rotary drum washer | henan deya machinery co., ltd

rotary drum washer | henan deya machinery co., ltd

Rotary drum washer, or drum washing machine is mainly used for washing more than 5 mm of various aggregates such as gravel, pebbles, coal, ceramic particles, all kinds of ore, etc., it reduces mud content with excellent efficiency, water and stone are separated completely, it effectively improves the strength of concrete and reduce the ore mud content.

Rotary Drum Washers are used to wash coarse sand or crushed stone and gravel with a maximum size of 100mm and to dewater the cleaned material sufficiently so it can be conveyed to storage. They are also used, in some installations to assist with the removal of lignite, mica, bark, leaves and trash. They are manufactured in different types while has function of separation at discharge side. The drum normally is equipped with manganese steel liners or alumina liners.

pumps

pumps

Bulk Oil Systems Lincoln pump 3:1 and 5:1 ratio packages include 600003 regulator, 600401 gauge, 815 coupler and 11659 nipple. Control valve, inline meter or handheld metering control valve must be ordered separately (except model 4283).

Manual Bucket and Filler Pumps Ideal for applications requiring portability and economy, Lincoln manual bucket pumps dispense lubricants with up to 5000 psi of operating pressure. Clamp the pump to any 25-50 lb... refinery container or select from models that include 30 or 40 lb. bulk containers. Use filler pumps to quickly fill grease guns and oil cans used in everyday operation.

major mines & projects | la encantada mine

major mines & projects | la encantada mine

Silver, lead and zinc oxide and sulphide mineralization at La Encantada occurs in vein, manto, chimney, breccia pipe, and irregular replacement and stockwork deposits. In general, shallower veins, mantos and breccias are oxidized whereas deeper mantos, skarn dissemination and stockworks contain primary sulphides. Most mining at La Encantada has been done in the oxidized mineral deposits and only some drilling and limited underground access has been done in the deposits with primary sulphides. Drilling indicates potential for deep seated disseminated or massive sulphide replacement mineralization.Structural controls on mineralization include intrusive contacts, faults, fold axes, fractures, fissures, and cavern zones. Intrusive contacts and intrusion-related faults are the most important controls in the skarn, whereas regional faults, folds, and fracture systems are dominant controls on mantos, chimneys and veins.Mineral deposition at La Encantada is recognized across a vertical extent of at least 500 metres. Primary sulphides generally occur 400 metres below surface at the skarn dome area (La Prieta) and Milagros area. Sulphide mineralization consists primarily of sphalerite, galena, pyrite and acanthite.In terms of volume, the mostimportant mineral depositsthat occur at La Encantada are mineralized tectonic breccias and breccia pipes. Skarn, hornfels and marble are developed at depth at the contact with the main stocks (Skarn dome and Milagros areas) and often contain sulphide mineralization. The La Encantada deposit is a typical example of a high temperature carbonate replacement deposit.The La Encantada property is located in the northern part of Coahuila within the Sierra Madre Oriental fold and thrust belt. The Sierra Madre Oriental extends in a south-southeasterly direction about 1,500 km, from the U.S.- Mexican border to approximately latitude 20N in the south. The geologic history initiated in the Permian-Triassic with the Ouachita-Marathon orogenic event, followed by Late Triassic to Middle Jurassic rifting of Pangaea, subsequent opening of the Gulf of Mxico and the formation of the Sabinas basin, passive-margin development through the Late Cretaceous, and Laramide foreland deformation and emplacement of intrusions through the early to mid-TertiaryLa Encantada lies on the southwestern flank of the NW-trending Sierra de La Encantada anticlinorium. The mines occur along a series of NE-trending faults and fractures that cut obliquely across the regional N-NW-trending anticlinorium. The most important ore-controlling structures appear to be the NE-trending faults and fractures that seem to control the localization of chimneys and vein shoots at the intersection with the NW trending structures. Major NW-NNW trending faults such as the main La Encantada front range fault do not appear to be mineralized

Mining the Veins System and other minor deposits at La Encantada is undertaken using primarily the conventional overhand cut-and fill mining method. Ramps are driven into the orebodies, and stopes are developed from sill drifts driven in the ore zones and slashed out the full width of the ore.Mining operations at La Encantada mine are partially mechanized. Drilling of access drifts and ramps is carried out using hydraulic jumbos, and most of the headings and cut-and fill stoping is accomplished using pneumatic hand-held jackleg machines.The cut-and-fill stoping cycle is started with blast holes drilled using hand-held jackleg drills, followed by blasting using conventional mining explosives. After blasting, LHDs are used to muck the blasted ore. The cut and fill stopes range between 50 to 150 m in length along strike, and extend between levels which are typically spaced 15 to 30 m apart vertically. Each cut is 2.5 to 3.0 m in height. Depending on ground conditions, the blast holes are drilled either upward or horizontally. Waste and mineralized material below cut-off grade is blasted down and used as backfill as needed. The minimum mining width is 2.0 m, and planned dilution is included in the mine design, which varies according to the ground conditions, vein width, and the dip of the vein. The dilution factors range from 5% to 20%, with an average of approximately 10%. Mined areas are measured to compare the width of the vein and the width of the cut on a regular basis; as mining advances, this comparison is used as means of reconciliation and to build the historical database of the dilution and mining recovery factors. Sills and access drifts are excavated at 2.5 m wide by 3.0 m high, cross-cuts and access ramps to the stopes are excavated 3.0 m wide by 3.0 m high, and main access ramps are excavated 4.0 m wide by 4.5 m high.Based on the geotechnical characteristics and the geometry of the breccias, First Majestic has started the implementation of a variant of Incined Caving for the San Javier and Milagros breccias. This configuration allows the extraction of ore by building draw-points at different elevations, starting from the outside of the deposits and working inwards as the lower levels are developed.Based on the constraining shapes generated for resource constraining, two main blocks have been identified, the first to be extracted is a block for the San Javier breccia with bottom elevation 1,740 masl. A second block has been delimited in the Milagros breccia with bottom elevation 1,550 masl.A review was carried out to assess the applicability of the possible mining methods, including caving and non-caving methods, to mine the Ojuelas deposit. Their applicability was scored using the University of British Columbia mining method selection matrix, or UBC method, to narrow down the potential methods that are appropriate given the characteristics of the orebody and country rocks.Based on a selection matrix, Block Caving, Sublevel Caving and Cut and Fill mining methods were the top three methods identified, in that order. Due to the geometry of the Ojuelas orebody, applying a standalone block caving or sublevel caving method could negatively impact the economics of the project, therefore, a combination of inclined caving and block caving methods is proposed as one of the methods to be further evaluated in mining the Ojuelas deposit. The upper zone, which has a moderate dipping and thin mineralization zone, will be mined with the incline caving method, whereas the lower portion that has a flatter dipping and thicker mineralization will be mined with a block caving method. The mining will be done in a top-down fashion.Inclined caving uses block caving principles with a truncated sub-level caving mining layout. The cave is initiated by undercutting the orebody at the inclined footwall area. The caving is advanced downward. The principles of gravity flows and draw interactions applied in block caving and sublevel caving are also applied in this mining method. Once all of the ore in the sublevels has been extracted, the block cave will be initiated at the bottom to remove the remainder of the reserve.

The installed plant capacity of the processing plant is 3,000 tpd for the crushing/grinding area. The processing plant (Plant No. 1) consists of three-stage crushing and ball mill grinding circuits.ROM material delivered from the mine is dumped into a steel-made coarse ore bin of 300 tonne capacity. The coarse ore bin is equipped with a steel rail grizzly in its upper part. The grizzly has openings of 12 x 12; oversize material is reduced in size using a hydraulic hammer.The coarse ore bin has a lower discharge chute that discharges into a vibrating feeder grizzly of 4 opening. The -12 + 4 material is fed into a 24 x 36 primary jaw crusher, and reduced to a minus 3 to 3-. This product is transported by a 30 wide belt conveyor to the two primary vibrating screens. These screens have only one sieve with an aperture of 3/8 x 3/8. The lower discharge of the screens contains material from 80 to 90% minus 1/4 (6,350 m).The grinding section is comprised of two ball mill circuits: a recently installed Metso ball mill and the two ball mills previously used, which are currently maintained as back-up. Current grinding capacity is of 125 tpd. The dimensions and sizes of the equipment are: - 12 x 24 Metso Mill (1800 HP)- D-26 Krebs cyclones, and- 10x8 250 HP Envirotech pumps- two back-up ball mills 9-1/2 x 11 and 9 x 12

The mill at the La Encantada cyanidation plant processes silver minerals through a leaching method producing dor bars. The installed plant capacity of the processing plant is 3,000 tpd for the crushing/grinding area, and 4,500 tpd for the cyanidation circuit. The processing plant flowsheet consists of three-stage crushing, ball mill grinding, a leaching circuit, and a Merrill-Crowe system, followed by smelting of the precipitate and final tailings filtration.The processing plant is divided into two parts: Plant No. 1 and Plant No. 2. Plant No. 1 is comprised of crushing and grinding circuits, while Plant No. 2 is comprised of leach tanks, countercurrent tanks (CCD), Merrill-Crowe section, smelting furnace, and tailings filtration.The product of the grinding circuit (Plant No. 1) is pumped to the Plant No. 2 (8X6 200 HP Envirotech pump) and fed into the primary thickener.The following reagent dosages are added to the process:- Cya ........

major mines & projects | simberi mine

major mines & projects | simberi mine

The Simberi Gold Project is located on Simberi Island in the Tabar Islands Group situated in the New Ireland Province of Papua New Guinea (PNG), approximately 80 km north-west of Lihir Island. Simberi is the oldest and northernmost island of the Tabar Group. It measures approximately 10 km east-west, 8 km north-south and rises to over 300 m above sea level. The currently known gold prospects (Sorowar, Pigiput, Pigibo, Botlu, Pigicow, Samat, Bekou and Monun Creek) on Simberi Island are located in the eastern half of the island within the central volcanic core. They are contained within a sub-cropping epithermal alteration system and structural corridor extending 4km north-south and 2km east-west. The host rocks for the mineralisation comprise Pliocene altered alkaline lava flows or intrusives (porphyries), volcaniclastics and tuffs.Of the eight separate deposits, Pigiput in the south is by far the largest gold resource. Monun Creek is located immediately to the north-east of Pigiput, with Sorowar, the second largest resource, further north again. Pigibo, Botlu, Samat and Bekou lie to the west and south of Pigiput, and while relatively small, are relatively higher grade. All deposits lie within 2 to 3km of each other. Sorowar, Pigiput and Botlu are currently being mined via open pit methods. Fine grained free gold in oxide material is the target of current operations. Within the sulphide zone gold is also fine grained (most grains are under 15 m in diameter) but is generally within pyrite. Modifications are required to the current processing plant to allow flotation of pyrite and recovery of the gold. The grade of the mineralisation is related to the natural porosity and degree of fracturing of the host rocks, greatest in the vicinity of steep and moderately dipping feeder structures interpreted to have been the pathways for both alteration and mineralising fluids.Gold does have lithological and structural controls, but these controls are complex and cannot be easily used to generate domains for resource estimation. Leapfrog software was used to generate a 0.25 g/t Au grade shell for resource estimation. A grade shell is needed to avoid smearing grades between mineralized and essentially unmineralized areas. This grade shell is sufficiently below the resource reporting cut-offs to not introduce any significant conditional bias during resource estimation.Oxidation domains (oxide, transitional and sulphide (fresh)) are based on logging from drill holes. The domains were defined in Leapfrog by a combination of offset surfaces and solids. The offset surfaces utilise the logging and depth below topography to create geologically realistic oxidation surfaces. The modelling of oxidation solids in Leapfrog was also required because a single oxidation surface could not always model the observed spatial complexity. For example, there can be pods of oxide completely enclosed by sulphide and vice versa.Based on statistical analysis and contact plots the 0.25 g/t Au grade shell was subdivided into two zones for the estimation of gold grades; a combined oxide and transitional zone and a sulphide only zone.The northernmost deposit is Sorowar, its bulk is aligned SE-NW (1,550 m) with minor (structurally controlled) orthogonal splays towards the southwest and northeast. These splays are less than 750 m long and 300 m wide.Pigibo is oriented W-E for approximately 740 m with a central bulge about 300 m wide and tapering to about 100 m at the western and eastern extremities. It is located about 1,500 m to the southwest of the central part of Sorowar.Pigiput is east of Pigibo and about 1000 m south of Sorowar. It is roughly equidimensional (640 m diameter) in plan.Monun Creek is between Pigiput and Sorowar however, there is now enough drilling to define continuous mineralisation between Pigiput and Sorowar.Botlu is about 800 m south of Pigibo. It strikes SE-NW for approximately 680 m with an average width of around 250 m. About 700 m to the SE of Botlu is the discontinuous Pigicow deposit which strikes SW-NE for nearly 600 m with a variable width (200-450 m).Samat is located about 700 m to the southeast of Pigicow and is aligned north-south for approximately 720 m with an average width of 300 m. Like Pigicow, Bekou is discontinuous and oriented towards the east-northeast with a strike length of around 600 m. Located about 650 m to the southwest of Samat, its width varies from 40 m to 170 m.

The Simberi Gold (SGCL) operation includes the mining of the Sorowar, Pigiput, Pigibo, Samat, Botlu, Pigicow and Bekou pits which are all located on Simberi Island and in close proximity to the Simberi processing facility. Of the seven separate deposits, Sorowar in the north is by far the largest oxide gold resource. Samat and Bekou lie to the south, and while relatively small, are relatively higher grade. Pigiput, Pigibo, Botlu and Pigicow lie between the Sorowar and Samat areas and have oxide gold resources of intermediate tonnage but at a grade similar to Sorowar. Pigiput has the largest sulphide gold resource. All deposits lie within 2 to 3km of each other. Sorowar, Pigiput and Botlu are currently being mined for oxides via open pit methods.

Material from the various Simberi Deposits is trucked and conveyed to the Simberi processing plant. The Simberi processing plant consists of a Wet Scrubber, Oversize Ball Mill, Semi-Autogenous Grinding Mill, Cyclone Circuit, CIL Circuit, elution and acid washing facilities, electrowinning cells, and Kiln.

Material from the various Simberi Deposits is trucked and conveyed to the Simberi processing plant. The Simberi processing plant consists of a Wet Scrubber, Oversize Ball Mill, Semi-Autogenous Grinding Mill, Cyclone Circuit, CIL Circuit, elution and acid washing facilities, electrowinning cells, and Kiln. Tails from the process are discharged using a Deep Sea Tailings Placement where the tailings is diluted with sea water, to the ratio of 8:1, prior to its disposal. Expansion of the Oxide Circuit was completed towards the end of 2013 and since then, the operation has been operating at 3.5 Mtpa.The technology associated with processing of Simberi ore is conventional carbon in leach and based on industry standard practices. Recovery performance is variable for the different weathering profile of the material. The metal recovery for oxides is calculated by a formula that uses gold grade as a predictor. The gold recovery has an upper limit of 93%The oxid ........

algerian kaolinite used for mullite formation - sciencedirect

algerian kaolinite used for mullite formation - sciencedirect

In the present study, mullite was synthesized through reaction sintering of Algerian kaolinite and high purity alumina. The raw powders were wet ball milled in a planetary ball mill. Powders' morphology and the microstructure of the sintered samples were characterized by means of a scanning electron microscope. An X-ray diffractometer equipped with a heating facility and a differential thermal analyzer were used to follow mullite formation. Cylindrical specimens were produced by uniaxial cold compaction at a pressure of 75MPa and sintered at different sintering temperatures for different sintering times. The heating rate was 10C/min. It was found that Algerian kaolinite was suitable for mullite production through reaction sintering with pure Al2O3. Formation of complete mullite occurred at 1550C. A relative density of 94% (of the theoretical density) was achieved at a relatively low sintering temperature of 1600C and a sintering time of 4h.

multi-proxy assessment of dieback causes in a mediterranean oak species | tree physiology | oxford academic

multi-proxy assessment of dieback causes in a mediterranean oak species | tree physiology | oxford academic

Michele Colangelo, J. Julio Camarero, Giovanna Battipaglia, Marco Borghetti, Veronica De Micco, Tiziana Gentilesca, Francesco Ripullone, A multi-proxy assessment of dieback causes in a Mediterranean oak species, Tree Physiology, Volume 37, Issue 5, May 2017, Pages 617631, https://doi.org/10.1093/treephys/tpx002

Drought stress causes forest dieback that is often explained by two interrelated mechanisms, namely hydraulic failure and carbon starvation. However, it is still unclear which functional and structural alterations, related to these mechanisms, predispose to dieback. Here we apply a multi-proxy approach for the characterization of tree structure (radial growth, wood anatomy) and functioning (13C, 18O and non-structural carbohydrates (NSCs)) in tree rings before and after drought-induced dieback. We aim to discriminate which is the main mechanism and to assess which variables can act as early-warning proxies of drought-triggered damage. The study was tailored in southern Italy in two forests (i.e., San Paolo (SP) and Oriolo (OR)) where declining and non-declining trees of a ring-porous tree species (Quercus frainetto Ten.) showing anisohydric behavior coexist. Both stands showed growth decline in response to warm and dry spring conditions, although the onset of dieback was shifted between them (2002 in SP and 2009 in OR). Declining trees displayed a sharp growth drop after this onset with reductions of 49% and 44% at SP and OR sites, respectively. Further, contrary to what we expected, declining trees showed a lower intrinsic water-use efficiency compared with non-declining trees after the dieback onset (with reductions of 9.7% and 5.6% at sites SP and OR, respectively), due to enhanced water loss through transpiration, as indicated by the lower 18O values. This was more noticeable at the most drought-affected SP stand. Sapwood NSCs did not differ between declining and non-declining trees, indicating no carbon starvation in affected trees. Thus, the characterized structural and functional alterations partially support the hydraulic failure mechanism of dieback. Finally, we show that growth data are reliable early-warning proxies of drought-triggered dieback.

Drought and heat stress cause forest dieback worldwide (Allen et al. 2010). However, a multi-proxy picture considering several structural and functional variables (growth, wood anatomy and water-use efficiency (WUE)) to infer how dieback occurs is still lacking. In the Allen et al. (2010) review, only 11% of the dieback cases involved ring-porous oak species with anisohydric behavior, i.e., showing large declines in leaf midday water potential during drought, which are supposed to better able to survive droughts under the forecasted warmer conditions than taller conifer isohydric species (McDowell and Allen 2015). When there is no water restriction, anisohydric species maintain higher stomatal conductance (g) and photosynthesis (A) rates than isohydric tree species, tracking changes in vapor pressure deficit (VPD) and showing little dependence on changes in soil moisture (Klein 2014). However, under intense or prolonged drought, anisohydric species are prone to xylem cavitation, while isohydric species rapidly close stomata (Klein 2014, Attia et al. 2015). If warmer conditions enhance atmospheric evaporative demand moving VPD toward a critical threshold, then drought stress could be magnified and trigger dieback, making anisohydric species vulnerable (Williams et al. 2013). In the case of ring-porous Mediterranean oak species, extreme droughts may surpass the limit of hydraulic safety at which they often function (Tognetti et al. 1998).

Drought-induced dieback is explained by two non-mutually exclusive mechanisms: (i) hydraulic failure due to a drastic loss of xylem conductivity and (ii) carbon starvation when carbon demands are not met (McDowell et al. 2008, 2011; McDowell 2011). The multiple links between these mechanisms are mediated by structural and functional changes affecting growth, wood anatomy and WUE. Regarding structural changes, wood anatomy is related to the occurrence of hydraulic failure and shoot dieback since wide xylem vessels are more prone to cavitation and loss of hydraulic conductivity than narrow vessels (Nardini et al. 2013). Cavitation resistance is linked to a reduction in lumen area or the mechanical reinforcement of fibers that prevent xylem embolism and drought damage (Hacke et al. 2001). In oaks, a reduction in lumen size of earlywood (EW) vessels and a sharp decrease in latewood (LW) production have been described as wood-anatomical responses to tolerate drought (Corcuera et al. 2004b, Eilmann et al. 2009). However, anisohydric oak species also show extensive xylem embolism if water shortage is prolonged (Hoffmann et al. 2011).

Altered structural conditions also include a decline in radial growth and abrupt shifts in climategrowth associations (Pellizzari et al. 2016). For instance, in several oak species, a growth decline and the weakening of climategrowth relationships started in response to severe droughts 2050 years prior to dieback onset (Dwyer et al. 1995, Drobyshev et al. 2007, Wyckoff and Bowers 2010, Haavik et al. 2011, Gea-Izquierdo et al. 2014, Stojanovi et al. 2015). Therefore, the relationships between dieback, growth and wood anatomy may also depend on functional traits affecting water use, and carbon uptake and storage (McDowell et al. 2008).

Water scarcity constrains growth and is responsible for loss in xylem conductivity affecting A and g, thus reducing carbon uptake (Hsiao 1973, Brda et al. 2006). Mild droughts impair the transport of assimilates through the phloem and trigger the conversion of stored carbon (e.g., starch) into osmolytes (e.g., soluble sugars (SS)) to maintain turgor pressure (Sala et al. 2010, 2012, Pantin et al. 2013). Contrastingly, severe droughts can impair photosynthesis and lead to carbon starvation and dieback (Sevanto et al. 2014). Shifts in intrinsic water-use efficiency (WUEi), i.e., the A/g ratio, may be inferred through the combined analysis of carbon (13C) and oxygen (18O) isotope compositions in leaves or tree-ring cellulose, the so-called dual-isotope approach (Scheidegger et al. 2000, Barbour et al. 2002, Ferrio and Voltas 2005, Ripullone et al. 2009, but see Roden and Farquhar 2012). Changes in gas exchange are recorded in tree rings through carbon isotope composition (13C), which gives insight into how trees respond to drought (Farquhar et al. 1989, Saurer et al. 2004, Battipaglia et al. 2010). On the other hand, 18O is correlated with soil water sources and g, making it a proxy for leaf transpiration and evaporative demand (Saurer et al. 1997).

Here we argue for a multi-proxy characterization of tree structure (i.e., the analysis of stem radial growth and wood-anatomical functional traits) and functioning (i.e., the determination of 13C, 18O, WUEi, non-structural carbohydrates (NSCs)) to determine how water and carbon limitations drive drought-induced dieback (Voltas et al. 2013, Billings et al. 2016). We explore the likely occurrence of structural and functional alterations in a ring-porous tree species (Quercus frainetto Ten.) (Chatziphilippidis and Spyroglou 2004) that presents ongoing dieback in southern Italy. This species shows mid to high growth rates and it is very sensitive to dry conditions in late spring and summer, and recovery rates after drought are rapid (Sanders et al. 2014). Since the 1980s, drought-induced dieback episodes have been detected in southern Italian oak forests showing a worrisome trend of increasing damage, as in other areas of the Mediterranean Basin (Ragazzi et al. 1989, Salvatore et al. 2002). We compare declining and non-declining Italian oak trees coexisting at two sites to improve our understanding of dieback.

We hypothesize that declining trees should show higher growth rates, wider lumen areas and stronger climategrowth associations prior to the dieback onset since these features make them more susceptible to drought-induced damage (Levani et al. 2011). After the dieback onset, we expect that declining trees would show lower growth rates, smaller lumen areas (i.e., less theoretical hydraulic conductivity) and higher WUEi than non-declining trees (Linares and Camarero 2012). Finally, 18O should increase in declining trees as compared with non-declining trees, indicating an enhanced stomatal control of water loss linked to canopy dieback (Scheidegger et al. 2000).

We selected two oak (Q. frainetto Ten.) forests showing recent dieback, located close to San Paolo Albanese (400120N, 162046E, 950 m above sea level (a.s.l.), mean slope 2530%; hereafter SP site) and Oriolo (400010N, 162330E; 770 m a.s.l., mean slope 25%; hereafter OR site) villages, situated in the Basilicata and Calabria regions (southern Italy), respectively. The SP stand accounts for a density of 348 trees ha1, while average diameter at 1.3 m (d.b.h.) and age are 40 cm and 145 years, respectively. In the OR site, the mean d.b.h. and age are 35 cm and 138 years, while mean density is 444 trees ha1. Both sites are located in high oak forests. The soil in both study sites is formed by sands and clays. No recent disturbance has been reported for the study sites (e.g., insect outbreaks or fires) and no silvicultural treatment has been applied in the last five decades.

The SP and OR study sites have shown dieback symptoms since the early 2000s (shoot dieback, leaf loss and withering, growth decline, high mortality). The densities of dead trees at the SP and OR sites are 35 stems ha1 and 25 stems ha1 (5%), respectively. The most affected areas encompass ca 250 and 600 ha in the SP and OR sites, respectively. The frequency of trees showing crown-transparency levels above the 50% threshold are 80% and 50% of trees in the SP and OR sites, respectively (see Figure S1available as Supplementary Data at Tree Physiology Online). Therefore, we considered that the SP site was more affected by drought than the OR site. Further, given the unavailability of microclimate data for the two sites, we used the percentage of dead and defoliated trees as an indirect measure of drought impact.

Climate in the study areas is Mediterranean, characterized by dry and warm summers (mean June to August precipitation is 79 mm) and wet and mild winters (mean December to February precipitation is 257 mm) with a mean annual temperature of 16.4 C and annual precipitation of 742 mm (data from Oriolo station, 400311N, 162647E, 445 m a.s.l., 19502015 period; see Figure S2 available as Supplementary Data at Tree Physiology Online). The warmest and coldest months are July (average maximum temperature of 33.5 C) and January (average minimum temperature of 4.0 C), respectively, whereas the driest and wettest months are July (22 mm) and December (99 mm). Drought occurs from June to September. Due to the shortness and heterogeneity of some local climate data, we used gridded (0.25 resolution) climate data from the E-OBS data set ver. 13.0 (Haylock et al. 2008) to quantify climate trends and climategrowth associations. Climate was extracted from the 0.25 grid with coordinates 40.0040.25N, 16.2516.50E. To evaluate droughtgrowth associations stress since 1950, we downloaded the Standardized Precipitation Evapotranspiration Index (SPEI) for the 0.5 grid where the study sites are located using the Global SPEI database webpage (http://sac.csic.es/spei/database.html). The SPEI is a multiscalar drought index, which considers the effects of temperature and evapotranspiration on drought severity and indicates wet (positive SPEI values) and dry (negative SPEI values) conditions (Vicente-Serrano et al. 2010).

First, seven circular plots (radius of 15 m) were randomly located in each of the SP and OR sites to describe the stand structure (density, basal area). Within each plot, dieback of all mature oak trees was characterized by a visual assessment of crown transparency made by two independent observations of the same tree (Dobbertin 2005, Camarero et al. 2016). Declining oaks (hereafter D trees) were considered those with crown transparency higher than 50%, whereas non-declining oaks (hereafter ND trees) were considered those with transparency lower than 50%. Using other crown-transparency thresholds (40% and 60%), the main results presented here did not change.

A total of 17 and 16 pairs of dominant and neighboring oak trees with contrasting vigor (NDD couples) were selected at SP and OR sites in summer 2014 to estimate their growth trends using dendrochronology. Within each NDD couple, trees were 1015 m apart at maximum. Four wood cores from each tree were sampled at breast height (1.30 m) using a 5-mm Pressler increment borer. Two cores were used for tree-ring analyses and wood anatomy, another core was used for isotope analyses and the last core for NSCs analysis in the sapwood. A total of 34 and 32 trees were used for tree-ring width measurements at SP and OR study sites, respectively (Table 1). Quantitative wood anatomy was performed on a subsample of 10 cores from each considered site, i.e., five trees for each of the two vigor classes. Table 1.Characteristics of the non-declining (ND) and declining (D) Italian oak trees sampled in the two study sites (OR and SP). Values are means SE. Different letters indicate significant differences (P < 0.05; MannWhitney U tests) between tree types within each considered site. Site .Tree type .No. of trees .D.b.h. (cm) .Height (m) .Age at 1.3 m (years) .SPND1731.9 0.8b11.3 0.5b142 5aD1727.9 1.0a10.2 0.3a144 4aORND1632.1 0.6a10.7 0.3a141 4aD1630.0 0.9a10.1 0.3a140 3aSite .Tree type .No. of trees .D.b.h. (cm) .Height (m) .Age at 1.3 m (years) .SPND1731.9 0.8b11.3 0.5b142 5aD1727.9 1.0a10.2 0.3a144 4aORND1632.1 0.6a10.7 0.3a141 4aD1630.0 0.9a10.1 0.3a140 3a Open in new tab Table 1.Characteristics of the non-declining (ND) and declining (D) Italian oak trees sampled in the two study sites (OR and SP). Values are means SE. Different letters indicate significant differences (P < 0.05; MannWhitney U tests) between tree types within each considered site. Site .Tree type .No. of trees .D.b.h. (cm) .Height (m) .Age at 1.3 m (years) .SPND1731.9 0.8b11.3 0.5b142 5aD1727.9 1.0a10.2 0.3a144 4aORND1632.1 0.6a10.7 0.3a141 4aD1630.0 0.9a10.1 0.3a140 3aSite .Tree type .No. of trees .D.b.h. (cm) .Height (m) .Age at 1.3 m (years) .SPND1731.9 0.8b11.3 0.5b142 5aD1727.9 1.0a10.2 0.3a144 4aORND1632.1 0.6a10.7 0.3a141 4aD1630.0 0.9a10.1 0.3a140 3a Open in new tab

Characteristics of the non-declining (ND) and declining (D) Italian oak trees sampled in the two study sites (OR and SP). Values are means SE. Different letters indicate significant differences (P < 0.05; MannWhitney U tests) between tree types within each considered site.

Characteristics of the non-declining (ND) and declining (D) Italian oak trees sampled in the two study sites (OR and SP). Values are means SE. Different letters indicate significant differences (P < 0.05; MannWhitney U tests) between tree types within each considered site.

To quantify climategrowth relationships, first we removed the long-term trends of tree-ring width series by detrending them through Friedman's super smoother, which preserves high-frequency (yearly) variability in the resulting ring-width indices. In addition, an autoregressive model was applied to each detrended series to remove most of the first-order autocorrelation related to the previous year of growth. We obtained series at the tree level of dimensionless ring-width indices. Finally, a biweight robust mean was used to obtain mean chronologies of ND and D trees at each site. Chronology development and standardization were carried out using the ARSTAN program (Cook and Krusic 2005). Climategrowth associations were calculated for the 195080 period, i.e., prior to the dieback onset, considering monthly and seasonal climatic variables (mean maximum and minimum temperatures, precipitation). The window of analyses included from the previous October to current September based on previous studies (Tessier et al. 1994). Droughtgrowth associations were calculated for the same period using 1- to 20-month-long SPEI values obtained from January to December.

Wood anatomy was analyzed for the 19802013 period. The five trees of each vigor class used in these and further (isotopes, carbohydrates) analyses were selected based on the highest correlations of their ring-width series with the mean chronology of each tree class developed for each study site. Semi-thin transversal sections (20 m) were obtained from one core per tree by dividing it into pieces of ~2 cm length. Sections were cut using a sliding microtome (Microm HM 400, Thermo Sci., Walldorf, Germany) and stained with safranin (1%) and astrablue (2%), dehydrated with ethanol (70%, 95% and 100%) and xylol, and mounted on microscope slides using Eukitt (Sigma-Aldrich, St Louis, MO, USA). Images were captured at 20 and 40 magnification using a transmitted light microscope (Zeiss Axiophot, Carl Zeiss Microscopy, Jena, Germany). Earlywood and LW vessels were analyzed in tangential windows of 2 and 0.3 mm, respectively. Earlywood vessels were considered those with lumen diameters larger than 50 m. Vessel lumen diameter (along the radial direction) and area were measured using the ImageJ software (Schneider et al. 2012). We quantified the following wood-anatomical traits following Scholz et al. (2013): ring area, EW and LW areas, absolute and relative (%) areas occupied by vessels in the EW and LW, EW and LW vessel lumen area (mean, minimum and maximum values), and EW and LW vessel density. We also calculated two additional variables (Dh, hydraulic diameter; Kh, potential hydraulic conductivity) by weighting individual vessel lumen diameters to correspond to the average HagenPoiseuille lumen theoretical hydraulic conductivity for a vessel size (Tyree and Zimmermann 2002). The Dh was calculated as the average of d5/d4, where d is the lumen diameter of each vessel (Sperry et al. 1994). The Kh was estimated as Kh = ( d4)/(128 Ar), where is the density of water at 20 C (998.2 kg m3 at 20 C), d is the vessel lumen diameter, is the viscosity of water (1.002 109 MPa s at 20 C) and Ar is the area imaged (Tyree and Zimmermann 2002). The Dh and the Kh were calculated considering all EW and LW measured vessels. Finally, we obtained the vulnerability index (VI) following Carlquist (1977) by calculating the ratio between the mean vessel lumen diameter and vessel density. The VI is a proxy for the tree resistance to drought- or frost-induced cavitation with low (VI < 1) and high (VI > 3) values indicating xeromorphy and mesomorphy, respectively.

The sapwood fraction of each core from the five trees selected in wood anatomy and isotope analyses was visually determined in the field and separated using a razor blade to determine the concentrations of NSCs. All sapwood samples were collected at the end of the growing season in 2015, transported to the laboratory in a portable cooler, and they were frozen and stored at 20 C until freeze-dried. Then, they were weighed and milled to a fine powder in a ball mill (Retsch Mixer MM301, Leeds, UK). Soluble sugars were extracted with 80% (v/v) ethanol and their concentration determined colorimetrically using the phenol-sulfuric method (Buysse and Merckx 1993). Starch and complex sugars remaining after ethanol extraction were enzymatically digested with an enzyme mixture containing amyloglucosidase to reduce glucose as described in Palacio et al. (2007). Non-structural carbohydrates measured after ethanol extractions are referred to as SS, and carbohydrates measured after enzymatic digestion are referred to as starch. The NSC concentrations (% dry matter) were calculated as the sum of SS and starch concentrations.

The carbon and oxygen isotopes analysis was performed on the same trees used for wood-anatomical analyses (five ND trees plus five D trees per study site). We used rings spanning the period 19812013 to avoid a possible juvenile effect and focused on this most recent period where ND and D performance is expected to differ. Groups of three contiguous rings (e.g., 198183, , 201113; n = 11 samples per tree) from the best five cross-dating trees per site were split using a scalpel and pooled together for each tree. These groups of rings were milled to a fine powder with a mixer mill (Retsch MM301, Haan, Germany). We then proceeded to extract -cellulose on 10 mg of wood per group of three rings to remove extractives and lignin with a double-step digestion (Boettger et al. 2007, Battipaglia et al. 2008).

The carbon and oxygen stable isotope compositions were measured at the CIRCE Laboratory (Center for Isotopic Research on the Cultural and Environmental Heritage, Caserta, Italy) by continuous-flow isotope ratio mass spectrometry (Delta V plus Thermo electron corporation, Bremen Germany). We processed 1 mg of wood obtaining a cellulose yield of 5060% using 0.06 mg for carbon isotope and 0.10.3 mg for oxygen isotope analysis.

Carbon isotope discrimination was used to account for the Suess effect (decrease in 13C of atmospheric CO2 since the beginning of industrialization) resulting from the emission of fossil CO2, which is depleted in 13C (McCarroll and Loader 2004). Stable isotope ratios were expressed as per mil deviations using the notation relative to Vienna Pee Dee Belemnite and Vienna Standard Mean Ocean Water standards in the case of carbon and oxygen isotopes, respectively. The accuracy of the analyses (SD of working standards) was 0.060.20.

Trends in climatic variables were assessed using the Kendall statistic. To compare mean values of the analyzed variables (BAI, wood anatomy, NSCs, isotopes) between the two vigor classes at each site we used the MannWhitney U test. The relationships between growth, wood-anatomical traits and isotope data were evaluated using Pearson correlations. We used the Wilcoxon rank-sum test to check if the changes through time of BAI, wood-anatomical variables or isotope ratios differed between ND and D trees (Hentschel et al. 2014). We chose this non-parametric test because it is robust against deviations from standard distributions and the presence of temporal autocorrelation (Gibbons and Chakraborti 2011).

We found a significant warming trend of minimum and maximum summer and spring temperatures at the study area since 1950 (see Figure S3 and Table S1 available as Supplementary Data at Tree Physiology Online). Dry conditions during the spring growing season occurred in several years (1957, 1975, 1977, 1992, 1995, 1999, 200102 and 201112; see Figure S3 available as Supplementary Data at Tree Physiology Online). The 10-month-long May SPEI, which captures drought severity during the growing season, detected severe droughts in 1957, 1995, 2000, 2002 and 2013 (Figure S4 available as Supplementary Data at Tree Physiology Online).

Growth trends (BAI) of non-declining (empty symbols) and declining (filled symbols) oak trees in the two study sites (SP and OR) from southern Italy. Values are means SE. The gray filled areas indicate the period when BAI of D and ND trees significantly (P < 0.05) differed.

Growth trends (BAI) of non-declining (empty symbols) and declining (filled symbols) oak trees in the two study sites (SP and OR) from southern Italy. Values are means SE. The gray filled areas indicate the period when BAI of D and ND trees significantly (P < 0.05) differed.

The D trees displayed a sharp growth drop after the onset decline comparing with ND trees, with reductions of 49% and 44% at sites SP and OR, respectively. Further, we found that D trees showed significantly lower BAI values than ND trees after 2002 and 2009 at the SP and OR study sites, respectively (Figure 1). Considering only the 19802013 period, when growth, wood anatomy and isotope data were available, BAI showed significantly (P < 0.05) higher values in ND than D trees at both study sites (SP, 3.1 0.2 cm2 vs 2.2 0.1 cm2; OR, 3.0 0.2 cm2 vs 2.5 0.1 cm2; Figure 1).

The D trees from the most affected SP site were the most sensitive to drought stress in terms of growth reduction (Figure S5 available as Supplementary Data at Tree Physiology Online). At both study sites, D trees showed more growth enhancement in response to cool and wet spring conditions than ND trees. Differences between ND and D trees were found for correlations of growth indices with May precipitation (SP site, ND trees, r = 0.20 0.03, D trees, r = 0.32 0.05; OR site, ND trees, r = 0.19 0.03, D trees, r = 0.28 0.04) and for spring mean maximum temperatures (SP site, ND trees, r = 0.32 0.03, D trees, r = 0.41 0.04; OR site, ND trees, r = 0.27 0.03, D trees, r = 0.35 0.04). Consequently, the D trees from SP site showed the highest SPEIgrowth correlation, specifically when considering spring months (May and June) and 2- to 10-month-long scales (Figure S6 available as Supplementary Data at Tree Physiology Online).

Wood-anatomical traits measured in non-declining (empty symbols) and declining (filled symbols) Italian oak trees in the two study sites (SP and OR). The gray filled areas indicate periods when the variables significantly (P < 0.05) differ between D and ND trees according to Wilcoxon tests. Values are means SE and correspond to the 19802013 period.

Wood-anatomical traits measured in non-declining (empty symbols) and declining (filled symbols) Italian oak trees in the two study sites (SP and OR). The gray filled areas indicate periods when the variables significantly (P < 0.05) differ between D and ND trees according to Wilcoxon tests. Values are means SE and correspond to the 19802013 period.

In agreement with the narrower EW vessel lumen area of D trees at the SP site, the Dh and the Kh were lower in D trees compared with ND trees but differences were significant only at the SP site (Table 2). Table 2.Wood-anatomical variables obtained for non-declining (ND) and declining (D) Italian oak trees in the two study sites (SP and OR). The data correspond to the 19802013 period and are presented as mean SE. Abbreviations: Dh, hydraulic diameter; Kh, potential hydraulic conductivity; VI, vulnerability index (ratio between mean vessel diameter and vessel density). Different letters indicate significant differences (for P < 0.05) between ND and D trees within each site (MannWhitney U tests). Site .Tree type .No. of measured vessels .Dh (m) .Kh (kg m1 MPa1 s1) 102 .Earlywood .Latewood .Vessel diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .Vessel lumen diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .SPND3501331.6 3.6b85.7 6.2b248.8 2.8a35.3 1.0a6 1a41.5 2.7a24.4 0.6a10.3 0.4a22 2a1.2 0.2bD4063321.2 2.8a66.9 3.5a246.5 4.3a35.6 1.5a6 1a41.1 3.0a26.1 0.7a12.0 0.8b37 4b0.6 0.1aORND4125311.2 1.6a50.6 4.4a231.2 5.1a31.9 1.3a7 1a33.0 4.5a26.8 0.7a10.6 0.6a19 1a1.4 0.5aD4194309.4 1.2a47.4 4.1a230.0 5.7a30.6 1.0a7 1a32.9 4.8a27.0 0.8a10.6 0.5a19 1a1.4 0.4aSite .Tree type .No. of measured vessels .Dh (m) .Kh (kg m1 MPa1 s1) 102 .Earlywood .Latewood .Vessel diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .Vessel lumen diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .SPND3501331.6 3.6b85.7 6.2b248.8 2.8a35.3 1.0a6 1a41.5 2.7a24.4 0.6a10.3 0.4a22 2a1.2 0.2bD4063321.2 2.8a66.9 3.5a246.5 4.3a35.6 1.5a6 1a41.1 3.0a26.1 0.7a12.0 0.8b37 4b0.6 0.1aORND4125311.2 1.6a50.6 4.4a231.2 5.1a31.9 1.3a7 1a33.0 4.5a26.8 0.7a10.6 0.6a19 1a1.4 0.5aD4194309.4 1.2a47.4 4.1a230.0 5.7a30.6 1.0a7 1a32.9 4.8a27.0 0.8a10.6 0.5a19 1a1.4 0.4a Open in new tab Table 2.Wood-anatomical variables obtained for non-declining (ND) and declining (D) Italian oak trees in the two study sites (SP and OR). The data correspond to the 19802013 period and are presented as mean SE. Abbreviations: Dh, hydraulic diameter; Kh, potential hydraulic conductivity; VI, vulnerability index (ratio between mean vessel diameter and vessel density). Different letters indicate significant differences (for P < 0.05) between ND and D trees within each site (MannWhitney U tests). Site .Tree type .No. of measured vessels .Dh (m) .Kh (kg m1 MPa1 s1) 102 .Earlywood .Latewood .Vessel diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .Vessel lumen diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .SPND3501331.6 3.6b85.7 6.2b248.8 2.8a35.3 1.0a6 1a41.5 2.7a24.4 0.6a10.3 0.4a22 2a1.2 0.2bD4063321.2 2.8a66.9 3.5a246.5 4.3a35.6 1.5a6 1a41.1 3.0a26.1 0.7a12.0 0.8b37 4b0.6 0.1aORND4125311.2 1.6a50.6 4.4a231.2 5.1a31.9 1.3a7 1a33.0 4.5a26.8 0.7a10.6 0.6a19 1a1.4 0.5aD4194309.4 1.2a47.4 4.1a230.0 5.7a30.6 1.0a7 1a32.9 4.8a27.0 0.8a10.6 0.5a19 1a1.4 0.4aSite .Tree type .No. of measured vessels .Dh (m) .Kh (kg m1 MPa1 s1) 102 .Earlywood .Latewood .Vessel diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .Vessel lumen diameter (m) .Vessel area (%) .Vessel density (no. mm2) .VI (m mm2) .SPND3501331.6 3.6b85.7 6.2b248.8 2.8a35.3 1.0a6 1a41.5 2.7a24.4 0.6a10.3 0.4a22 2a1.2 0.2bD4063321.2 2.8a66.9 3.5a246.5 4.3a35.6 1.5a6 1a41.1 3.0a26.1 0.7a12.0 0.8b37 4b0.6 0.1aORND4125311.2 1.6a50.6 4.4a231.2 5.1a31.9 1.3a7 1a33.0 4.5a26.8 0.7a10.6 0.6a19 1a1.4 0.5aD4194309.4 1.2a47.4 4.1a230.0 5.7a30.6 1.0a7 1a32.9 4.8a27.0 0.8a10.6 0.5a19 1a1.4 0.4a Open in new tab

Wood-anatomical variables obtained for non-declining (ND) and declining (D) Italian oak trees in the two study sites (SP and OR). The data correspond to the 19802013 period and are presented as mean SE. Abbreviations: Dh, hydraulic diameter; Kh, potential hydraulic conductivity; VI, vulnerability index (ratio between mean vessel diameter and vessel density). Different letters indicate significant differences (for P < 0.05) between ND and D trees within each site (MannWhitney U tests).

Wood-anatomical variables obtained for non-declining (ND) and declining (D) Italian oak trees in the two study sites (SP and OR). The data correspond to the 19802013 period and are presented as mean SE. Abbreviations: Dh, hydraulic diameter; Kh, potential hydraulic conductivity; VI, vulnerability index (ratio between mean vessel diameter and vessel density). Different letters indicate significant differences (for P < 0.05) between ND and D trees within each site (MannWhitney U tests).

We did not find any significant differences in SS, starch, NSC sapwood concentrations between ND and D trees either at the SP or at the OR site (Table 3). Interestingly, the NSC sapwood concentrations at the most drought affected SP site were significantly lower than at the least affected OR site (U = 46, P = 0.0005) due to differences between sites regarding SS (U = 39, P = 0.0002) but not starch concentrations (U = 135, P = 0.5194). Table 3.Concentrations of NSCs in the sapwood of non-declining (ND) and declining (D) Italian oak trees sampled at the SP and OR study sites. Values are means SE. Different letters indicated significant differences (P < 0.05; MannWhitney U tests) between tree types within each site. Site .Tree type .SS (%) .Starch (%) .NSC (%) .SPND1.45 0.15a2.03 0.19a3.48 0.23aD1.41 0.15a1.73 0.16a3.14 0.15aORND2.09 0.31a2.24 0.21a4.33 0.45aD2.95 0.41a1.94 0.29a4.89 0.39aSite .Tree type .SS (%) .Starch (%) .NSC (%) .SPND1.45 0.15a2.03 0.19a3.48 0.23aD1.41 0.15a1.73 0.16a3.14 0.15aORND2.09 0.31a2.24 0.21a4.33 0.45aD2.95 0.41a1.94 0.29a4.89 0.39a Open in new tab Table 3.Concentrations of NSCs in the sapwood of non-declining (ND) and declining (D) Italian oak trees sampled at the SP and OR study sites. Values are means SE. Different letters indicated significant differences (P < 0.05; MannWhitney U tests) between tree types within each site. Site .Tree type .SS (%) .Starch (%) .NSC (%) .SPND1.45 0.15a2.03 0.19a3.48 0.23aD1.41 0.15a1.73 0.16a3.14 0.15aORND2.09 0.31a2.24 0.21a4.33 0.45aD2.95 0.41a1.94 0.29a4.89 0.39aSite .Tree type .SS (%) .Starch (%) .NSC (%) .SPND1.45 0.15a2.03 0.19a3.48 0.23aD1.41 0.15a1.73 0.16a3.14 0.15aORND2.09 0.31a2.24 0.21a4.33 0.45aD2.95 0.41a1.94 0.29a4.89 0.39a Open in new tab

Concentrations of NSCs in the sapwood of non-declining (ND) and declining (D) Italian oak trees sampled at the SP and OR study sites. Values are means SE. Different letters indicated significant differences (P < 0.05; MannWhitney U tests) between tree types within each site.

Concentrations of NSCs in the sapwood of non-declining (ND) and declining (D) Italian oak trees sampled at the SP and OR study sites. Values are means SE. Different letters indicated significant differences (P < 0.05; MannWhitney U tests) between tree types within each site.

Trends in carbon (13C) and oxygen (18O) isotope ratios of tree-ring wood and WUEi of non-declining (empty symbols) and declining (filled symbols) Italian oak trees sampled in the two study sites (SP and OR). The gray filled areas indicate periods when the variables significantly (P < 0.05) differ between D and ND trees according to Wilcoxon tests. Values correspond to means SE for 3-year ring segments (n = 11 samples per tree, n = 5 trees per site).

Trends in carbon (13C) and oxygen (18O) isotope ratios of tree-ring wood and WUEi of non-declining (empty symbols) and declining (filled symbols) Italian oak trees sampled in the two study sites (SP and OR). The gray filled areas indicate periods when the variables significantly (P < 0.05) differ between D and ND trees according to Wilcoxon tests. Values correspond to means SE for 3-year ring segments (n = 11 samples per tree, n = 5 trees per site).

Scatter plots showing the associations between carbon (13C) and oxygen (18O) isotope ratios considering non-declining (empty symbols) and declining (filled symbols) oak trees at the SP and OR sites. The black arrows show the major trends of mean values of 18O and 13C comparing non-declining vs declining trees (small and big symbols show individual and mean SD values, respectively), whereas the insets indicate the inferred changes of stomatal conductance (g) vs photosynthesis rates (A). Data correspond to 3-year ring segments for the 19812013 period.

Scatter plots showing the associations between carbon (13C) and oxygen (18O) isotope ratios considering non-declining (empty symbols) and declining (filled symbols) oak trees at the SP and OR sites. The black arrows show the major trends of mean values of 18O and 13C comparing non-declining vs declining trees (small and big symbols show individual and mean SD values, respectively), whereas the insets indicate the inferred changes of stomatal conductance (g) vs photosynthesis rates (A). Data correspond to 3-year ring segments for the 19812013 period.

Diverse relationships observed between the hydraulic diameter and the WUEi for non-declining (empty symbols) and declining (filled symbols) Italian oak trees at the two study sites (SP and OR). The continuous lines highlight the significant (P < 0.05) associations for non-declining trees (correlation statistics are shown for each tree type).

Diverse relationships observed between the hydraulic diameter and the WUEi for non-declining (empty symbols) and declining (filled symbols) Italian oak trees at the two study sites (SP and OR). The continuous lines highlight the significant (P < 0.05) associations for non-declining trees (correlation statistics are shown for each tree type).

Contrasting relationships observed between the mean LW vessel diameter and the wood oxygen isotope ratio (18O) for the non-declining (empty symbols) and declining (filled symbols) Italian oak trees sampled at the SP and OR study sites. Continuous and dashed lines show these associations for non-declining and declining trees, respectively (correlation statistics are shown for each tree type). Data correspond to the 19802013 period.

Contrasting relationships observed between the mean LW vessel diameter and the wood oxygen isotope ratio (18O) for the non-declining (empty symbols) and declining (filled symbols) Italian oak trees sampled at the SP and OR study sites. Continuous and dashed lines show these associations for non-declining and declining trees, respectively (correlation statistics are shown for each tree type). Data correspond to the 19802013 period.

The very warm and dry springsummer conditions of the 1990s and 2000s triggered dieback of Italian oaks characterized by leaf shedding, shoot dieback and growth decline, which was particularly evident at the most affected site SP (Figure 1, and Figure S1available as Supplementary Data at Tree Physiology Online). These climatic conditions have been shown to cause a reduction of growth in other deciduous oak species growing at Mediterranean drought-prone sites (Tessier et al. 1994). In the present study, declining and non-declining trees did not show differences in growth rate and vessel lumen area prior to the 2000s dieback episodes (Figures 1 and 2). Only during wet decades such as the 1970s, declining trees tended to show higher BAI values, but this occurred only at the OR site, which was less affected by drought. This is in contrast to other studies where physiological differences were found between declining and non-declining trees prior to dieback in oaks (Levani et al. 2011) and conifers (Voltas et al. 2013). Therefore, our findings do not support a predisposition of the most affected trees in terms of wood anatomy, i.e., declining trees would produce wider conduits but more susceptible to drought-induced cavitation. The smaller size (d.b.h., height) of declining trees from the SP site does not indicate that the most susceptible trees to drought stress should support more leaf biomass, which could involve higher respiration costs (Levani et al. 2011). Oaks display multiple structural strategies of drought resistance including adjustments of the ratio of leaf area to sapwood area or changes in the root system and xylem anatomy (Brda et al. 2006, Klein et al. 2014). Other drought-avoidance strategies include partial or premature leaf shedding (this study and Peguero-Pina et al. 2015) or water storage in sapwood (Meinzer et al. 2003). More insight on oak dieback could be gained by quantifying biomass allocation as related to microsite conditions (slope, soil depth and texture) of declining trees, since oak mortality may be also associated with impaired root systems (Thomas and Hartmann 1996). This approach could also consider genetic information (e.g., Lloret and Garca 2016) since we still lack data to explain why in neighboring trees subjected to similar climate conditions, some are more vulnerable to drought stress than others.

As expected, the sharp drop in growth rates of declining trees started in the late 2000s (Figure 1). This recent decline in growth was accompanied by changes in some wood-anatomical traits such as the reduction of EW vessel lumen area (Figure 2). Concurrently, the decrease of 13C and 18O in declining trees also happened during that period when dieback intensified (200813). These drops in growth and alterations in wood anatomy were most patent in the more affected SP site (Figures 13). It is also possible that the growth increase observed for non-declining trees at the OR site is the consequence of reduced competition for water and nutrients with the coexisting declining and dying trees. The decrease in 13C and 18O in declining trees as compared with non-declining trees (Figure 4) would correspond to increased stomatal conductance (g) according to Scheidegger et al. (2000), and agrees with the recent reduction of WUEi in the declining trees, which was again more notable in the SP site (Figure 3). This is strictly linked with canopy dieback and with a less tight stomatal control of transpiration, perhaps as a mechanism to compensate drought stress, whereas carbon uptake through photosynthesis (A) remains stable.

Considering wood anatomy, we detected minor differences between vigor classes, suggesting a low plasticity of xylem traits as compared with leaf physiological features in response to drought. Perhaps, different functional thresholds exist regarding the detection of xylem (wood anatomy) and leaf (photosynthesis) responses to drought stress. Our findings agree with observations from a 6-year-long partial throughfall, where the evergreen diffuse-porous Quercus ilex responded to the induced drought by reducing leaf transpiring area and increasing leaf-specific conductivity, whereas vessel diameter and density remained largely unaffected albeit vessel lumen fraction (i.e., ratio between mean vessel lumen area and number per unit area) increased (Limousin et al. 2010). Contrastingly, drought reduced the lumen area of EW vessels in the ring-porous Quercus pubescens, thus lowering the hydraulic conductivity and reducing the risk of cavitation (Eilmann et al. 2009). This was consistent with our data from the declining trees at the SP site, where the production of smaller EW vessels by declining trees would explain their smaller hydraulic diameter and potential hydraulic conductivity (Figure 2). Others reported declines in vessel diameter as water availability decreases (Villar-Salvador et al. 1997) or in response to an extreme drought (Corcuera et al. 2004a), which can be consequent to reduced turgor-driven cell enlargement during xylogenesis. Overall, water shortage during spring and summer causes a severe reduction of LW production in ring-porous oaks species (Alla and Camarero 2012), which might increase their susceptibility to further water deficit if some EW vessels lose functionality due to cold- (Cavender-Bares and Holbrook 2001) and drought-induced cavitation in late winter and spring (Corcuera et al. 2004b, 2006). This could explain the trend toward forming wider LW vessels (Figure 2) and the consequent significant decrease of the vulnerability index in declining trees at the SP site (Table 2). The tendency to form LW vessels with lower minimum diameters in ND than D trees would guarantee continuous, although slow, water flow in the former under conditions promoting embolism.

We found no support to the carbon limitation hypothesis because sapwood NSC levels did not differ between declining and non-declining trees, but NSC and SS decreased in the most affected SP site (Table 3). According to Hoch (2015), hydraulic failure and growth cessation precede carbon starvation in most studies on drought-induced dieback. For instance, a seasonal study of NSC dynamics in Quercus petraea found that stem growth ceased when soil water content was depleted but the accumulation of NSCs was not affected and continued (Barbaroux and Brda 2002). Sapwood NSCs can be too coarse a variable to test carbon starvation, which may be restricted to organs such as roots located away from leaves, thus suggesting hydraulic constraints on the transport of photoassimilates (Sala et al. 2010). The finding that NSC levels decreased in the most affected SP site as compared with the less affected OR site suggests that drought (e.g., the reduced spring precipitation observed in the early 2000s with 3562% of the 19502015 mean record corresponding to 1.7 to 2.0 SPEI values) can constrain carbon storage to some level, but this threshold seems not to be as lethal as has been observed in another Mediterranean oak species (Quercus faginea) that experienced drought-induced dieback (Camarero et al. 2016). Our results do not exclude the involvement of carbon starvation in the studied drought-induced dieback episode since trees were exposed to prolonged water shortage, which could constrain carbon uptake due to stomatal limitation (Ogasa et al. 2013). However, hydraulic failure seems to be the most plausible cause of dieback as it has been observed under experimental conditions, where trees experienced extensive cavitation during advanced drought (Barigah et al. 2013, Hartmann et al. 2013).

Regarding isotope data, our results partially agree with those presented by Hentschel et al. (2014), who reported lower 13C values in declining Picea abies trees and interpreted them as an increase of stomatal conductance rates, while non-declining trees had low stomatal conductance, allowing them to prevent water loss during dry periods. The higher 13C of non-declining trees from the SP site (Figure 3) indicates an overall more conservative water use (Scheidegger et al. 2000) in comparison with declining individuals, suggesting the capacity of those individuals to effectively close stomata to prevent excessive water loss. In contrast, the dual-isotope approach allowed inferring a sustained increase of stomatal conductance in declining trees, probably leading to a positive feedback characterized by a drought- and heat-enhanced water loss through transpiration that is not compensated by reduced growth rates, smaller EW lumen areas and steady WUEi (Ponton et al. 2001). This cascade of altered functioning would explain the uncoupling between growth and WUEi in declining trees (Figure S7 available as Supplementary Data at Tree Physiology Online) and the lack of association observed at the most affected SP site between LW vessel lumen diameter, a proxy for summer hydraulic conductivity, and 18O, a proxy for evaporative demand (Figure 6). It may be also the case that the drought-induced sharp reduction of growth uncouples the tree-ring and leaf isotope signatures (Pflug et al. 2015), thus making the dual-isotope approach unreliable (Roden and Siegwolf 2012).

Finally, the study forests have not been recently managed, but their previous management during early stand development may have shaped their current stand structure (e.g., stand density, hydraulic architecture and height of trees), which could influence the recent tree responses to droughts. Stand density reductions could improve the water balance of these forests (McDowell et al. 2006), but currently they are not dense forests. More importantly, the fact that anysohydric species with ring-porous wood such as the Italian oak succumb to warm and severe dry conditions questions the validity of thinning once VPD has reached such minimum values that stomatal conductance reaches very low values and xylem cavitation becomes widespread.

Although a further advance in the knowledge has been achieved from the plethora of recent studies on drought-induced dieback, this process remains a poorly understood phenomenon. The main reason is that the comprehension of causes and mechanisms involved necessarily requires a multi-proxy approach, which is often lacking from studies. Our approach, by investigating on either physiological or structural aspects, partially supports hydraulic failure as the potential cause of oak dieback. We observed that declining tress presented a reduced WUEi after the onset of dieback, especially in the most drought-affected site, which is coherent with an enhanced water loss. These joint evidences, with the parallel absence of any significant differences in sapwood NSC concentrations, partially support the hydraulic failure mechanism of dieback. In future scenarios of increased severity and duration of droughts, as foreseen under warmer conditions, dieback phenomena may affect drought-tolerant anysohydric species, as they keep a narrow margin of hydraulic safety.

This research was financially supported by the project Alarm of forest mortality in Southern Italy (Gorgoglione Administration, Basilicata Region, Italy) and by the CGL201569186-C2-1-R project (Spanish Ministry of Economy). M.C. was supported by the PhD program from the course of Agricultural, Forest and Food Science at the University of Basilicata (Italy).

Alla AQ, Camarero JJ (2012) Contrasting responses of radial growth and wood anatomy to climate in a Mediterranean ring-porous oak: implications for its future persistence or why the variance matters more than the mean. Eur J For Res 131:15371550.

Barbaroux C, Brda N (2002) Contrasting distribution and seasonal dynamics of carbohydrate reserves in stem wood of adult ring-porous sessile oak and diffuse-porous beech trees. Tree Physiol 22:12011210.

Barigah TS, Charrier O, Douris M, Bonhomme M, Herbette S, Amglio T, Fichot R, Brignolas F, Cochard H (2013) Water stress-induced xylem hydraulic failure is a causal factor of tree mortality in beech and poplar. Ann Bot 112:14311437.

Battipaglia G, Jggi M, Saurer M, Siegwolf RTW, Cotrufo MF (2008) Climatic sensitivity of 18O in the wood and cellulose of tree rings: results from a mixed stand of Acer pseudoplatanus L. and Fagus sylvatica. Palaeogeogr Palaeoclimatol Palaeoecol 261:193202.

Battipaglia G, De Micco V, Brand WA, Linke P, Aronne G, Saurer M, Cherubini P (2010) Variations of vessel diameter and 13C in false rings of Arbutus unedo L. reflect different environmental conditions. New Phytol 188:10991112.

Billings SA, Boone AS, Stephen FM (2016) Tree-ring 13C and 18O, leaf 13C and wood and leaf N status demonstrate tree growth strategies and predict susceptibility to disturbance. Tree Physiol 36:576588.

Boettger T, Haupt M, Knller K et al.. (2007) Wood cellulose preparation methods and mass spectrometric analyses of 13C, 18O, and nonexchangeable 2H values in cellulose, sugar, and starch: an interlaboratory comparison. Anal Chem 79:46034612.

Brda N, Huc R, Granier A, Dreyer E (2006) Temperate forest trees and stands under severe drought: a review of ecophysiological responses, adaptation processes and long-term consequences. Ann For Sci 63:625644.

Chatziphilippidis G, Spyroglou G (2004) Sustainable management of coppice forests in Greece. Towards the sustainable use of Europe's forestsForest Ecosystem and Landscape Research: Scientific Challenges and Opportunities. In: Andersson F, Birot Y, Paivinen R (eds) EFI Proceedings 49, European Forest Institute, Joensuu, Finland, pp 5160.

Cook ER, Krusic PJ (2005) Program ARSTAN: A tree-ring standardization program based on detrending and autoregressive time series modeling, with interactive graphics. Columbia University, Palisades, New York, NY.

Corcuera L, Camarero JJ, Sis S, Gil-Pelegrn E (2006) Radial-growth and wood-anatomical changes in overaged Quercus pyrenaica coppice stands: functional responses in a new Mediterranean landscape. Trees 20:9198.

Ferrio JP, Araus JL, Bux R, Voltas J, Bort J (2005) Water management practices and climate in ancient agriculture: inferences from the stable isotope composition of archaeobotanical remains. Veg Hist Archaeobot 14:510517.

Gea-Izquierdo G, Viguera B, Cabrera M, Caellas I (2014) Drought induced decline could portend widespread pine mortality at the xeric ecotone in managed Mediterranean pine-oak woodlands. For Ecol Manage 320:7082.

Hentschel R, Rosner S, Kayler ZE, Andreassen K, Brja I, Solverg S, Tveito OE, Priescak E, Gessler A (2014) Norway spruce physiological and anatomical predisposition to dieback. For Ecol Manage 322:2736.

Ogasa M, Miki NH, Murakami Y, Yoshikawa K (2013) Recovery performance in xylem hydraulic conductivity is correlated with cavitation resistance for temperate deciduous tree species. Tree Physiol 33:335344.

Peguero-Pina JJ, Sancho-Knapik D, Martn P, Saz MA, Gea-Izquierdo G, Caellas I, Gil-Pelegrn E (2015) Evidence of vulnerability segmentation in a deciduous Mediterranean oak (Quercus subpyrenaica E. H. del Villar). Trees 29:19171927.

Pellizzari E, Camarero JJ, Gazol A, Sangesa-Barreda G, Carrer M (2016) Wood anatomy and carbon-isotope discrimination support long-term hydraulic deterioration as a major cause of drought-induced dieback. Glob Chang Biol 22:21252137.

Pflug EE, Siegwolf R, Buchmann N, Dobbertin M, Kuster TM, Gnthardt-Goerg MS, Arend M (2015) Growth cessation uncouples isotopic signals in leaves and tree rings of drought-exposed oak trees. Tree Physiol 35:10951105.

Ponton S, Dupouey JL, Brda N, Feuillat F, Bodns C, Dreyer E (2001) Carbon isotope discrimination and wood anatomy variations in mixed stands of Quercus robur and Quercus petraea. Plant Cell Environ 24:861868.

Salvatore M, Alessandro R, Irene D (2002) Current situation of oak decline in Italy and in other European countries. In Villemant C, Sousa E (eds) IOBC/WPRS Bulletin vol. 25. Working Group Integrated Protection in Oak Forests. Lisbon, Portugal.

Tognetti R, Longobucco A, Raschi A (1998) Vulnerability of xylem to embolism in relation to plant hydraulic resistance in Quercus pubescens and Quercus ilex co-occurring in a Mediterranean coppice in central Italy. New Phytol 139:437447.

Voltas J, Camarero JJ, Carulla D, Aguilera M, Oriz A, Ferrio JP (2013) A retrospective, dual-isotope approach reveals individual predispositions to winter-drought induced tree dieback in the southernmost distribution limit of Scots pine. Plant Cell Environ 36:14351448.

Get in Touch with Mechanic
Related Products
Recent Posts
  1. valve repair wet ball mill

  2. ball mill 2100 * 7000

  3. feed rods in to mill

  4. rod mill 1 at open

  5. belgium efficient stone milling production line price

  6. ball mill catalytic converter

  7. feldspar lumps bauxite ball mill

  8. ball mills south africa

  9. quartz mining wet ball mill quartz mining growing

  10. sops for ball mill equepments

  11. harper efficient iron ore toothed roll crusher price

  12. oran portable brick and tile industrial dryer manufacturer

  13. gold crushing equipment for sale

  14. ball mill vs roller feeder

  15. mobile iron ore beneficiation machine

  16. hammer crusher vs roll crusher

  17. sousse economic calcium carbonate crushing production line

  18. high efficiency jaw crusher rock breaker

  19. price of granite crushing plant 250 tonnes

  20. tajikistan portable bentonite spiral chute separator