Water, nitrogen and phosphorus use efficiencies of four tree species in response to variable water and nutrient supply

Feike A. Dijkstra & Yolima Carrillo & Michael J. Aspinwall & Chelsea Maier &Alberto Canarini & Hero Tahaei & Brendan Choat & David T. Tissue

Published: 31 March 2016


Background and Aims The commonly observed trade-off between plant water use efficiency (WUE) and ni-trogen use efficiency (NUE) has been attributed tophysiological constraints in the leaf. We examined if asimilar trade-off can occur between WUE and phospho-rus use efficiency (PUE) and if changes in NUE andPUEinresponsetowaterandnutrientsupplycanberelated to microbial N and P immobilisation.Methods We studied water and nutrient use efficienciesin leaves of four tree species (Eucalyptus sideroxylon,Eucalyptus tereticornis, Casuarina cunninghamiana,and Pinus radiata) that were grown under rainout shel-ters for 16 months at low and high levels of water andnutrient supply.Results Across all four species, WUE increased whenwater supply was low and nutrient supply was high,while NUE increased when water supply was high andnutrient supply was low. As a consequence, a trade-offwas found between WUE and NUE for all four species.

In contrast, no trade-off was found between WUE andPUE, likely because PUE and microbial P immobilisa-tion in the soil unexpectedly increased with high nutri-ent supply.Conclusions With variable water and nutrient supply,physiological constraints generate a trade-off betweenWUE and NUE, but not between WUE and PUE; thelatter may have been obscured by microbial control overPsupplytoplants.


When water and nutrients are scarce, efficient use ofthese resources may improve plant growth and chancesof survival (Aerts and Chapin III 2000; Tilman 1988).Plants can respond to conditions of low water availabil-ity by transpiring less water per unit carbon fixed andbiomass produced (i.e., they have a high water useefficiency, WUE, Ehleringer and Cooper 1988; Toftet al. 1989). Likewise, in environments with low nitro-gen (N) availability plants often fix more carbon andproduce more biomass per unit of N (i.e., they have ahigh N use efficiency, NUE, Aerts and Chapin III 2000;Vitousek 1982). However, plants often decrease WUEwith increased NUE and vice versa, suggesting a trade-off between WUE and NUE (Field et al. 1983; Gonget al. 2011; Patterson et al. 1997). This trade-off can beexplained by physiological constraints in the leaf: when plants increase leaf intercellular CO2concentration byopening their stomata, this increases photosynthesis perunit of leaf N, but also increases transpiration therebyreducing WUE (or plant δ13C, an integrative index ofleaf-scale WUE, Farquhar and Ric hards 1984).Therefore, reduced leaf WUE induced by greater wateravailability frequently increases leaf NUE (Li et al.2003; Reich et al. 1989). Similarly, when N availabilityincreases, leaf tissue N increases (and NUE decreases),which enhances the photosynthetic capacity in leavesthereby increasing WUE (Livingston et al. 1999;Ripullone et al. 2004).The trade-off between WUE and NUE in leaves hasalso been explained by N availability in the soil (Crameret al. 2009). It was argued that when N availability islow, plants may enhance transpiration rates to increaseN uptake through mass flow in the soil. As a result,WUE may decrease with increased NUE induced byconditions of lower soil N availability, at least whenplant growth is not limited by water.In contrast, there is no clear evidence for a trade-offbetween leaf WUE and phosphorus (P) use efficiency(PUE) (but see Brown et al. 2011). Addition of P had noeffect on WUE in the tropical tree Ficus insipida(Garrish et al. 2010), while leaf PUE was unrelated toWUE in Pinus pinaster (Warren et al. 2005), or waseven positively related with WUE in Populus and grass-land species (DesRochers et al. 2006;Zhouetal.2013).

Increased leaf P concentrations in trees and lianas wereexplained by enhanced transpiration rates increasingmass flow of P (Cernusak et al. 2010;Cernusaketal.2011), so that variation in transpiration rates may actu-ally result in positive relationships between WUE andPUE.In all of these studies, plant-soil interactions have notbeen considered. However, changes in water and nutri-ent availabi lity in the soil may affect microbialmineralisation and immobilisation of N and P (Austinet al. 2004; Bünemann et al. 2012; Manzoni andPorporato 2007), with consequences for the absoluteand relative amounts of N and P that can be taken upby plants. Microbial requirement for P is usually muchhigher than for N (Cleveland and Liptzin 2007), whichcould potentially result in imbalanced supply of N and Pto plants (Jonasson et al. 1996). Therefore, changes inmicrobial activity due to variation in water and nutrientavailability could affect plant N and P uptake, andthereby influence relationships among WUE, NUEand PUE.Here we examined leaf δ13C, C/N and C/P ratios (asindices of leaf-scale WUE, NUE and PUE, respectively)in four tree species in response to changes in soil waterand nutrient supply. The species were grown in localnative forest soil in large 75 L bags for 16 months underrainout shelters with two levels of water supply (con-stantly watered vs. drought periods) and two levels of Nand P fertiliser additions in a full factorial design. Wethen examined relationships between leaf δ13CandC/N,and between leaf δ13C and leaf C/P, and determinedwhether relationships could be explained solely byphysiological constraints in the leaf, or also by changesin microbial N and P immobilisation.