods are not interchangeable: Due to differences in growth–climate sensitivities and climatic memory, parameters representing LNP growth fail to predict OY growth and vice versa (decreases in R2 up to 0.63; lowest R2 = 0.06). Temporal relationships between growth and antecedent climate (memory) show warmer springs and longer growing seasons negatively impact growth following dry La Niña winters, but that NAM moisture can rescue trees after these events. Increased importance of monsoonal precipitation during LNP is key, as the largest La Niña‐related precipitation deficits and monsoonal precipitation contributions both occur in the southern part of the region. Decreases in first order autocorrelation during LNP were largest in the heart of the monsoon region, reflecting both the greatest initial growth declines and the largest recovery. Understanding the unique climatic controls on growth in Southwest conifers requires consideration of both the influences and interactions of drought, ENSO, and NAM, each of which is likely to change with continued warming. While plasticity of growth sensitivity and memory has allowed relatively quick recovery in the tree‐ring record, recent widespread mortality events suggest conditions may soon exceed the capacity for adjustment in current populations.

Legacies of more frequent drought in ponderosa pine across the western United States

“…When there have been no recent droughts ( d = 0), all three regions respond positively to precipitation (Figure a), negatively to temperature and PDSI (Figure b,c), and negatively to the P ant × D ant interaction (though magnitudes may strongly differ, Figure d). A negative PDSI effect is consistent with our previous work that analyzed a subset of these climate and tree‐ring datasets using a similar modeling framework (Peltier & Ogle, ). Since this covariate is analogous to a precipitation by (negative) temperature interaction, the negative PDSI effects simply reflect greater sensitivity of ring width to precipitation when it is warm (that is, when ‘negative temperature' is low) (Peltier & Ogle, ).…”

“…A negative PDSI effect is consistent with our previous work that analyzed a subset of these climate and tree‐ring datasets using a similar modeling framework (Peltier & Ogle, ). Since this covariate is analogous to a precipitation by (negative) temperature interaction, the negative PDSI effects simply reflect greater sensitivity of ring width to precipitation when it is warm (that is, when ‘negative temperature' is low) (Peltier & Ogle, ).…”

“…In particular, during widespread drought, drought‐related changes in growth–climate sensitivities in interior populations (SW and IM, Figure ) result in greater growth than predicted if trees did not adjust their sensitivities (Figure a,b). Interior P. ponderosa , particularly in the SW region, experience strong interannual variability in both cold‐ and warm‐season precipitation, and another study indicates P. ponderosa and other conifer species show plasticity in precipitation responses at subannual timescales following drought (Peltier & Ogle, ). Sensitivity to climate in general is highest in interior varieties, potentially due to local adaptation to highly variable climate (McCullough et al, ).…”

“…To investigate the potential impacts of compounded drought and changes in growth–climate sensitivities due to associated drought legacies, we constructed a novel model based upon the SAM framework (Ogle et al, ). This framework has been applied to a number of problems (Dal Bello, Rindi, & Benedetti‐Cecchi, ; Guo & Ogle, ; Ogle et al, ; Ryan et al, , ), including applications to tree‐ring chronologies (Peltier et al, ; Peltier & Ogle, ). For a detailed description of the SAM model as applied to tree‐ring data, see Peltier et al ().…”

“…Thus, the assumption of many terrestrial ecosystem models that recovery is both immediate and complete (or that growth–climate sensitivities are constant through time) is probably invalid, at least at large scales (Kolus et al, ). Furthermore, induced changes in growth–climate sensitivities have been shown to vary across space within conifer species such as Pinus ponderosa , suggesting the need to consider variation among populations (Anderegg, Schwalm, et al, ; McCullough, Davis, & Williams, ; Peltier & Ogle, ). While tree rings cannot be used to estimate carbon fluxes directly due to both historic sampling design of available datasets focusing on climate sensitive trees (Klesse et al, ; Nehrbass‐Ahles et al, ) and the complexity of terrestrial net ecosystem exchange (Babst et al, ), they can provide directional information on the major drivers of forest net primary productivity (Babst et al, ), with strong implications for overall fluxes.…”