Phylogenetic variation in the shoot mineral concentration of angiosperms

The calcium (Ca) concentration of plant shoot tissues varies systematically between angiosperm orders. The phylogenetic variation in the shoot concentration of other mineral nutrients has not yet been described at an ordinal level. The aims of this study were (1) to quantify the shoot mineral concentration of different angiosperm orders, (2) to partition the phylogenetic variation in shoot mineral concentration between and within orders, (3) to determine if the shoot concentration of different minerals are correlated across angiosperm species, and (4) to compare experimental data with published ecological survey data on 81 species sampled from their natural habitats. Species, selected pro rata from different angiosperm orders, were grown in a hydroponic system under a constant external nutrient regime. Shoots of 117 species were sampled during vegetative growth. Significant variation in shoot carbon (C), calcium (Ca), potassium (K), and magnesium (Mg) concentration occurred between angiosperm orders. There was no evidence for systematic differences in shoot phosphorus (P) or organic-nitrogen (N) concentration between orders. At a species level, there were strong positive correlations between shoot Ca and Mg concentration, between shoot P and organic-N concentration, and between shoot K concentration and shoot fresh weight:dry weight ratio. Shoot C and cation concentration correlated negatively at a species level. Species within the Poales and the Caryophyllales had distinct shoot mineralogies in both the designed experiment and in the ecological survey.     

Applications of mineral nutrients to heavily N-fertilized Scots pine trees: Effects on arginine and mineral nutrient concentrations

The aim of the investigation was to study if improved nutrient status in Scots pine (Pinus sylvestris L) trees would be reflected in decreased concentrations of arginine in the needles. The studies trees had imbalanced mineral nutrient composition and elevated needle arginine concentrations caused by long-term fertilization with N. Concentrations of arginine and mineral nutrients in needles were followed over three consecutive years of additional fertilization with N alone or with P, K, Mg and micronutrients in combination with and without N.Analysis of needle mineral concentrations suggested that there were deficiencies only in K and Mg. The N concentration increased both in trees fertilized with N alone and in trees fertilized with N in combination with mineral nutrients. In the control treatment and in trees fertilized with mineral nutrients other than N the N concentration remained fairly constant. The highest Ca/N, K/N and P/N ratios were found in trees fertilized with mineral nutrients other than N while the lowest ratios were found in trees fertilized with N alone. Arginine concentrations in needles from trees fertilized with N alone remained at a high level throughout the experiment while arginine concentrations in trees given the other treatments decreased.The results show that the mineral nutrient balance can be improved with appropriate fertilization and that this improvement is reflected in decreasing arginine levels. Furthermore the study demonstrates that when N supply is reduced the arginine concentration also decreases also as an effect of reduced N supply per se. The study also indicates that arginine may be a better measure of the N status in pine trees than total N.     

Litter fall and nutrient dynamics in a Nigerian rain forest seven years after a ground fire

Abstract. Seasonal litter fall and mineral element content (N, P, Ca, Mg, K) of regrowth forest communities at the base and on the slope of an inselberg in Ile-Ife, Nigeria, were studied 7 yr after a ground fire ravaged the forest. Litter fall (tha^?1 yr^?1) was 4.6 (total), 4.2 (leaf), 0.3 (small wood < 2.5 cm diameter) and 0.1 (reproductive parts: fruits and flowers) in the base community and 6.4 (total), 5.4 (leaf), 0.9 (small wood) and 0.1 (reproductive parts) in the slope community. There was significant monthly variation in litter fall in the two communities with lowest amount of litter recorded during the wettest months of the year (May - August) and the highest amount during the dry season. Significant monthly variation (P<0.05) in Ca, Mg and K concentration in leaf litter and for Mg (P < 0.01) in fruit litter occurred, with the lowest concentration recorded during the wettest months (May-August). In leaf and wood litter the order of mineral element concentration was Ca>N>K> Mg > P while in fruit litter it was N > K > Ca > Mg > P. Quantities of mineral element (kg ha^-1 yr¹) returned to the soil via litterfall were N: 66; P: 4; Ca: 97; Mg: 15; K: 45 in base forest, and N: 112; P: 5; Ca: 142; Mg: 20; K: 66 in slope forest. Through leaf litter >88.5% of these elements was returned into the two communities, through wood > 4.0% and through reproductive organs > 0.3%. The order of quantities of these elements returned in leaf and wood litter was Ca > N > K > Mg > P, in fruit litter N ～ K > Ca > Mg > P. Significant monthly variation in the amounts of the various elements returned were recorded in leaf litter, but not in wood and fruit litter. The lowest amount of various elements was returned during the wettest months (May-August) which coincided with the period of the lowest element concentration and litter fall.     

Variations in the foliar nutrient content of mire plants: effects of growth-form based grouping and habitat

We determined concentrations of major nutrients in the vegetation of six habitat types (hummock, scrub, lawn, fen meadow, hollow and marginal stream), spanning a broad range of environmental conditions as regards water-table depth and water chemistry, in five mires on the southern Alps of Italy. Our study was based on chemical analyses of living tissues of plant species, grouped into growth-form based plant functional types (PFTs). We aimed at assessing to what extent the observed differences in tissue nutrient content were accounted for by community composition (both in terms of species and PFTs) and by habitat. Nutrient concentrations were overall lowest in Sphagnum mosses and highest in forbs, although the latter showed large variations presumably due to heterogeneity in mechanisms and adaptations for acquiring nutrients among species within this PFT. Nutrient content patterns in the other three PFTs varied greatly in relation to individual nutrients, with evergreen shrubs showing low nitrogen (N) concentrations, graminoids showing high N concentrations but low potassium (K) and magnesium (Mg) concentrations and deciduous shrubs showing rather high phosphorus (P) concentrations. Habitat accounted for a modest fraction of variation in tissue concentration of all nutrients except P. We concluded that the nutrient status of mire vegetation is primarily controlled by community composition and structure although habitat does exert a direct control on P concentration in the vegetation, presumably through P availability for plant uptake.     

福建省马尾松和杉木针叶中7种营养元素含量特征

为了解福建省马尾松和杉木人工林养分需求特征,选取25个马尾松(Pinus massoniana)林和31个杉木(Cunninghamia lanceolata)林,测定了针叶中7种主要营养元素(C、N、P、S、K、Ca、Mg)含量。结果表明,马尾松针叶K、Ca、Mg含量显著低于杉木,分别是杉木的68%、14%、50%,表明杉木对矿质养分尤其Ca的需求远高于马尾松,提示马尾松比杉木更能适应矿质养分尤其Ca较贫瘠的立地。两树种针叶C、N、P、S含量及其化学计量比均表现出相似的内稳态特征,而Ca、Mg、K含量及其化学计量比的变异大,表明这3种矿质元素受立地条件的影响较大。两树种对N、P、K需求表现为协同正相关关系,N与Ca含量则为负相关关系,暗示持续N沉降或营林实践中长期施N肥可能抑制Ca的吸收。     

Significance of sequential leaf development for nutrient balance of the cotton sedge,Eriophorum vaginatum L.

Summary The sedgeEriophorum vaginatum in an interior Alaskan muskeg site produced leaves sequentially at about 1.5-month intervals. Each leaf remained active for two growing seasons. Young leaves (even those initiated late in the season) always had high concentrations of N, P, K and Mg and were low in Ca. Stems had high concentrations of nutrients, sugar, amino acid N and soluble organic P in autumn and spring but low concentrations in summer. Growth of leaves in spring was strongly supported by translocation from storage. Leaves approached their maximum nutrient pool before nutrient uptake began in late spring, one month before maximum biomass. Retranslocation of nutrients from aging leaves could support nutrient input into new, actively growing leaves as a consequence of the sequential leaf development. For instance retranslocation from aging leaves accounted for more than 90 and 85% of P and N input to new leaves appearing in early summer and 100% to leaves that appeared later. Leaching losses were negligible. Half time for decay of standing dead litter was 10 years. We suggest that sequential leaf development paired with highly efficient remobilization of nutrients from senescing leaves enables plants to recycle nutrients within the shoot and minimize dependence upon soil nutrients. This may be an important mechanism enablingEriophorum vaginatum to dominate nutrient-poor sites. This may also explain why graminoids with sequential leaf production cooccur with evergreen shrubs and dominate over forbs and deciduous shrubs in nutrient-poor sites in the boreal forest (e.g., in bogs) and at the northern limit of the tundra zone.     

兰竹荔枝叶片营养元素适宜含量的研究

本研究旨在探讨福建荔枝最重要品种兰竹的叶片营养元素适宜含量。统计分析表明,不同地点、年份对同一品种叶片元素含量存在明显的差异;荔枝叶片含氮量变异系数最小,并按磷、镁、钙、钾依次增大。本研究初步提出的丰产兰竹荔枝秋梢叶片营养元素适宜含量为:氮1.5—2.2％,磷0.12—0.18％,钾0.7—1.4％,钙0.3—0.8％,及镁0.18一0.88％;其叶片氮、磷、钾、钙、镁的适宜比例是1:0.08:0.57:0.30:0.12。上列指标可供荔枝营养诊断指导施肥之参考。     

元谋干热河谷燥红土和变性土上植物叶片的元素含量及其重吸收效率

采用盆栽试验,研究元谋干热河谷燥红土和变性土上生长的植物叶片以及凋落叶营养元素含量,并分析养分重吸收效率对土壤类型与物种互作的响应.结果表明: 土壤类型对叶片N、P、Ca、Mg、Cu、Zn、Fe、N∶P以及凋落叶N、P、Mn、N∶P均有显著影响;燥红土植物叶片与凋落叶N、Mn含量和N∶P显著高于变性土,而燥红土植物叶片P、Ca、Mg、Fe、Cu、Zn和凋落叶P含量显著低于变性土.燥红土植物叶片N含量较变性土高34.8%,而P含量低40.0%;在叶片凋落时,N、P、K表现为重吸收,而其他元素呈富集状态.燥红土凋落叶Ca、Mg、Mn富集系数显著高于变性土.物种仅对叶片N含量有显著影响,物种与土壤交互作用对植物叶片和凋落叶元素含量影响不显著,表明各土壤类型对不同物种元素含量的影响方式较为一致.土壤类型对植物元素含量的影响可进一步作用于干热河谷植物凋落物分解、植物-土壤的养分反馈以及生物地球化学循环.     

An analysis of the growth response of carrot seedlings to deficiency in some mineral nutrients

Critical plant concentrations for a reduction in relative growth rate to 90% of that of fully nourished plants were estimated by a novel method for several mineral nutrients. Carrot plants were grown from seed for 28 days in a range of nutrient solutions omitting N, P, K, Ca, S, Mg, Fe, B, Mn, Zn, Cu and Mo as separate treatments. All treatments except -Mn, -Zn, -Cu and -Mo resulted in effects on plant growth and the development of deficiency symptoms. Estimates of critical concentrations were based on a simple simulation model incorporating the principle of nutrient dilution with increasing plant weight and on mineral analysis of the plants. Parameters governing the shape of the relationship between fractional relative growth rate and plant nutrient concentration were altered until the model predicted the observed final mean dry weight of deficient plants and time of divergence of this growth curve from that of fully nourished plants. Critical concentrations so obtained were higher than those previously reported for Ca, Fe, N and P in carrots and lower for K, Mg and S.     

The diagnosis and recommendation integrated system (DRIS) for diagnosing the nutrient status of grassland swards: I. Model establishment

Herbage analysis offers a definitive means of determining the N, P, K and S status of perennial ryegrass swards. Unfortunately, the results of such analyses can be difficult to interpret, simply because the minimum or 'critical' concentration of a nutrient in plant tissue for optimum growth, varies both with crop age and with changes in the concentrations of other nutrients. The Diagnosis and Recommendation Integrated System (DRIS) could help to improve the reliability of such interpretations. Diagnoses made using DRIS are based on relative rather than on absolute concentrations of nutrients in plant tissue, and as such should be comparatively independent of crop age.The aim of this study was to establish and test DRIS methodology for high-yielding perennial ryegrass swards. Because of prohibitive costs, setting up a whole new series of field experiments to evaluate DRIS model parameters for perennial ryegrass was out of the question. Instead, the diagnostic norms and associated coefficients of variation for the model were evaluated using data from a single (large) multi-factorial glasshouse experiment.Of the nutrient ratios selected to form the diagnostic norms, K/N and S/N had the clearest physiological rationale, whereas those involving Ca and Mg in combination with N, P, K and S appeared to have little physiological basis. It was reasoned, though, that because Ca and Mg uptake by plants are largely passive processes (ultimately governed by plant growth), the DRIS indices for these nutrients, together reflected the degree to which growth may be limited by non-nutritional (environmental) factors relative to nutritional ones. Both indices were combined to form a single reference (Rⁱ) index. Without such an internal reference, plant growth could be limited by multiple nutrient deficiencies, and yet N, P, K and S indices might all be close to, or equal to zero (i.e. the optimum), simply because the absolute concentrations of each nutrient (while low) had been in the correct state of balance. Moreover, by effectively using Ca and Mg as internal reference parameters in DRIS, 'nutrient concentrations' which previously formed the basis of the critical value approach, were essentially incorporated into the DRIS model, thus combining the strengths of the two diagnostic approaches; the only difference being that Ca and Mg, and not dry matter, were the internal references against which the levels of the major nutrients were compared.     

Nutrient cycling and foliar status in an urban pine forest in Athens, Greece

The nutrient cycling and foliar status for the elements Ca, Mg, K, N, P, S, Fe, Mn, Zn and Cu were investigated in an urban forest of Aleppo pine (Pinus halepensis) in 2004 in Athens, Greece in order to draw conclusions on the productivity status and health of the ecosystem. The fluxes of bulk and throughfall deposition were characterized by the high amounts of Ca, organic N and sulfate S. The magnitude of the sulfate S fluxes indicated a polluted atmosphere. The nutrient enrichment in throughfall was appreciable for ammonium N, P and Mn. The mineral soil formed the largest pool for all the elements followed by the forest floor, trunk wood and trunk bark. The understory vegetation consisting of annual plants proved important for storing N, P and K. Compared to current year needles of Aleppo pine in remote forests of Spain, the needles of the Aleppo pine trees in Athens had significantly higher concentrations of Ca, N, P and Cu and significantly lower concentrations of Mg and Zn. The soil had a high concentration of calcium carbonate and accordingly high pH values. When all inputs to the forest floor were taken into account, the mean residence time of nutrients in the forest floor followed the order Fe > Mn > Cu > Ca > Mg > P > Zn > N > K > S.     

Foliar macronutrients (N,P, K,Ca and Mg) in lime (Tilia spp.)

Summary The effect of sampling technique upon the measurement of variation in nutrient (N, P, K, Ca and Mg) concentration was examined. Sampling and analytical precision over two years were indicated by coefficients of variation which ranged from 5.7 to 14.0 per cent for sampling error and 4.3 to 16.1 per cent for analytical error. Concentration of N, P and K in the leaves was found to be independent of sampling height. Ca and Mg concentrations, however, decreased with increased height on the crown. The differences in concentration between the bottom and top of the crown were 15 and 19 per cent for Ca and Mg respectively. The nutrient concentrations of samples taken from basal coppice shoots were significantly different from those in the true crown. Nutrient concentration was independent of aspect. Leaf area varied throghout the season, but the largest leaves were always at the base of the crown and the smallest at the top.     

What controls plant nutrient use in high elevation ecosystems?

The importance of rock-derived mineral nutrients (P, K, Mn, Mg, and Ca) in plant physiological function is well established. However, one important and relatively unexplored question is whether or not the same rules of plant nutrient use efficiency apply to these essential elements even if they are not limiting to primary production. We examined conifer growth and nutrient use dynamics across sites with contrasting geologies (sedimentary and volcanic) that vary in both rock-derived mineral nutrient and N availability. Differences in bedrock geochemistry generally corresponded to differences in available soil nutrients, such that the volcanic site tended to have greater available nutrients. Foliar nutrient concentrations reflected both differences in bedrock chemistry and indices of available soil nutrients for P, K, and Mn. Aboveground biomass production did not follow expected patterns and was greater for trees growing on low nutrient sites, but only with respect to the annual woody increment. Fine litter production did not differ between sites. Finally, we found evidence for trade-offs between two commonly examined components of nutrient use efficiency (NUE): nutrient productivity (A _n) and mean residence time of nutrients. However, we did not find evidence for higher plant NUE in soils with lower nutrient availability for N or rock-derived nutrients.     

Relationships of Starch Concentration with Specific Leaf Weight and Mineral Concentrations in Potato Leaves Under Varied CO2 and Temperature

Foliar concentrations of starch and major nutrients N, P, K, Ca, and Mg along with specific leaf weight (SLW) were determined in the potato (Solaruan tuberosum L. ) cvs "Denali", "Norland” and "Russet Burbank" grown for 35 days under the CO2 concentrations of 500, 1 000, 1 500 and 2 000 μmol' mol-l at both 16 and 20 ℃ air temperature. The starch concentration, pooled from the three cuhivars, increased with increasing CO2 concentration at both 16 and 20 ℃, and was consistently higher at 16 ℃ than at 20 ℃. The SLW (g·m-2) was positively related to the foliar starch concentration on the basis of leaf area or dry weight. The concentrations of N, P, Ca, and Mg in leaves were negatively related to starch concentration under 14% starch on a dry weight basis. Above 14%, there was no significant relationship between nutrient and starch concentrations. The similar patterns were seen when the SLW and nutrient concentrations were expressed on a starch-free basis. In contrast, the leaf concentration of K was not closely related to the starch concentration. The results indicated that the changes in SLW and concentrations of N, P, Ca, and Mg in potato leaves only partially resulted from the changed starch concentration.     

东灵山地区辽东栎叶养分的季节动态与回收效率

为探讨辽东栎叶的养分回收对种群的适应能力和生态系统的养分循环的意义,在东灵山地区分析了辽东栎叶中Ｎ、Ｐ、Ｋ、Ｃa、Ｍg、Fe等６种元素的浓度和单位叶面积含量,确定了了辽东栎叶养分的季节动态与回收效率。结果发现：１）６种元素的浓度和单位叶面积含量在叶的生活史中都发生了明显的季节变化,其中Ｎ、Ｐ、Ｋ等３种元素的浓度在展叶初期下降很快,阴后逐渐下降的,但单位叶面积含量是在７月下旬达到最大值;２）Ｃa浓度和单位叶面积含量整个生长季中一直保持上升趋势,Ｍg的季节变化无一定规律;３）Ｎ、Ｐ、Ｋ、Ｍg都有一定程度的回收,其中Ｎ、Ｐ的回收率分别为６７％和８７％,Ｃa、Ｆe在叶有所积累。为探讨叶养分回收效率在常绿和落叶植物中的差异,对栎属植物养分动态进行了比较分析。结果表明：１）落叶植物成熟叶中的Ｎ浓度稍高于常绿植物,但不显著,Ｎ的回收率与成熟叶中的浓度成显著正相关,;２）栎属植物的成熟叶和落叶中的Ｎ、Ｐ浓度间呈显著正相关。     

Carbon,Nitrogen, Phosphorus,and Potassium Stoichiometry in an Ombrotrophic Peatland Reflects Plant Functional Type

Ombrotrophic bog peatlands are nutrient-deficient systems and important carbon (C) sinks yet the stoichiometry of nitrogen (N), phosphorus (P) and potassium (K), essential for plant growth and decomposition, has rarely been studied. We investigated the seasonal variation in C, N, P, and K concentrations and their stoichiometric ratios in photosynthetically active tissues of 14 species belonging to five plant functional types (PFTs) (mosses, deciduous trees/shrubs, evergreen shrubs, graminoids, and forb) at Mer Bleue bog, an ombrotrophic peatland in eastern Ontario, Canada. Although we observed variations in stoichiometry among PFTs at peak growing season, there was convergence of C:N:P:K to an average mass ratio of 445:14:1:9, indicating N and P co-limitation. Nitrogen, P, and K concentrations and stoichiometric ratios showed little seasonal variation in mosses, evergreens, and graminoids, but in forb and deciduous species were the largest in spring and decreased throughout the growing season. Variations in nutrient concentrations and stoichiometric ratios among PFTs were greater than seasonal variation within PFTs. Plants exhibit N and P co-limitation and adapt to extremely low nutrient availability by maintaining small nutrient concentrations in photosynthetically active tissues, especially for evergreen shrubs and Sphagnum mosses. Despite strong seasonal variations in nutrient availabilities, few species show strong seasonal variation in nutrient concentrations, suggesting a strong stoichiometric homeostasis at Mer Bleue bog.     

鼎湖山马尾松林降水再分配对养分动态影响

在鼎湖山马尾松林中 ,降水到达森林重新分配后 ,pH值的大小变化顺序为 :茎流 >径流 >穿透雨 >降水 ,且它们均呈酸性。与 pH值变化基本上相类似 ,养分元素浓度在各过程中均以茎流最高 ,降水最低。不同元素比较 ,其大小次序为 :Ca>N >Mg>P。Ca、P和Mg三种元素浓度在各过程中的月变化较明显 ,但pH值和N元素浓度的月际变化不强。该生态系统养分元素呈积累状态 ,大气降水的养分输入为 (kg·hm 2 ·a 1) :2 0 .62 9(N)、1 .0 1 5 (P)、46.3 3 2 (Ca)、3 .5 0 8(Mg) ;输出为 (kg·hm 2 ·a 1) :7.1 2 0 (N)、0 .63 2 (P)、2 3 .640 (Ca)、2 .1 3 3 (Mg)。此外 ,还对该生态系统生物地球化学循环特征进行了讨论。     

Controls of Bedrock Geochemistry on Soil and Plant Nutrients in Southeastern Utah

The cold deserts of the Colorado Plateau contain numerous geologically and geochemically distinct sedimentary bedrock types. In the area near Canyonlands National Park in Southeastern Utah, geochemical variation in geologic substrates is related to the depositional environment with higher concentrations of Fe, Al, P, K, and Mg in sediments deposited in alluvial or marine environments and lower concentrations in bedrock derived from eolian sand dunes. Availability of soil nutrients to vegetation is also controlled by the formation of secondary minerals, particularly for P and Ca availability, which, in some geologic settings, appears closely related to variation of CaCO₃ and Ca-phosphates in soils. However, the results of this study also indicate that P content is related to bedrock and soil Fe and Al content suggesting that the deposition history of the bedrock and the presence of P-bearing Fe and Al minerals, is important to contemporary P cycling in this region. The relation between bedrock type and exchangeable Mg and K is less clear-cut, despite large variation in bedrock concentrations of these elements. We examined soil nutrient concentrations and foliar nutrient concentration of grasses, shrubs, conifers, and forbs in four geochemically distinct field sites. All four of the functional plant groups had similar proportional responses to variation in soil nutrient availability despite large absolute differences in foliar nutrient concentrations and stoichiometry across species. Foliar P concentration (normalized to N) in particular showed relatively small variation across different geochemical settings despite large variation in soil P availability in these study sites. The limited foliar variation in bedrock-derived nutrients suggests that the dominant plant species in this dryland setting have a remarkably strong capacity to maintain foliar chemistry ratios despite large underlying differences in soil nutrient availability.     

Nutrient retention by Pseudoscleropodium purum and its relation to growth

Abstract

A complete mineral nutrient fertilizer solution was added to carpets of Pseudoseleropodium purum growing in Windsor Forest, Berkshire, to study the efficiency of nutrient capture from dilute solutions and its effect on growth. In the experiment different frequencies of application were balanced by employing different concentrations of fertilizer so that all treatment plots received the same quantities of nutrients over a six-month period. At the end of the experiment the concentrations of macronutrients (N, P, K, Mg, Ca) within the moss tissues were determined and shoot growth was assessed by the tagged shoot method.

Significant luxury uptake occurred only in the case of P where shoot concentrations in fertilized plots were 50% above the control. Smaller net increases were noted for other elements, particularly Mg. Shoot N concentration was hardly affected by fertilizer application possibly due to efficient internal recycling of the element in P. purum. Levels of exchange ably held cations were not markedly increased by fertilizer addition in the field experiment but when shoots were incubated with fertilizers in the laboratory for 30 min the concentrations of exchangeable Ca and Mg rose appreciably and exchangeable K fell. These changes appear to be quickly reversed under field conditions without transfer of cations to the protoplasts. The importance of elevated cation exchange capacity in the initial sequestration of nutrient cations is, therefore, questioned. Growth stimulation due to fertilizer treatment was negligible. Maximum net uptake of P, Mg and Ca occurred when plots were watered at weekly intervals with a dilute fertilizer and least occurred where a concentrated solution was applied at less frequent intervals implying that contact time between P. purum and its potential nutrients is an important factor. The period elapsing between fertilizer additions may also have significance, however, due to leaching away of nutrients and the re-establishment of a natural ionic balance at the cell wall exchange sites. Levels of nutrient absorption and growth were greatest at a site where the moss remained moist longest. This result indicates that the ability of P. purum to absorb and utilize additional nutrients is governed by general metabolic performance which is itself restricted by water supply in Windsor Forest.