CaptureReducing greenhouse gas emissions from agriculture and other sectors of California’s economy has become one of the most important environmental concerns of state and federal regulatory organizations. This is the result of the June 2006 passage of the California Global Warming Solutions Act, Assembly Bill 32, which calls for reducing the state’s greenhouse gas (GHG) production to 1993 levels by 2020; and the U.S. Environmental Protection Agency’s recent endangerment finding for the primary greenhouse gases (GHGs) produced by agriculture, namely carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4).

The finding subjects these GHGs to regulation under the California Environmental Quality Act (CEQA). For these reasons more attention is now focused on carbon footprints, here defined as the net amount of GHGs emitted and consumed by a vineyard in production up to the vineyard gate.

Our ultimate goal is to lessen (mitigate) emissions of GHGs like N2O, enhance both CH4 oxidation and photosynthetic CO2 assimilation by California vineyards and foster carbon sequestration into soils, all of which can lower a system’s carbon footprint, even to the point where it can be negative.

__________________________________________________________________________________

Capture

This article as originally published in 2013 in Practical Winery and Vineyard Journal and is available HERE. Used with permission.

__________________________________________________________________________________

There is limited knowledge available about how well current “sustainable” farming practices mitigate GHG emissions in vineyards (reference 6). We know of no complete vineyard carbon balances or carbon footprints.

Our long-term research in this area will soon complete its 10th year. We assembled the first working budget of the three major GHGs in a single vineyard in Napa Valley. We have compiled extensive information about carbon cycling and soil carbon sequestration under minimum-tillage and conventional-tillage conditions.

We are working closely with the California Sustainable Winegrowing Alliance to help fill data gaps identified by E. Carlisle with the ultimate goal of perfecting a carbon footprint decision-support system (reference 6). The project also coordinates with efforts by Applied Geosolutions (cooperators Dr. William Salas, Dr. Changsheng Li) to calibrate the DeNitrification DeComposition model for vineyards. This model will be embedded in the decision-support system for use by grapegrowers and other practitioners to facilitate carbon sequestration and assess carbon footprints for a variety of management practice options, soils and climates.

The data also are being used by cooperators Dr. Alissa Kendall, Elias Marvinney and Sonja Brodt (UC Davis) to assemble life-cycle analyses for carbon footprints of vineyards and orchards.

Experiment

A winter barley cover crop was directly seeded to build up soil organic carbon.

A winter barley cover crop was directly
seeded to build up soil organic carbon.

The trial is being carried out in a 2.1 acre Cabernet Sauvignon vineyard at the University of California, Davis, Department of Viticulture & Enology’s Oakville Research Station in Napa Valley. The trellis is a six-wire VSP system with bilateral cordons. Yield and biomass production have been recorded for the rootstock 101-14 Mgt, which has a relatively shallow root system with moderate vigor and was the most widely planted in the area at the onset of the experiment.

The vineyard was conventionally tilled for 12 years (1991 – 2002). In 2003, we initiated an experiment utilizing three tillage management practices: 1) “minimum tillage” with a dwarf barley cover crop (shallow-disked to about 1-inch depth with one pass every second year to facilitate cover crop seedling establishment),
2) “tillage with dwarf barley cover crop” (disked 8 to 12 inches deep twice per year), and 3) “conventional tillage” with resident vegetation (disked 8 to 12 inches deep once per year).

As opposed to a no-tillage treatment, minimum tillage conformed to the needs of the vineyard manager for water management, while at the same time requiring fewer cultivation passes through the tractor rows.

Tillage breaks up soil aggregates and accelerates oxidation of soil carbon by microbial organisms, which leads to its loss. About 19% of anthropogenic CO2 in the atmosphere is a consequence of soil disturbance by cultivation and deforestation (reference 3).

Exploring ways to foster increases in the carbon content of these soils (soil carbon sequestration) can improve vineyard carbon footprints. We have been assessing soil carbon sequestration through periodic analyses of carbon in the plow horizon (“Ap horizon”). The soil carbon content is expected to increase to some degree when tillage frequency is reduced, but little research has been done in California to measure this expected increase over time under minimum tillage.

Soil carbon sequestration is a multiyear process because most carbon in biomass that is incorporated into soils, or simply falls to the soil surface, is decomposed (released as CO2). It can require several years to get good measures of how much carbon is retained in the soil by a changed tillage or cover cropping practice (in this case nearly a decade).

At the same time we have tracked annual “above-ground net primary productivity” as carbon (ANPP-C). This is a measure of how much carbon in biomass is grown (for example wood and canes), in essence, how much carbon is retained following photosynthetic CO2 assimilation and respiration. While we can gather measures of ANPP-C relatively easily, it is much more difficult to acquire root production belowground (BNPP-C), but we have assessed some root production.

By definition, a vineyard carbon footprint should account for production of other GHGs, primarily N2O and CH4. The International Panel on Climate Change uses conversion factors for CH4 and N2O to calculate their global warming potential (GWP) in the same units as CO2 (reference 2). Per molecule, N2O is 298 times more potent than CO2 as a GHG, while CH4 is 25 times stronger on a 100-year time horizon. Thus a substantial amount of vineyard carbon sequestration can be required to offset a seemingly low level of N2O production.

Both CH4 and N2O are produced by microbes in soil through the processes of nitrification, denitrification and methanogenesis. In upland soils under nonwater- saturated conditions, methane can also be consumed through methanotrophy, which oxidizes CH4.

For three years, we carefully quantified CO2, N2O and CH4 emissions from soils of the vineyard’s three tillage treatments. Measurements of gas fluxes were taken at least once every two weeks, and often more frequently, using static (N2O and CH4) and dynamic (CO2) gas-flux chambers.

CO2 data illuminate the relationship between tillage and loss of soil carbon. Nitrous oxide (N2O) emissions show peaks of production as a result of nitrogen fertilization and rain. Quantification of such event-related emissions is crucial to understanding a farming system’s carbon footprint. Methane should be less important in scale than the other two gases; however, more research is needed.

Major results

Although we are in the 10th year of this investigation, the most comprehensive farm-gate carbon footprint for the treatments outlined above was assembled during 2008–10, when the American Vineyard Foundation was supporting the project. With a farm-gate footprint we describe cultivation practices and operations in grape production up to the point when grapes leave the vineyard, and no other processing that occurs beyond the vineyard, as in wine production.

Changes in biomass production: Aboveground net primary production of carbon (C) is generally defined as the net flux of CO2-C from the atmosphere into green plants per unit time. Comprehensive measurements of biomass showed that average ANPP-C for the vineyard across all treatments was about 3.43 metric tons of carbon per hectare per year in 2009 (mT C ha-1 year-1). In 2010, a wetter year, an average ANPP of 3.55 mT C ha-1 year-1 was produced. These sums include carbon in the fruit.

Figure 1: Minirhizotron images capture the maturation of grapevine roots over two weeks in July.

Figure 1: Minirhizotron images capture the maturation of grapevine roots over two weeks in July.

In 2009, the conventional tillage treatment with a barley cover crop assimilated 18% more carbon than the minimumtilled cover crop, while the difference jumped to 37% in 2010. In the 2009 fruit harvest, approximately 17.4% of ANPP-C, or about 0.6 mT C ha-1 year-1 was removed from the vineyard, whereas in 2010 this percentage was somewhat higher at 27.3%, or roughly 1.0 mT C ha-1 year-1.

Biomass carbon other than harvested fruit is either retained as wood in trunk and cordon growth or incorporated into the tractor row by mowing and tillage (canes and leaves).

Figure 2: RBM Grape root biomass distributions (kg per m3) measured at five 30 cm depths in the three tillage treatments. Fine root biomass (less than 2 mm) increased in the top plow (Ap) horizon under minimum tillage.

Figure 2: RBM Grape root biomass distributions
(kg per m3) measured at five 30 cm
depths in the three tillage treatments. Fine root
biomass (less than 2 mm) increased in the top
plow (Ap) horizon under minimum tillage.

A decline in pruning weights and yield under minimum tillage was noted during the course of the experiment. In 2009 this treatment produced 29% less fruit and 25% less ANPP-C than the tilled cover crop treatment. In 2010, these discrepancies progressed to 33% and 28% less. Yield under conventional tillage was intermediate in both years. The “devigoration” noted in the minimum-till treatment offset about 5% of the carbon it sequestered into soils, because vines were accumulating less woody biomass.

It is surmised that this devigoration was due mainly to water competition, because the cover crop consumes water and is likely to compete with vines in early spring. The wider differences seen in 2010 may be the result of long-term competition in the eighth year of treatment.

The impact of these changes in production in the minimum-tillage treatment depends on site characteristics. This is a moderate vigor site where reduction in vigor is somewhat desirable, so yield reductions may contribute to improved quality. Overall yields in 2009 and 2010 ranged from approximately 1.7 to 3.9 tons per acre, with lowest yields from the minimum-tillage treatment and highest yields from the tilled cover crop treatment. See Figure 1.

Below ground carbon sequestration into roots: Several key findings emerged from efforts to quantify the influence of minimum tillage on grape root biomass production and distribution. In Figure 2 are root biomasses at 30 cm intervals of soil depth. Results indicated that with the cessation of tillage in the minimum-tilled treatment, roots grew back into the soil horizon that was previously disturbed annually, best characterized as an Ap or plow horizon. Much of this additional growth was in the form of fine roots.

This supports the importance of maintaining these treatments for long time periods. It is likely that roots proliferate into this horizon in spring prior to tillage or may grow between the first tillage event (April) and second event (May). Thus root biomass changes were increasing carbon sequestration for the minimum tillage treatment. See Figure 2.

Carbon sequestered into soil organic carbon pool

A long-term increase of carbon in soils occurred under the minimum-tillage treatment. Between 1991 and 2003, soil carbon content (% by weight) remained constant without a significant change in the Ap horizon (0 to 15 cm).

Figure 3: Total soil carbon in samples taken from the plow horizon (depth of 15 cm) showing changes over seven years of management of a barley cover crop (CC) under minimumtillage conditions (left bar), twice per year tilled cover crop (center bar), and once per year conventional-till/weed management treatment (right bar).

Figure 3: Total soil carbon in samples taken from the plow horizon (depth of 15 cm) showing changes over seven years of management of a barley cover crop (CC) under minimum tillage conditions (left bar), twice per year tilled cover crop (center bar), and once per year conventional-till/weed management treatment (right bar).

After seven years (2003 – 10) of imposed minimum tillage, soil carbon in the plow horizon of the minimum-tilled rows had significantly increased by an average of 8.4%. Taking into account a soil bulk density at approximately 1.32 grams cm-3, and the drip zones where no change is expected, this translates into an overall sequestration of 2,640 kg carbon per hectare; annually, 377 kg C or 1,383 kg CO2 were removed from the atmosphere.

The tilled-only once per year treatment, which was a control, also showed increased soil carbon, so treatment differences had to be adjusted. A slight decline in soil carbon was reported for the treatment tilled twice per year, although it was not statistically significant. (See Figure 3)

Grape production and ripening

Yield generally declined following initiation of the trial (2004 – 06). In 2009 and 2010, yields were also low, from 1.8 tons to 3.3 tons per acre. Production from minimum-tilled plots was consistently lowest among the three management practices. In all treatments, canopy size (leaf production) declined, which was reflected in pruning weights of one-year old canes, and was significantly lower under minimum tillage over the course of the study.

There were no significant differences in individual berry weight, pH, or °Brix for the 2009 and 2010 vintages among the three tillage treatments.

The decreased yield and growth in the minimum tillage treatment seemed to be primarily due to long-term water stress (reference 7). Most water consumed by these vines in summer is derived from deeper soil reservoirs and late spring rains. In the summer of 2009, after a winter of low rainfall, significantly greater water stress occurred in the minimum-tilled vines, and it occurred early in the summer (see Figure 4). In contrast, in the subsequent year, after a winter of heavy rainfall, there were no significant differences (reference 7).

Figure 5: Soil respiration in 2009-10. The importance of rain is clear in comparing the profile of 2009 to 2010, the latter being a year of much higher rainfall. Also evident are the contribution of root flush and higher temperatures in the spring.

Figure 5: Soil respiration in 2009-10. The importance of rain is clear in comparing the profile of 2009 to 2010, the latter being a year of much higher rainfall. Also evident are the contribution of root flush and higher temperatures in the spring.

Production and consumption of greenhouse gases:

Figure 4: Mid-day vine water potentials (MPa) from 101-14 Mgt vines. Significant differences were seen in summer 2009 but none in 2010. Samples were taken between 1 and 3 pm on 16 vines per treatment, and significance was declared using standard errors of the mean.

Figure 4: Mid-day vine water potentials (MPa) from 101-14 Mgt vines. Significant differences were seen in summer 2009 but none in 2010. Samples were taken between 1 and 3 pm on 16 vines per treatment, and significance was declared using standard errors of the mean.

Carbon dioxide from soil respiration. We monitored the flux of CO2 from the soil, commonly referred to as soil respiration respiration (Figure 5). The majority of this flux comes from microbial decomposition of organic matter, but plant roots also produce a good deal of CO2 during respiration of the carbon-based sugars produced in leaves during photosynthesis and transported to belowground.

Soils are constantly accumulating carbon but also oxidizing and losing carbon as CO2. Sequestration requires the soil carbon pool to be refreshed at a higher rate than carbon is lost. The loss of soil carbon is accelerated after each tillage event, as formerly inaccessible organic matter in soil aggregates is mobilized and consumed by microorganisms.

Other events like rainfall can accelerate soil carbon cycling and loss as CO2. It was therefore particularly important to monitor soil respiration following rainfall and tillage events and during the spring root-flush.

Our results show a close connection between total rainfall and soil respiration for this Mediterranean-climate vineyard: In 2010 there was 56% more rain than in 2009, and 67% higher soil respiration. The higher precipitation in spring 2010 was particularly important because it led to greater vine root growth, evident in heightened soil respiration in the herbicide-treated drip zone.

These observations highlight how dynamic the soil carbon pool is and how the equilibrium state of carbon gain and loss depends upon soil moisture. Soil nitrogen can control the rate of soil respiration since soil microbial organisms more easily consume organic matter with a low C:N ratio (6-12), leading to greater growth and respiration of organic matter and root exudates (reference 4).

Figure 6: Average vineyard N2O-N effluxes over two years (2009-10), showing especially the importance of first fall rains in October and KNO3-fertigations in late June/early July. Also evident are rains after tillage in May of both years -- and April of 2010.

Figure 6: Average vineyard N2O-N effluxes over two years (2009-10), showing especially the importance of first fall rains in October and KNO3-fertigations in late June/early July. Also evident are rains after tillage in May of both years — and April of 2010.

Overall results indicated that the tilled cover-cropped tractor rows had the lowest soil respiration rates. This was not unexpected, given this treatment also had the lowest soil carbon content. As soil organic carbon levels decrease, soil respiration will also generally decrease. In contrast, in the minimum-tilled tractor rows, undisturbed grape roots and increased root biomass from the barley cover crop seem to have produced higher soil respiration.

The tilled cover crop treatment did not appear to sequester as much carbon as the conventional tillage treatment, where resident vegetation (weeds) were tilled once per year. These observations point to the lasting effects of soil environmental changes created by tillage, more than to the loss of soil carbon witnessed directly after tillage.

Nitrous oxide (N2O) emissions from vineyard soils

Figure 8: Proportional sources of N2O emissions from the vineyard’s soil, averaged across tractor rows and drip zones in 2009–10. “Seasonal Rains and 48 hours after” include all rain events and the following two days, from the period after the first fall rainstorm and before the year’s final rain event, which occurred after spring tillage. Although N-fertigation at 7.5–15 lbs/acre accounted for a small fraction of yearly N2O emissions, vineyards with higher application of N may benefit in future from more sophisticated N-fertigation practices, such as micro-sprinklers, or improved management of N-pulses and N-concentrations in fertigation.

Figure 8: Proportional sources of N2O
emissions from the vineyard’s soil, averaged across tractor rows and drip zones in 2009–10. “Seasonal Rains and 48 hours after” include all rain events and the following two days, from the period after the first fall rainstorm and before the year’s final rain event, which occurred after spring tillage. Although N-fertigation at 7.5–15 lbs/acre accounted for a small fraction of yearly N2O emissions, vineyards with higher application of N may benefit in future from more sophisticated N-fertigation practices, such as micro-sprinklers, or improved management of N-pulses and N-concentrations in fertigation.

For N2O emissions, a field assessment was made at least twice per month. But climatic or cultural events with a clear potential to cause high short-term emissions, such as fertigation, tillage and precipitation, were measured more frequently —often only hours apart (Figure 8).

Nitrous oxide emissions are often characterized by short intervals of very high emissions in areas where resources like water and nitrogen converge, in particular when irrigation or precipitation water saturate soil pores and low oxygen conditions prevail.

Vineyards represent complex spatial environments since the drip zone is farmed in one manner (herbicide applications and drip irrigation), while the tractor rows are farmed in another manner (tractor cultivation, seeded cover crops and dry land conditions, for example). In the drip zone, most emissions come from point sources centered under the emitter and during times of drip fertigation and irrigation.

N2O emissions across a given tractor row area were similar and uniformly elevated mainly after early fall rain events but also immediately following cultivation. The highest rates of N2O emission were seen from the tilled cover crop tractor rows, despite the common belief that tillage will reduce N2O production since it increases aeration.

We believe the increase in N2O emission observed here was due to the pooling of water and organic matter in a plow-pan that developed in this treatment, aided by the soil’s high clay fraction. If true, this would be an unintended, negative consequence of tillage.

To describe and quantify the emissions from soil under the drip emitter required careful work that, to our knowledge, has not been approached by other researchers. Using smaller gas-collection chambers spaced at various distances from drippers, previous work in our laboratory found that emission “plumes” (emission patterns in 3-D around a point source) from drip fertigation could be described using two-dimensional Gaussian distribution patterns.

In 2010, we observed that “chasing” N with irrigation water (a common practice) led to more complex emission patterns. In this case, the zone of major emissions took place on “shoulders” around the center of the drip zone, where a high-nitrate irrigation solution was present. Initially, emissions were lower directly under the dripper where chasing water had diluted applied NO3-, but, over time, the plume became more even as NO3- and water presumably diffused through the drip zone (Figure 7).

Overall, our results have not suggested great opportunities to mitigate N2O emissions in wine grape vineyards with low N applicati ons. However, vineyards with higher application of N may benefit in the future from more sophisticated N-fertigation practices, such as micro-sprinklers, or improved management of N-pulses and N-concentrations in fertigation.

We are pursuing further research to describe how total N2O emissions from N-fertilization are affected by patterns of application. Applying N in lower concentrations at higher frequencies, or at earlier and cooler points in spring or summer when irrigation is not necessarily needed may result in lowered N2O emissions. Such investigation must also consider the potential for increased leaching of nitrates (NO3-) into groundwater.

Figure 7: Spatial models of emissions after drip fertigation with 7.5 pounds N per acre using KNO3 show how the N2O plume in the drip zone can be quantified spatially. Left panel shows a plume resulting from continuous N application in 20 gallons of fertigation solution over six hours. Right panel shows effects when nitrogen applied through the drip system was pulsed at hour-3, then chased with two hours of irrigation water (REPRINTED WITH PERMISSION OF THE AMERICAN CHEMICAL SOCIETY, OR, from D.R. Smart et al. N2O emissions and water management in California perennial crops. pp. 227-255 In Guo, L, AS Gunasekara and LL McConnell (Eds.) Understanding Greenhouse Gas Emissions from Agricultural Management, American Chemical Society, Baltimore, Md.

Figure 7: Spatial models of emissions after drip fertigation with 7.5 pounds N per acre using KNO3 show how the N2O plume in the drip zone can be quantified spatially. Left panel shows a plume resulting from continuous N application in 20 gallons of fertigation solution over six hours. Right panel shows effects when nitrogen applied through the drip system was pulsed at hour-3, then chased with two hours of irrigation water (REPRINTED WITH PERMISSION OF THE AMERICAN CHEMICAL SOCIETY, OR, from D.R. Smart et al. N2O emissions and water management in California perennial crops. pp. 227-255 In Guo, L, AS Gunasekara and LL McConnell (Eds.) Understanding Greenhouse Gas Emissions from Agricultural Management, American Chemical Society, Baltimore, Md.

Methane fluxes

Methane (CH4) is a major greenhouse gas, responsible for approximately 30% of radiative forcing, the driving mechanism behind global warming (reference 2). Methane has a longer lifespan in the atmosphere than CO2, but a much shorter lifespan than N2O, partly because it is readily oxidized (consumed) by soil bacteria in many upland farming and natural ecosystems (reference 1).

Fostering CH4 oxidation may be possible in cultivated upland perennial crops like grapes. Two years of data indicated that a small amount of CH4 was oxidized in tractor rows. Treatment differences due to tillage were not apparent.

Emissions are sometimes seen following rain, although they are difficult to predict or to model. We have seen consistent, high emissions of CH4 during KNO3 fertigation. Like N2O production, this may be related to the establishment of low-oxygen or anaerobic conditions, which is a requirement for methano­genesis.

Vineyard operations

A farm-gate carbon footprint for a vineyard must include tractor and other fuel-consuming operations. Table I (below) lists the CO2 equivalent production of each of the operations conducted in the field during the two years of interest when we were conducting intensive research on N2O and CH4 fluxes. The operations component of the GHG footprint was highly dependent on disease pressure (fungicide and sulfur dust applications) and cultivation practices of chopping, mowing and tillage.

Table I: Kg of CO2 released per hectare in each vineyard management operation two years for three tillage systems.

Table I: Kg of CO2 released per hectare in each vineyard management operation two years for three tillage systems.

Conclusion

We are close to achieving a total carbon footprint for a number of test case research vineyards. Once we have confirmed the carbon equivalent cost of the herbicides and fungicides used in this investigation the footprint should be forthcoming, but we are exercising caution in the reporting process.

It has been important to investigate the capacity of a high C:N cover crop such as dwarf barley to sequester carbon when accompanied by minimum tillage practices (shallow tillage every other year). The use of the barley cover crop under conventional tillage practices (including an “extra” tillage pass for seeding bed preparation in the fall), did not lead to measurable carbon sequestration. Presumably the increased disturbance and aeration of the soil counterbalanced the higher inputs of organic carbon into the soil.

If the fuel emissions from tillage passes are included, both conventionally tilled treatments had negative effects on carbon sequestration. Still, the most frequently tilled treatment outperformed the other treatments in fruit yield, probably due to lower competition for water from the cover crop.

The results obtained serve as a good comparison to D.S. Kroodsma and C.B. Field’s work on carbon sequestration in California agriculture, in which they estimated that vineyards currently sequester 240 kg C ha-1 yr-1, and could sequester above 480 with cessation of tillage (reference 3).

Including woody NPP as sequestered carbon, our study found C sequestration under all treatments. However, if we anticipate the decomposition of, or burning of vines at the end of the vineyard’s life span, we found only minimum tillage sequestered C, at an averageof 374 kg C ha-1 yr-1. This was because minimum tillage resulted in carbon sequestration into soil. N2O emissions constituted between 6% and 17% of net GWP in the latter case. CH4 oxidation offset about 1% of net GWP.

We estimate that with seven years of carbon sequestration into these soils, minimum tillage offset about 35 years of potential GHG emissions from its own management and fertilizer requirements. It is impossible to say at this time when the treatment will stop adding carbon to the soil, but we can expect that the rate will be considerably reduced as the soil carbon pool becomes saturated.

What might the future bring for carbon footprints in vineyards? The question is raised whether any other cover crops might successfully sequester carbon under conventional tillage practices. This question stands apart from the use of leguminous cover crops to provide nitrogen to grapevines. But we expect that the two questions will converge as we learn better management strategies for cover crops in vineyards.

Likewise, the inclusion of pomace as a soil amendment could affect fertility and soil carbon sequestration. New possibilities may arise for the use of prunings and vineyard wood. Vineyard carbon footprints could be significantly affected if these are used to generate energy, or else converted into products such as biochar, which allow long-term fixation of carbon.

Bibliography

  1. Chan, A.S.K., Parkin, T.B., 2001 “Effect of land use on methane flux from soil.“ J. Env. Qual. 30, 786–797.
  2. IPCC, 2007 Technical Summary 2.5. In: 1, W.G. (Ed.), Fourth Assessment Report: Climate Change 2007. Intergovernmental Panel on Climate Change.
  3. Kroodsma, D.A., Field, C.B., 2006 “Carbon sequestration in California agriculture, 1980–2000.” Ecol. Appl. 16, 1975–1985.
  4. Lal, R., 2004 “Soil carbon sequestration impacts on global climate change and food security.” Science (Washington D. C.) 304, 1623–1627.
  5. Liljeroth, E., Schelling, G.C., Van Veen, J.A., 1990 “Influence of Different Application Rates of Nitrogen to Soil on Rhizosphere Bacte ria.” Netherlands J. Ag. Sci. 38, 255–264.
  6. Carlisle, E., Smart, R., Bronde, J., Arnold, A. 2009. “Carbon Footprints of Vineyard Operations.” Practical Winery & Vineyard September/October, 15–25.
  7. Smart, D.R., Schwass, E., Lakso, A., Morano, L., 2006 “Grapevine rooting patterns: A comprehensive analysis and a review.” Am. J. Enol. & Vit. 57, 89–104