plants

Article Vertical Strata and Stem Carbon Dioxide Efflux in Cycas Trees

Thomas E. Marler 1,* and Murukesan V. Krishnapillai 2 1 College of Natural and Applied Sciences, University of Guam, Mangilao, Guam 96923, USA 2 Cooperative Research and Extension, College of Micronesia-FSM, Yap Campus, Yap 96943, Micronesia; [email protected] * Correspondence: [email protected]



Received: 5 January 2020; Accepted: 9 February 2020; Published: 11 February 2020


Abstract: Stem respiration is influenced by the vertical location of tree stems, but the influence of vertical location on stem respiration in a representative cycad species has not been determined. We quantified the influence of vertical strata on stem carbon dioxide efflux (Es) for six arborescent Cycas L. species to characterize this component of stem respiration and ecosystem carbon cycling. The influence of strata on Es was remarkably consistent among the species, with a stable baseline flux characterizing the full mid-strata of the pachycaulous stems and an increase in Es at the lowest and 2 1 highest strata. The mid-strata flux ranged from 1.8 µmol m− s− for Cycas micronesica K.D. Hill to 3.5 2 1 · · µmol m s for Cycas revoluta Thunb. For all species, Es increased about 30% at the lowest stratum · − · − and about 80% at the highest stratum. A significant quadratic model adequately described the Es patterns for all six species. The increase of Es at the lowest stratum was consistent with the influence of root-respired carbon dioxide entering the stem via sap flow, then contributing to Es via radial conductance to the stem surface. The substantial increase in Es at the highest stratum is likely a result of the growth and maintenance respiration of the massive cycad primary thickening meristem that constructs the unique pachycaulous cycad stem.

Keywords: conservation physiology; Cycas edentata; Cycas micronesica; Cycas nitida; Cycas revoluta; Cycas wadei; Cycas zambalensis; stem respiration

1. Introduction Carbon dioxide is among the greenhouse gases that are emitted by tree stems and the amounts are great enough for this source of carbon dioxide to influence regional and global carbon cycles [1,2]. Vertical variations in stem respiration have been reported, with the stem surfaces directly above the root collar exhibiting greater carbon dioxide efflux (Es) than stem surfaces at higher strata [3–5]. Interpretations focus on respired carbon dioxide in root tissues that enters root xylem then moves to stem tissues in sap flow. This root-derived carbon dioxide responds to a radial conductance gradient toward the free air at these basal stem strata and some estimates indicate that about half of the root-respired carbon dioxide may enter the atmosphere by way of Es [3]. The ability of stem xylem to dissolve locally respired carbon dioxide then transport it in the transpiration stream may also influence Es in higher strata and is evinced in diel cycles of Es as the changes in sap flow directly modify the conductance of carbon dioxide toward the stem surface [6–9]. As a result of sap flow and localized storage or refixation, about 40% of the respired carbon dioxide in stems may not be emitted from stem surfaces closest to the site of respiration [10]. A direct determination of the influence of vertical strata on tree Es has been the subject of several studies [4,5,11–16]. Results either revealed little influence of vertical strata on Es or indicated that an increase in Es occurred in the highest strata of a tree’s bole and branches. The reported increase in

Plants 2020, 9, 230; doi:10.3390/plants9020230 www.mdpi.com/journal/plants Plants 2020, 9, 230 2 of 10

Es at higher strata may be a result of greater respiration in the more active stem tissues in the strata that are closest to the highly active leaf crown, a general reduction in the resistance to radial carbon dioxide conductance in the smaller branches at higher strata, and/or the influence of vertical strata on refixation or storage of carbon dioxide in stem tissues. The literature on Es is biased toward one stem form: pycnoxylic eudicot and gymnosperm trees that increase in stem diameter by bifacial secondary vascular cambium. Cycads produce manoxylic stems and, to our knowledge, only one report on Es of cycads exists [17]. Cycads comprise an interesting group of gymnosperms that portray primitive features and cycads are intriguing to botanists because of their antiquity, unique traits, and beauty [18]. The cycad stem is constructed in a unique manner to produce a pachycaulous form that contains persistent pith and cortex and is comprised mostly of living parenchyma tissues [18,19]. Radial enlargement of the cycad stem is mostly restricted to the primary thickening meristem [20–22], the largest known apical meristem in the plant kingdom [18]. Clearly, our current level of knowledge of Es is deficient because manoxylic, pachycaulous trees have been ignored in the research to date. The simplicity of cycad stem architecture removes some of the complications in interpreting the influence of tree height on Es. Indeed, the published studies on the subject reported Es from the bole for lowest strata but from small side branches for the highest strata of the canopy. Stem diameter and radial organization of tissues may exert a direct influence on Es [11,14], so the influences of vertical strata on Es of these species may be mediated via the canopy architecture and stem design rather than via vertical strata per se. The pachycaulous cycad stem is essentially a cylinder of similar diameter for most of the tree height, providing a model tree form where the vertical gradient of Es can be studied without the complication of elaborate branch architecture and disparate internal organization of different sized stems. Our objectives were to determine the influence of vertical strata on Es for six representative arborescent Cycas species (Figure1). We predicted little influence of vertical strata on Es from the root collar to the mid-strata of each tree, then an increase in Es close to the stem apex due to respiratory Plantsactivity 2020, 9, xof FOR the PEER highly-active REVIEW primary thickening meristem. 3 of 10

Figure 1. Cont.

Figure 1. The general appearance of six Cycas tree species in the study locations. (a) Cycas edentata in an agroforest in Angeles City, Philippines; (b) Cycas micronesica in native Yap habitat; (c) Cycas nitida in its native northern Samar, Philippines habitat; (d) Cycas revoluta in an urban landscape in Angeles City, Philippines; (e) Cycas wadei in its endemic Culion Island, Philippines habitat; (f) Cycas zambalensis in its endemic San Antonio, Philippines habitat.

2. Results Midday air temperature, stem surface temperature, and relative humidity were similar among the six locations of study. The tallest stratum of measurements ranged from 2.8 m for C. revoluta to 3.9 m for C. micronesica and C. nitida (Figure 2). The mid-section of the stems exhibited homogeneous Es for all six arborescent Cycas species in this study (Figure 2). In ascending order, the mid-strata Es was 1.8 µmol·m-2·s-1 for C. micronesica, 1.9 µmol·m-2·s-1 for Cycas edentata de Laub., 2.0 µmol·m-2·s-1 for Cycas nitida K.D. Hill & A. Lindstrom, 2.1 µmol·m-2·s-1 for Cycas wadei Merrill, 2.5 µmol·m-2·s-1 for Cycas zambalensis Madulid & Agoo, and 3.5 µmol·m-2·s-1 for C. revoluta. The Es of the lowest stratum for all six species increased about 30% above that of the mid-strata (Figure 2). An increase in stem diameter also occurred at the lowest strata of the stems (Figure 2). The Es at the lowest stratum was 2.4 µmol·m-2·s-1 for C. micronesica and C. edentata, 2.6 µmol·m-2·s-1 for C. nitida and C. wadei, 3.1 µmol·m-2·s-1 for C. zambalensis, and 4.1 µmol·m-2·s-1 for C. revoluta. An 80% increase in Es above the mid-strata Es occurred in the highest measurement stratum (Figure 2). This apical stem Es was 3.3 µmol·m-2·s-1 for C. micronesica, 3.4 µmol·m-2·s-1 for C. edentata, 3.6 µmol·m-2·s-1 for C. nitida and C. wadei, 4.5 µmol·m-2·s-1 for C. zambalensis, and 5.9 µmol·m-2·s-1 for C. revoluta. The influence of vertical strata on Es could be fitted with significant quadratic regression models for every individual and every species (Table 1, Figure 2). The Es of the second lowest stratum was

Plants 2020, 9, x FOR PEER REVIEW 3 of 10

Plants 2020, 9, 230 3 of 10

FigureFigure 1. The 1. generalThe general appearance appearance of six of Cycas six Cycas tree treespecies species in the in study the study locations. locations. (a) Cycas (a) Cycas edentata edentata in in an agroforestan agroforest in Angeles in Angeles City, City, Philippines; Philippines; (b) (Cycasb) Cycas micronesica micronesica in nativein native Yap Yap habitat; habitat; (c) ( cCycas) Cycas nitida nitida in in itsits native native northern northern Samar, Samar, Philippines Philippines habitat; habitat; ( (dd)) CycasCycas revoluta revoluta inin an urban landscapelandscape inin AngelesAngeles City, City,Philippines; Philippines; ( e(e) )Cycas Cycas wadei wadeiin in its its endemic endemic Culion Culion Island, Island, Philippines Philippines habitat; habitat; ( f()f)Cycas Cycas zambalensis zambalensisin its in itsendemic endemic San San Antonio, Antonio, Philippines Philippines habitat. habitat. 2. Results 2. Results Midday air temperature, stem surface temperature, and relative humidity were similar among the Midday air temperature, stem surface temperature, and relative humidity were similar among six locations of study. The tallest stratum of measurements ranged from 2.8 m for C. revoluta to 3.9 m the six locations of study. The tallest stratum of measurements ranged from 2.8 m for C. revoluta to for C. micronesica and C. nitida (Figure2). The mid-section of the stems exhibited homogeneous E for 3.9 m for C. micronesica and C. nitida (Figure 2). The mid-section of the stems exhibited homogeneouss all six arborescent Cycas species in this study (Figure2). In ascending order, the mid-strata Es was Es for all six arborescent Cycas species in this study (Figure 2). In ascending order, the mid-strata Es 1.8 µmol m 2 s 1 for C. micronesica, 1.9 µmol m 2 s 1 for Cycas edentata de Laub., 2.0 µmol m 2 s 1 for was 1.8 µmol··m−-2·s−-1 for C. micronesica, 1.9 µmol·m· -2−·s·-1 −for Cycas edentata de Laub., 2.0 µmol·m·-2·s−-1 ·for− Cycas nitida K.D. Hill & A. Lindstrom, 2.1 µmol m 2 s 1 for Cycas wadei Merrill, 2.5 µmol m 2 s 1 for Cycas nitida K.D. Hill & A. Lindstrom, 2.1 µmol·m·-2·s−-1 ·for− Cycas wadei Merrill, 2.5 µmol·m·-2·s−-1 ·for− Cycas zambalensis Madulid & Agoo, and 3.5 µmol m 2 s 1 for C. revoluta. Cycas zambalensis Madulid & Agoo, and 3.5 µmol·m-2· s-1− for· − C. revoluta. The Es of the lowest stratum for all six species increased about 30% above that of the mid-strata The Es of the lowest stratum for all six species increased about 30% above that of the mid-strata (Figure2). An increase in stem diameter also occurred at the lowest strata of the stems (Figure2). The (Figure 2). An increase in stem diameter also occurred2 1 at the lowest strata of the stems (Figure 2). 2The1 Es at the lowest stratum was 2.4 µmol m− s− for C. micronesica and C. edentata, 2.6 µmol m− s− for Es at the lowest stratum was 2.4 µmol·m-2·s-1 for· C. micronesica and C. edentata, 2.6 µmol·m-2·s·-1 for· C. C. nitida and C. wadei, 3.1 µmol m 2 s 1 for C. zambalensis, and 4.1 µmol m 2 s 1 for C. revoluta. An nitida and C. wadei, 3.1 µmol·m-2·s-1· for− ·C−. zambalensis, and 4.1 µmol·m-2·s-1 ·for− C.· −revoluta. An 80% 80% increase in Es above the mid-strata Es occurred in the highest measurement stratum (Figure2). increase in Es above the mid-strata Es occurred2 1 in the highest measurement stratum2 1 (Figure 2). This This apical stem Es was 3.3 µmol m− s− for C. micronesica, 3.4 µmol m− s− for C. edentata, 3.6 apical stem Es was 3.3 µmol·m-2·s-1 for ·C. micronesica,· 3.4 µmol·m-2·s-1 for C. edentata· · , 3.6 µmol·m-2·s-1 µmol m 2 s 1 for C. nitida and C. wadei, 4.5 µmol m 2 s 1 for C. zambalensis, and 5.9 µmol m 2 s 1 for for C. nitida· −and· − C. wadei, 4.5 µmol·m-2·s-1 for C. zambalensis· − · −, and 5.9 µmol·m-2·s-1 for C. revoluta· . − · − C. revoluta. The influence of vertical strata on Es could be fitted with significant quadratic regression models The influence of vertical strata on Es could be fitted with significant quadratic regression models for every individual and every species (Table 1, Figure 2). The Es of the second lowest stratum was for every individual and every species (Table1, Figure2). The Es of the second lowest stratum was

more similar to the mid-strata Es than the Es at the lowest stratum for these Cycas trees. In contrast, the Es of the second highest stratum revealed a moderate increase above that of the mid-strata Es for four of the Cycas species, but not for C. edentata or C. revoluta. Plants 2020, 9, x FOR PEER REVIEW 4 of 10 more similar to the mid-strata Es than the Es at the lowest stratum for these Cycas trees. In contrast, the Es of the second highest stratum revealed a moderate increase above that of the mid-strata Es for Plants 2020, 9, 230 4 of 10 four of the Cycas species, but not for C. edentata or C. revoluta.

Figure 2. The relationship between vertical strata and stem carbon dioxide efflux (Es) for six arborescent FigureCycas 2. species.The relationship (a) Cycas edentata between;(b ) verticalCycas micronesica strata and;(c) Cycas stem nitida carbon;(d ) dioxideCycas revoluta efflux;( e ()ECycass) for six arborescentwadei; and Cycas (f) Cycas species. zambalensis (a) Cycas. Solid edentata lines represent; (b) Cycas the micronesica quadratic model; (c) Cycas calculated nitida from; (d) the Cycas mean revoluta of ; (e) Cycaseach parameterwadei; and after (f) Cycas fitting zambalensis the data for. each Solid of lines the six represent replications the per quadratic species, with model axes calculated transposed. from the meanNumbers of each on theparameter right vertical afteraxis fitting represent the data the for mean each stem of t diameterhe six replications for the six replicationsper species of, with each axes 2 2 transposedspecies.. ModelsNumbers for each on the species right were: vertical (a) Es = axis2.63 represent0.01*height the+ mean0.00004*height stem diameter; P < 0.001; forR the= six 2 − 2 replications0.78; (b) ofEs each= 2.54 species.0.01*height Models+ 0.00003*heightfor each species; P were:< 0.001; R = 0.88; (c) Es = 2.77 0.01*height + 2 − 2 2 − 2 0.00003*height ; P < 0.001; R = 0.87; (d) Es = 4.61 0.22*height + 0.00008*height ; P < 0.001; R = 2 − 2 0.71; (e) Es = 2.93 0.01*height + 0.00004*height ; P < 0.001; R = 0.85; (f) Es = 3.24 0.01*height + (a)− Es = 2.63 – 0.01*height + 0.00004*height2; P < 0.001; R2 = 0.78 − 0.00004*height2; P < 0.001; R2 = 0.79. (b) Es = 2.54 – 0.01*height + 0.00003*height2; P < 0.001; R2 = 0.88 Stem carbon dioxide(c) Es = e2.77fflux – increased0.01*height with + 0.00003*height stem diameter2; P (Figure < 0.001;3 ).R2 The = 0.87 inclusion of 300 data pairs in the regression(d) generatedEs = 4.61 –a 0.22*height highly significant + 0.00008*height linear model,2; P < but 0.001; the considerableR2 = 0.71 scatter reduced the fitness. (e) Es = 2.93 – 0.01*height + 0.00004*height2; P < 0.001; R2 = 0.85 (f) Es = 3.24 – 0.01*height + 0.00004*height2; P < 0.001; R2 = 0.79

Plants 2020, 9, x FOR PEER REVIEW 5 of 10

Table 1. Characteristics of the six study locations used for Cycas stem carbon dioxide efflux measurements. Range of six replications.

Air Stem Relative Quadratic Model Quadratic Species Temperature Temperature Humidity Plants 2020, 9, 230 P Value Model5 R of2 10 (°C) (°C) (%) Cycas edentata 27–32 26–32 59–65 0.0079–0.0018 0.78–0.82 CycasTable micronesica 1. Characteristics 28– of31 the six study26– locations31 used60– for69 Cycas stem0.0021 carbon–0.0011 dioxide 0.88 efflux–0.90 measurements. Range of six replications. Cycas nitida 26–30 25–30 63–71 0.0014–0.0018 0.87–0.90 Cycas revoluta Air Temperature27–33 Stem27 Temperature–31 59–Relative66 0.0087Quadratic–0.0015 Quadratic0.71–0.77 Species 2 Cycas wadei (29◦C)–34 27(–◦C)33 60Humidity–68 (%) 0.0500Model–0.0501P Value 0.85Model–0.88 R CycasCycas zambalensis edentata 27–3226–30 25 26–32–29 58–64 59–65 0.0139 0.0079–0.0018–0.0110 0.79 0.78–0.82–0.84 Cycas micronesica 28–31 26–31 60–69 0.0021–0.0011 0.88–0.90 Cycas nitida Stem carbon dioxide26–30 efflux increased with 25–30 stem diameter 63–71 (Figure 3). 0.0014–0.0018 The inclusion of 0.87–0.90 300 data Cycas revoluta 27–33 27–31 59–66 0.0087–0.0015 0.71–0.77 pairsCycas in the wadei regression generated29–34 a highly 27–33 significant linear 60–68 model, but 0.0500–0.0501 the considerable 0.85–0.88 scatter reducedCycas zambalensis the fitness. 26–30 25–29 58–64 0.0139–0.0110 0.79–0.84

FigureFigure 3. 3.The The relationship relationship between between stem stem diameter diameter and stem and carbon stem dioxide carbon e dioxidefflux (Es ) efflux for six arborescent(Es) for six Cycasarborescentspecies. Cycas species.

3.3. Discussion 3.1. Cycad Stems 3.1. Cycad Stems The homogeneous cycad pachycaulous stem construction design [23] may explain the similarity The homogeneous cycad pachycaulous stem construction design [23] may explain the similarity in Es for our mid-strata measurements. Pachycaulous stem forms have evolved in other plant groups, in Es for our mid-strata measurements. Pachycaulous stem forms have evolved in other plant groups, such as palms and other arborescent monocot species, but the cycad stem is constructed by the largest such as palms and other arborescent monocot species, but the cycad stem is constructed by the largest known primary thickening meristem of any plant group [18]. Radial enlargement in sections of the known primary thickening meristem of any plant group [18]. Radial enlargement in sections of the cycad stem subtending the leaf crown occurs by way of lethargic disjunct mitotic activity that is cycad stem subtending the leaf crown occurs by way of lethargic disjunct mitotic activity that is scattered among the tissues throughout the cross-section of the stem [24]. These traits of the cycad scattered among the tissues throughout the cross-section of the stem [24]. These traits of the cycad pachycaulous stem may explain the remarkable similarity in Es throughout most of the vertical strata pachycaulous stem may explain the remarkable similarity in Es throughout most of the vertical strata of the Cycas stems for all six species that we studied and this partly conformed to our first prediction. of the Cycas stems for all six species that we studied and this partly conformed to our first prediction. The moderate increase in Es at the lowest stratum was not consistent with our first prediction. The moderate increase in Es at the lowest stratum was not consistent with our first prediction. Four anatomical and physiological features of the cycad stem may explain this increase in Es near the Four anatomical and physiological features of the cycad stem may explain this increase in Es near the root collar. First, increased Es at basal stem strata may result from the contribution of root-respired root collar. First, increased Es at basal stem strata may result from the contribution of root-respired carbon dioxide that is xylem-transported to the stem [3–5]. Second, the peripheral tissues of a cycad carbon dioxide that is xylem-transported to the stem [3–5]. Second, the peripheral tissues of a cycad stem are comprised of a distinct corky periderm layer and persistent leaf petiole scars that include stem are comprised of a distinct corky periderm layer and persistent leaf petiole scars that include living and dead structures [18,23,25]. The thickness of this cork tissue is greatest at the base of cycad trees [18]. The secondary vascular cambium and bark tissues of pycnoxylic eudicot and gymnosperm trees may act as conductance barriers to gases [26,27]. Similarly, the peripheral structures of a cycad stem may act as conductance barriers in a manner that influences carbon dioxide diffusion to the Plants 2020, 9, 230 6 of 10 open air and this trait may vary among cycad species and along the vertical axis of the pachycaulous cycad stem. Third, the absolute diameter of the cortex increases with cycad stem height and cortex width may influence Es [18,24]. Fourth, vascular strands traverse the cycad stem cortex in a non-linear manner, connecting the vascular cylinders with each leaf, and these strands persist in the cortex tissue following leaf mortality [18,22,28,29]. These vascular strands at the base of a tall cycad tree may be decades or centuries in age, yet the vascular strands near the stem apex are relatively young. The need to more fully understand these vertical gradients in tree Es in the face of ongoing climate change is underscored by the reported influence of elevated atmosphere carbon dioxide concentration on the relationship between vertical strata and Es [30]. The substantial increase in Es at the highest stem stratum conformed to our second prediction. The mitotic tissues that construct radial cycad stem growth are uniquely constricted to the massive primary thickening meristem [20–22], so restriction of the greatest growth and maintenance respiration to the stem stratum that contains this structure was expected. The relative Es of the second highest stem stratum varied among the six species. This contrast in behavior was probably a function of how close these measurements were to the primary thickening meristem in each replication’s stem apex, which was controlled by the axial location of the oldest living petioles of the leaf crowns of each species. The poorer fit of the quadratic models for C. edentata and C. revoluta than for the other species (Table1) was likely due to the fact that the observed Es values of this second highest stratum for these two species were less aligned with the non-linear upper portions of the regressions’ predicted values. The range in Es was 1.9-fold for the baseline flux of the mid-sections of stems for these Cycas species, the range in Es was 1.8-fold for the flux near the leaf crown, and the range in Es was 1.7-fold for the flux near the root collar. Therefore, the variation of Es among Cycas plants appears more constrained at basal and apical stem locations than at mid-locations of the stem. This variation among only six congeneric cycad taxa reveals the need to add more pachycaulous tree species to the global Es database.

3.2. Stem Respiration Literature Bias

The Es literature is comprised almost exclusively of pycnoxylic eudicot and gymnosperm tree species that grow radially by bifacial secondary vascular cambium. We are aware of only one other report that included Es of pachycaulous tree species. An impressive data set collection of woody tissue carbon dioxide efflux was conducted in Costa Rica and one growth form in this study was palms [14]. The arborescent palm stem is also classified as pachycaulous, however, the palm stem anatomy and methods of radial stem growth are inconsistent with cycad stems. One similarity is evident from the Costa Rica palm results and our cycad results: the mid-strata of the pachycaulous palm and cycad stem is essentially a cylinder with minimal variation in Es. The Es of this mid-section was less than 1 µmol m 2 s 1 for the palm stems (no species were named) in the Costa Rica study [14] and up to 3.5 · − · − µmol m 2 s 1 for the Cycas stems in our study. · − · − Our understanding of Es is hindered by the lack of data on tree species with growth forms that do not expand radially by secondary vascular cambium. We have addressed this limitation by adding six cycad species to the literature. The generality of the quadratic pattern of the relationship between stem height increment and Es remains unexplored for other arborescent cycad taxa. Clearly, more data are urgently needed for cycads, palms, and other monocot arborescent species to fully understand Es across all tree species and develop models that adequately predict all arborescent functional groups. Ecosystem carbon cycling models based on the current Es literature are not relevant for localities with numerous palms, cycads, or other monocot trees.

4. Materials and Methods Access to a large number of tall cycad trees within homogeneous edaphic and climatic conditions is difficult to achieve. We addressed these limitations by accessing high-density in situ localities for four Cycas species and urban landscape specimens within one municipality for two common garden species. Plants 2020, 9, x FOR PEER REVIEW 7 of 10

4. Materials and Methods Access to a large number of tall cycad trees within homogeneous edaphic and climatic conditions is difficult to achieve. We addressed these limitations by accessing high-density in situ localities for four Cycas species and urban landscape specimens within one municipality for two common garden species. For each species, the populations were carefully surveyed for the tallest available, containing at least six representative trees. Although cycads do not produce axillary buds during primary growth, the stems do possess the ability to produce adventitious buds that generate Plants 2020, 9, 230 7 of 10 branching [18]. We restricted the selection of replications to monopodial individuals and did not include trees with multiple branches. The initiation of reproductive structures on dioecious cycad For each species, the populations were carefully surveyed for the tallest available, containing at least six trees is infrequent and we did not include any trees with developing reproductive structures. These representative trees. Although cycads do not produce axillary buds during primary growth, the stems methodsdo ensured possess thethat ability this tofirst produce look adventitiousat cycad Esbuds was thatnot generate complicated branching by [ambiguities18]. We restricted introduced the by multi-stemselection trees ofand replications sink activity to monopodial of expanding individuals reproductive and did not organs include trees, which with are multiple always branches. positioned at the stemThe apex initiation. These of methods reproductive also structures excluded on dioeciousthe use of cycad the treestallest is infrequent individuals and wein each did not locality include in order to enableany appropriate trees with developing replication. reproductive structures. These methods ensured that this first look at cycad E was not complicated by ambiguities introduced by multi-stem trees and sink activity of expanding In situs Cycas micronesica trees were studied from 5 to 10 Feb 2018 on Yap island (Figure 4). The reproductive organs, which are always positioned at the stem apex. These methods also excluded the locality wasuse of a theclosed tallest-canopy individuals forest in each growing locality on in orderacid toschist enable soils appropriate that ha replication.ve been the subject of several investigationsIn situ[31–Cycas34]. Stem micronesica heighttreess of were the studied replications from 5 tow 10ere Feb 405 2018–415 on cm. Yap islandSlope (Figure, as determined4). The by a clinometerlocality, was was 21° a to closed-canopy 32°. In situ forest Cycas growing nitida ontrees acid were schist studied soils that from have been 10 to the 14 subject Mar of2019 several in a habitat located investigations on a small [ barrier31–34]. Stem island heights in of Northwestern the replications Samarwere 405–415 near cm. the Slope, San as Bernardino determined by Strait. a The clinometer, was 21 to 32 . In situ Cycas nitida trees were studied from 10 to 14 Mar 2019 in a habitat population was growing◦ 20–30◦ m from the shore as understory components of littoral forests in sand located on a small barrier island in Northwestern Samar near the San Bernardino Strait. The population substratewas with growing karst 20–30 outcrops m from and the shore had asbeen understory the subject components of one of investigation littoral forests in [ sand35]. substrateStem height with s of the replicationskarst w outcropsere 415 and–42 had5 cm been and the slope subject was of one 2° investigation to 4°. In situ [35 ].Cycas Stem heightszambalensis of the replicationstrees were were studied on 27 Apr 415–425 2019 in cm San and slopeAnton wasio, 2 ◦ Zambales,to 4◦. In situ Cycas Philippines. zambalensis Thetrees population were studied on was 27 Apr growing 2019 in inSan full sun conditionsAntonio, in ultramafic Zambales, soils. Philippines. Stem Theheight populations of the was replications growing in fullwere sun 405 conditions–415 cm in ultramaficand slope soils. was 34° to Stem heights of the replications were 405–415 cm and slope was 34 to 46 . In situ Cycas wadei trees 46°. In situ Cycas wadei trees were studied on 4 May 2019 within the◦ only◦ known endemic population were studied on 4 May 2019 within the only known endemic population on Culion Island, Philippines. on CulionThe plants Island, studied Philippines. were growing The in full plants sun conditions studied in impoverishedwere growing soils and in the full locality sun has conditions been in impoverishedthe subject soils of and one investigationthe locality [ 36has]. Stembeen heights the subject of the replicationsof one investiga were 355–365tion [3 cm6]. and Stem slope hei wasghts of the replications5◦ to were 17◦. 355–365 cm and slope was 5° to 17°.

FigureFigure 4. Map 4. Map of the of the study study locations locations in in western western Pacific Pacific islands. islands.

Cycas edentataCycas edentata and Cycasand Cycas revoluta revoluta gardengarden trees trees in thethe Angeles Angeles City, City, Philippines Philippines urban landscapeurban landscape were studied from 16 to 24 Dec 2019. These two cycad species have long been favored specimens for were studiedmunicipal from landscapes 16 to 24 inDec the 2019. Philippines, These where two theycycad are species used at Cityhave Hall, long Barangay been favored Hall, and specimens school for municipalbuildings. landscapes Old trees in the with Philippines substantial heights, where are they easily are located used in at most City municipalities. Hall, Barangay The nativeHall, soiland school buildings.is aOld neutral trees coarse with loam, substantial but the landscape heights substrates are easily had located been heavily in most modified. municipalit The C. edentataies. Thestem native soil is a neutralheights coarse were loam, 355–370 but cm the and landscape the C. revoluta substratesstem heights had were been 300–315 heavily cm. modified. Angeles City The is positioned C. edentata stem heights wereon a plain, 355– so370 slope cm wasand not the measurable C. revoluta for stem any ofheights the replications. were 300–315 cm. Angeles City is positioned Stem vertical strata categories for Es measurements were determined in a fixed manner for all six on a plain,replications so slope of was all six not species. measurable The lowest for stratum any of was the fixed replicat at 30 cmions. height, then vertical increments of

Plants 2020, 9, x FOR PEER REVIEW 8 of 10

Plants Stem2020, 9 ,vertical 230 strata categories for Es measurements were determined in a fixed manner for8 of all 10 six replications of all six species. The lowest stratum was fixed at 30 cm height, then vertical increments of 50 cm were included, beginning at a height of 50 cm, and continued until the ultimate 50height cm wereof each included, tree. The beginning maximum at a height stratum of for 50 cm,stem and measurement continued untils was the adjusted ultimate to height the posi of eachtion tree.directly The below maximum the living stratum petioles for stem of measurements the leaf crown wasfor adjusted each tree to. This the positionapproach directly generated below fixed the livingvertical petioles strata for of theevery leaf height crown category for each tree.except This the approach tallest height generated category fixed, which vertical varied strata among for every the heightreplications category and exceptspecies the. tallest height category, which varied among the replications and species. A CIRAS EGM-4EGM-4 analyzer fitted fitted with a SRC-1SRC-1 closedclosed system chamber (PP Systems, Amesbury, Amesbury, MA,MA. U.S.A.) USA) was was used used to to quantify quantify the the e effluxfflux of of carbon carbon dioxide dioxide fromfrom stemstem surfacessurfaces atat each pre-definedpre-defined stratum. The use of a horizontally-oriented soil chamber has been previously reported for tree Es stratum. The use of a horizontally-oriented soil chamber has been previously reported for tree Es measurements [[1717,,3737,38,38].]. AA ringring ofof modeling modeling clay, clay approximately, approximately 10 10 cm cm in diameter,in diameter was, was placed placed on theon stemthe stem surface surface to form to form a malleable a malleable seal, thenseal, thethen SRC-1 the SRC chamber-1 chamber was inserted was inserted into the into modeling the modeling clay to provideclay to provide a sealed a chamber sealed chamber of 1.171 Lof volume 1.171 L (Figure volume5). (Figure The use 5) of. The modeling use of claymodeling to form clay sealed to form gas exchangesealed gas chambers exchange onchambers tree stems on hastree been stems previously has been previously reported [17 reported,39]. [17,39].

Figure 5. Depiction ofof thethe infraredinfraredgas gas exchange exchange instrument instrument (left) (left) and and chamber chamber in in use use near near the the base base of aofCycas a Cycasstem stem with with red red malleable malleable modeling modeling clay clay used used for for seal. seal.

The EGM-4EGM-4 recordedrecorded air air temperature temperature and and the the increase increase in carbonin carbon dioxide dioxide above above ambient ambient for afor 2 min a 2 period.min period. The changeThe change in carbon in carbon dioxide dioxide was used was to calculateused to calculateEs. Three E periodss. Three of periods efflux were of efflux conducted were forconducted each measurement for each measurement and the mean andE sthewas mean used E ass was the used value as for the each value stem for height each stem for each height replication. for each Thereplication stem surface. The stem temperature surface wastemperature measured was with measured an infrared with thermometer an infrared (Milwaukee thermometer Model (Milwaukee 2267-20, MilwaukeeModel 2267 Tool,-20, Milwaukee Brookfield, Tool, WI, USA). Brookfield, Relative WI, humidity USA). Relative was determined humidity with was a slingdetermined psychrometer. with a Stemsling psychrometer. diameter at the Stem height diameter of each at flux the measurement height of each and flux total measurement stem height and were total recorded. stem height The were time ofrecorded. day was The limited time to of 10:00–14:00 day was limi to restrictted to 1 time0:00– of14 day:00 to to restrict the middle time of of the day photoperiod to the middle as the of diel the cyclephotoperiod of Es is notas the known diel cycle for cycads. of Es is not known for cycads. The Es datadata for for each each replication replication were were plotted plotted in in scatter scatter plots plots to toreveal reveal the the general general relationship relationship of ofEs toEs verticalto vertical strata strata.. These These plots plots revealed revealed the approximate the approximate fit of afit quadratic of a quadratic relationship, relationship, so the general so the generallinear m linearodel (Proc model GLM, (Proc SAS GLM, Institute, SAS Institute, Cary, NC, Cary, USA) NC, was USA) employed was employed to fit the to quadratic fit the quadratic models modelsfor each forreplication. each replication. Then, one Then, regression one regression was conducted was conducted for each species for each to species determine to determine one universal one universalquadratic quadratic model. We model. also used We also the used Proc the GLM Proc to GLM determine to determine the linear the linear relationship relationship between between stem stemdiameter diameter and E ands. Es.

AuthorAuthor Contributions: Contributions: Conceptualization,Conceptualization T.E.M.;, T.E.M. methodology,; methodology, T.E.M.; T.E.M. software,; software, T.E.M.; T.E.M. formal; formal analysis, analysis, T.E.M.; investigation,T.E.M.; investigation, T.E.M.; resources, T.E.M.; resources,T.E.M. and M.V.K;T.E.M. writing—original and M.V.K; writing draft— preparation,original draft T.E.M.; preparation, writing—review T.E.M.; and editing, M.V.K.; project administration, T.E.M.; funding acquisition, T.E.M. All authors have read and agreed to the published version of the manuscript.

Plants 2020, 9, 230 9 of 10

Funding: This research was funded by United States Forest Service Cooperative Agreement number 17-DG-11052021-217 and United States Department of Agriculture NIFA Grant number 2013-31100-06057. Acknowledgments: M.V.K. thanks the College of Micronesia land grant program for their support. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Yang, J.; He, Y.; Aubrey, D.P.; Zhuang, Q.; Teskey, O. Global patterns and predictors of stem CO2 efflux in forest ecosystems. Glob. Chang. Biol. 2016, 22, 1433–1444. [CrossRef][PubMed] 2. Vargas, R.; Barba, J. Greenhouse gas fluxes from tree stems. Trends Plant Sci. 2019, 24, 296–299. [CrossRef] [PubMed]

3. Bloemen, J.; McGuire, M.A.; Aubrey, D.P.; Teskey, R.O.; Steppe, K. Transport of root-respired CO2 via the transpiration stream affects aboveground carbon assimilation and CO2 efflux in trees. New Phytol. 2013, 197, 555–565. [CrossRef][PubMed]

4. Kunert, N. A case study on the vertical and diurnal variation of stem CO2 effluxes in an Amazonian forest tree. Trees 2018, 32, 913–917. [CrossRef] 5. Tarvainen, L.; Wallin, G.; Lim, H.; Linder, S.; Oren, R.; Ottosson Löfvenius, M.; Räntfors, M.; Tor-ngern, P.;

Marshall, J. Photosynthetic refixation varies along the stem and reduces CO2 efflux in mature boreal Pinus sylvestris trees. Tree Physiol. 2018, 38, 558–569. [CrossRef]

6. Martin, T.A.; Teskey, R.O.; Dougherty, P.M. Movement of respiratory CO2 in stems of loblolly pine (Pinus taeda L.) seedlings. Tree Physiol. 1994, 14, 481–495. [CrossRef] 7. Bowman, W.P.; Barbour, M.M.; Turnbull, M.H.; Tissue, D.T.; Whitehead, D.; Griffin, K.L. Sap flow rates and

sapwood density are critical factors in within- and between-tree variation in CO2 efflux from stems of mature Dacrydium cupressinum trees. New Phytol. 2005, 167, 815–828. [CrossRef]

8. McGuire, M.A.; Cerasoli, S.; Teskey, R.O. CO2 fluxes and respiration of branch segments of sycamore (Platanus occidentalis L.) examined at different sap velocities, branch diameters, and temperatures. J. Exp. Bot. 2007, 58, 2159–2168. [CrossRef]

9. Saveyn, A.; Steppe, K.; Lemeur, R. Daytime depression in tree stem CO2 efflux rates: Is it caused by low stem turgor pressure? Ann. Bot. 2007, 99, 477–485. [CrossRef] 10. Hilman, B.; Muhr, J.; Trumbore, S.E.; Kunert, N.; Carbone, M.S.; Yuval, P.; Wright, S.J.; Moreno, G.;

Pérez-Priego, O.; Migliavacca, M.; et al. Comparison of CO2 and O2 fluxes demonstrate retention of respired CO2 in tree stems from a range of tree species. Biogeosciences 2019, 16, 177–191. [CrossRef] 11. Sprugel, D.G. Components of woody-tissue respiration in young Abies amabilis (Dougl.) Forbes trees. Trees 1990, 4, 88–98. [CrossRef] 12. Damesin, C.; Ceschia, E.; Le Goff, N.; Ottorini, J.-M.; Dufrêne, E. Stem and branch respiration of beech: From tree measurements to estimations at the stand level. New Phytol. 2002, 153, 159–172. [CrossRef] 13. Ryan, M.G.; Hubbard, R.M.; Pongratic, S.; Raison, R.J.; McMurtrie, R.E. Foliage, fine-root, woody-tissue and stand respiration in Pinus radiata in relation to nitrogen status. Tree Physiol. 1996, 16, 333–343. [CrossRef]

14. Cavaleri, M.A.; Oberbauer, S.F.; Ryan, M.G. Wood CO2 efflux in a primary tropical rain forest. Glob. Chang. Biol. 2006, 12, 2442–2458. [CrossRef] 15. Katayama, A.; Kume, T.; Komatsu, H.; Ohashi, M.; Matsumoto, K.; Ichihashi, R.; Kumagai, T.O.; Otsuki, K.

Vertical variations in wood CO2 efflux for live emergent trees in a Bornean tropical rainforest. Tree Physiol. 2014, 34, 503–512. [CrossRef][PubMed]

16. Tarvainen, L.; Räntfors, M.; Wallin, G. Vertical gradients and seasonal variation in stem CO2 efflux within a Norway spruce stand. Tree Physiol. 2014, 34, 488–502. [CrossRef][PubMed]

17. Marler, T.E. Stem CO2 efflux of Cycas micronesica is reduced by chronic non-native insect herbivory. Plant Signal. Behav. 2020, 15, e1716160. [CrossRef] 18. Norstog, K.J.; Nicholls, T.J. The Biology of the Cycads; Cornell University Press: Ithaca, NY, USA, 1997. 19. Niklas, K.J.; Cobb, E.D.; Marler, T. A comparison between the record height-to-stem diameter allometries of pachycaulis and leptocaulis species. Ann. Bot. 2006, 97, 79–83. [CrossRef] 20. Fosberg, A.S. Structure and growth of the shoot apex of Cycas revoluta. Am. J. Bot. 1939, 26, 372–385. 21. Fosberg, A.S. Further studies on zonal structure and growth of the shoot apex of Cycas revoluta Thunb. Am. J. Bot. 1940, 27, 487–500. Plants 2020, 9, 230 10 of 10

22. Stevenson, D.W. Radial growth in the Cycadales. Am. J. Bot. 1980, 67, 465–475. [CrossRef] 23. Fisher, J.B.; Lindström, A.; Marler, T. Tissue responses and solution movement after stem wounding in six Cycas species. HortScience 2009, 44, 848–851. [CrossRef] 24. Terrazas, T. Origin and activity of successive cambia in Cycas (Cycadales). Am. J. Bot. 1991, 78, 1335–1344. [CrossRef] 25. Marler, T.E.; Lindström, A.; Fisher, J.B. Stem tissue dimensions correlate with ease of horticultural management for six Cycas species. HortScience 2010, 45, 1293–1296. [CrossRef]

26. Steppe, K.; Saveyn, A.; McGuire, M.A.; Lemeur, R.; Teskey, R.O. Resistance to radial CO2 diffusion contributes to between-tree variation in CO2 efflux of Populus deltoides stems. Funct. Plant Biol. 2007, 34, 785–792. [CrossRef]

27. Ryan, M.G.; Cavaleri, M.A.; Almeida, A.C.; Penchel, R.; Senock, R.S.; Stape, J.L. Wood CO2 efflux and foliar respiration for Eucalyptus in Hawaii and Brazil. Tree Physiol. 2009, 29, 1213–1222. [CrossRef] 28. Langdon, L.M. Stem anatomy of Dioon spinulosum. Bot. Gaz. 1920, 70, 110–120. [CrossRef] 29. Chamberlain, C.J. Gymnosperms: Structure and Evolution; University Chicago Press: Chicago, IL, USA, 1935.

30. Salomón, R.L.; Steppe, K.; Crous, K.Y.; Noh, N.J.; Ellsworth, D.S. Elevated CO2 does not affect stem CO2 efflux nor stem respiration in a dry Eucalyptus woodland, but it shifts the vertical gradient in xylem [CO2]. Plant Cell Environ. 2019, 42, 2151–2164. 31. Marler, T.E.; Krishnapillai, M.V. Cycas micronesica trees alter local soil traits. Forests 2018, 9, 565. [CrossRef] 32. Marler, T.E.; Krishnapillai, M.V. Does plant size influence leaf elements in an arborescent cycad? Biology 2018, 7, 51. [CrossRef] 33. Marler, T.E.; Krishnapillai, M.V. Distribution of elements along the rachis of Cycas micronesica leaves: A cautionary note for sampling design. Horticulturae 2019, 5, 33. [CrossRef] 34. Marler, T.E.; Krishnapillai, M.V. Incident light and leaf age influence leaflet element concentrations of Cycas micronesica trees. Horticulturae 2019, 5, 58. [CrossRef] 35. Marler, T.E.; Ferreras, U.F. Disruption of leaf nutrient remobilization in coastal Cycas trees by tropical cyclone damage. J. Geogr. Nat. Disasters 2015, 5, 2. [CrossRef] 36. Marler, T.E.; Ferreras, U.F. Current status, threats, and conservation needs of the endemic Cycas wadei Merrill. J. Biodivers Endanger. Species 2017, 5, 193. [CrossRef] 37. Xu, M.; DeBiase, T.A.; Qi, Y. A simple technique to measure stem respiration using a horizontally oriented soil chamber. Can. J. For. Res. 2000, 30, 1555–1560. [CrossRef]

38. Zhao, G.; Liu, G.; Zhu, W.; Zhao, J.; Wang, X.; Wang, Y.; Jia, M. Stem CO2 efflux of Abies fabri in subalpine forests in the Gongga Mountains, Eastern Tibetan Plateau. J. Plant Ecol. 2017, 10, 1001–1011. 39. Ávila-Lovera, E.; Tezara, W. Water-use efficiency is higher in green stems than in leaves of tropical tree species. Trees 2018, 32, 1547–1558. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).