Disentangling feed conversion efficiency and feed sourcing in global livestock water footprint trends, 1972–2023

Disentangling feed conversion efficiency and feed sourcing in global livestock water footprint trends, 1972–2023 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Disentangling feed conversion efficiency and feed sourcing in global livestock water footprint trends, 1972–2023 Mesfin Mekonnen, Betelhem Demeke, Fitsume Wolkeba This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9360861/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Livestock water use is dominated by the water required to grow feed, yet long‑term, globally consistent reconstructions remain limited. We compile a country‑level, animal‑specific assessment of green and blue water footprints (WFs) for seven livestock categories across 1972–2023 using a harmonized accounting framework linking feed requirements, trade‑weighted feed WFs, and livestock production statistics. Global livestock WF increased from 1,979 km³ yr⁻¹ in 1972 to 3,044 km³ yr⁻¹ in 2023. Ruminants remain dominant water users, with beef cattle (34%) and dairy cattle (25%) contributing the most. In 2023, feed production accounted for 98% of total livestock WF, while direct water uses were 2%. China showed the largest increase (147→586 km³ yr⁻¹; +299%) and contributed 19% of the global total livestock WF; the top five countries accounted for 44%. Unit WFs declined markedly (e.g., pork −70%, eggs −68%, chicken −66%), yet unsustainable surface and groundwater use contributed 4% of the total and 52% of the blue WF. Earth and environmental sciences/Hydrology Physical sciences/Engineering/Civil engineering Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Livestock systems are central to global food security and rural livelihoods, supplying approximately 18% of global calories and 42% of protein while supporting the livelihoods of more than one billion people worldwide (Alexandratos and Bruinsma, 2012; FAO, 2025). At the same time, livestock production is among the most resource‑intensive components of food systems, placing substantial demands on land and freshwater resources, much of which are embodied in animal feed (Alexandratos and Bruinsma, 2012; Heinke et al., 2020; Mekonnen and Hoekstra, 2012; Poore and Nemecek, 2018; Steinfeld et al., 2006). Livestock occupies roughly 30% of the Earth’s ice‑free land surface and relies on grasslands and croplands to supply animal biomass intake, with pronounced regional and system‑level variation in feed efficiency and production practices (Herrero et al., 2013; Steinfeld et al., 2006). These biophysical differences translate into large disparities in environmental burdens and improvement potential across species and production systems. Water is a particularly salient constraint. Agriculture accounts for roughly 70% of global blue water withdrawals, and livestock‑related water use is dominated by the water required to produce feed, with drinking and service water playing comparatively minor roles (Mekonnen and Hoekstra, 2012). Global assessments estimate that livestock feed production consumes approximately 2,100–4,400 km³ yr⁻¹ of green and blue water, corresponding to up to about 40% of agricultural consumptive water use (de Fraiture, 2007; Heinke et al., 2020; Mekonnen and Hoekstra, 2012; Weindl et al., 2017). These estimates underscore the tight coupling between livestock water use, feed sourcing, and productivity. At the same time, methodological differences in accounting for green, blue, and grey water, and in reporting volumes versus impact‑based metrics, complicate comparisons across studies and can yield divergent results for similar products (Boulay et al., 2021; FAO, 2019). A substantial body of literature has quantified the water footprints of animal products and identified feed conversion efficiency, feed composition, and feed origin as dominant drivers of unit water footprints across species and production systems (Gerbens-Leenes et al., 2013; Heinke et al., 2020; Mekonnen and Hoekstra, 2012; Mekonnen et al., 2019; Poore and Nemecek, 2018; Wisser et al., 2024). Seminal work showed that animal products generally exhibit higher unit water footprints than nutritionally comparable plant‑based foods (Mekonnen and Hoekstra, 2012). Subsequent analyses documented strong gradients from beef to pork to poultry. They demonstrated how transitions from grazing to mixed and industrial systems simultaneously improve feed efficiency while often increasing reliance on irrigated concentrate feeds, thereby shifting the balance between green and blue water use (Gerbens-Leenes et al., 2013; Heinke et al., 2020). Despite these advances, the feed side of livestock water use, the amount of feed required, its composition, and source, remains incompletely resolved in a long‑term, country‑specific context. Global datasets have quantified biomass use and feed efficiencies across regions and systems, highlighting the central role of grasslands and mixed crop–livestock systems (Herrero and Thornton, 2013). More recent efforts provide national‑scale water footprint estimates for feed items in the late 2010s and increasingly link feed production, hydrology, and trade (Govoni et al., 2024; Wisser et al., 2024). However, these datasets do not resolve country‑ and animal‑specific feed composition as a consistent time series spanning multiple decades, limiting the ability to reconstruct the historical co‑evolution of feed baskets, feed sourcing (rainfed versus irrigated), and livestock productivity. Evidence from regional case studies illustrates the importance of this historical perspective. In the United States, for example, the water footprint of meat and milk production declined markedly between 1960 and 2016 due to improvements in animal productivity, feed conversion efficiency, and crop yields that reduced the water footprint of feed ingredients (Mekonnen et al., 2019). Comparable, globally consistent reconstructions remain scarce, constraining efforts to disentangle structural changes in livestock systems from yield‑driven reductions in feed water requirements and to benchmark trajectories across countries. Against this backdrop, three key gaps persist: (i) limited temporal coverage of country‑level, animal‑specific feed use and composition; (ii) the absence of harmonized, long‑run green and blue water footprint estimates for livestock production; and (iii) insufficient attribution of changes in unit water footprints to evolving feed baskets, feed conversion efficiencies, and crop water requirements. The current study addresses these gaps by assembling a global, country‑level, animal‑specific time series of feed use and composition spanning 1961–2023, harmonized with FAOSTAT production statistics and informed by the livestock systems literature. We use this framework to (i) trace the evolution of livestock feed baskets across species and countries; (ii) estimate green and blue consumptive water use attributable to livestock production consistently over 1972–2023; and (iii) derive unit water footprints that decompose temporal changes into contributions from feed composition, feed conversion efficiency, and crop water requirements. By providing a unified six‑decade perspective, this analysis complements existing global snapshots and recent feed water footprint datasets, offering new insights into how nutrition, technology, and agronomy have jointly shaped the water intensity of animal‑source foods. Results Global water footprint of livestock production (1972–2023) The total consumptive water footprint (WF) of global livestock production increased by 54% over the past five decades, increasing from 1,979 km³ yr⁻¹ in 1972 to 3,044 km³ yr⁻¹ in 2023 (Fig. 1). This long‑term growth trend and the relative increases across livestock categories are depicted in Fig. 1a, while the 2023 structural composition of WF by feed sources and water types is summarized in Fig. 1b. Detailed WFs by feed sources, livestock type, and water sources are provided in Table-S1 (Supplementary Information). Breaking down by species, ruminants (beef cattle, dairy cattle) remain the dominant contributors to global livestock WF, while broilers show the fastest relative growth. The evolution of species‑level totals and percentage changes are visualized in Fig. 1a; exact species totals and shares for the period 1972 to 2023 are reported in Table S1.2. At the national level, growth patterns were highly heterogeneous (Fig. 2). China exhibited the strongest absolute increase, rising from 147 km³ yr⁻¹ to 586 km³ yr⁻¹ (+299%), and overtaking the United States to become the world’s largest contributor (19% of global WF in 2023). India (+259%) and Brazil (+136%) also showed large increases consistent with strong expansion in dairy and meat demand. In contrast, the United States reduced its livestock WF by 13%, reflecting efficiency gains, shifts toward more productive production systems, and improvements in feed conversion (Mekonnen et al., 2019). Pakistan showed the largest relative growth (+605%), though from a low baseline. Together, China, India, Brazil, the United States, and Pakistan accounted for 44% of global livestock WF in 2023 (Fig. 2). Detailed WFs per country, livestock type, and water sources are provided in Table-S2 (Supplementary Information). The feed‑source decomposition reveals the structural drivers behind these changes (Fig. 3). Pasture remains the single largest component (48% of total WF in 2023), growing by 83% since 1972, while maize now contributes the largest WF among feed crops (495 km³ yr⁻¹ in 2023). The most dramatic relative increase occurred for soybean (≈20‑fold since 1972), reflecting global shifts toward concentrate‑rich rations. The 2023 partitioning of green versus blue water, and the sustainability categories within blue water, is illustrated in Fig. 1b, which links water sources to feed streams and livestock outputs. The results confirm that (i) livestock scaling, not just yield or feed efficiency, remains the dominant driver of global WF growth, (ii) rapid shifts toward concentrate‑heavy diets amplify pressures on cropland water resources, and (iii) large increases in maize and soybean WFs mirror the global rise in mixed and industrial production systems. Unsustainable blue water footprint of livestock production The blue component of the livestock water footprint (WF) shows a marked shift toward unsustainable sources over the past five decades. Aggregated across countries, total blue WF rose from 158 km³ yr⁻¹ in 1972 (45% surface–sustainable, 23% groundwater–sustainable, 8% surface–unsustainable, 23% groundwater depletion/overdraft) to 252 km³ yr⁻¹ in 2023 (41% surface–sustainable, 6% groundwater–sustainable, 12% surface–unsustainable, 41% groundwater depletion/overdraft), indicating that unsustainable blue WF increased by 2.7 times (from 50 to 133 km³ yr⁻¹, +170%). Country contributions to the global unsustainable blue total evolved unevenly (Fig. 4). China’s absolute unsustainable blue volume more than doubled (from 15 to 33 km³ yr⁻¹), yet its share of the global unsustainable total fell from 30% (1972) to 24% (2023) because other countries, especially those with rapidly expanding groundwater dependence, grew faster. India moved in the opposite direction: its unsustainable blue volume rose from 2 to 25 km³ yr⁻¹, and its contribution to the global unsustainable total almost quadrupled from 5% to 19%, reflecting strong growth in groundwater depletion/overdraft for livestock feed and production. The US maintained almost similar absolute magnitudes but reduced its global contribution from 24% to 10%, indicating that, relative to worldwide trends, US livestock systems now account for a smaller share of global unsustainable blue water use. Among other notable contributors, Spain increased unsustainable blue volumes (≈1→3 km³ yr⁻¹) while its share of the global unsustainable total reduced from 3% to 2%; Egypt rose from ~2 to 5 km³ yr⁻¹, with its global contribution decreasing from5% to 4%. Table S1 and Table S2 provide the sustainable and unsustainable surface and groundwater per feed and country, respectively. Taken together, the country‑level trajectories in Fig. 4 show rising groundwater depletion/overdraft dependence alongside differing national contributions to the global unsustainable blue total. Even where some large producers reduced their share of the global unsustainable blue total (e.g., China, United States), absolute unsustainable blue volumes generally increased; meanwhile, countries such as India account for an expanding portion of the global unsustainable blue WF, with growth driven especially by groundwater depletion/overdraft. This underscores that efficiency gains must be paired with basin‑level limits and source‑shifting in regions where livestock and feed production depend on unsustainable blue water. Unit water footprint of livestock products Unit water footprints declined substantially across most livestock products between 1972 and 2023 (Fig. 5). Notable reductions include pork (−70% in L kg⁻¹), eggs (−68%), and chicken meat (−66%), consistent with improvements in feed conversion efficiency, genetics, and management in intensive systems. Milk declined by ~52% across units, while cattle, sheep, and goat meat showed more modest decreases. The unit WFs of the major livestock products are provided in Table S3. Unit WF reductions are widespread and consistent with known global improvements in livestock productivity observed in the literature (e.g., Mekonnen and Hoekstra (2012) for US productivity trends). The fact that total WFs increased despite lower unit WFs underscores the overwhelming role of production scale compared to efficiency gains. Across Figs. 1, 2, 3, and 5, three consistent patterns emerge. First, scale effects dominate changes in total water footprints, as global livestock WF increased despite widespread declines in unit WF, indicating that growth in production outpaced efficiency gains. Second, changes in feed composition strongly shape water outcomes: increasing reliance on maize and soybean feeds, combined with the continued dominance of pasture, explains much of the observed spatial and temporal variation in livestock WFs (Figs. 1b and 3). Third, efficiency gains differ markedly across species, with monogastric systems achieving the largest reductions in unit WFs (Fig. 5), while ruminant production continues to account for the majority of global livestock WF due to its scale and dependence on pasture-based feeds (Fig. 1a). Drivers of Unit Water Footprints of Livestock Products Across all livestock types, the model explains the vast majority of observed variation in unit WFs (R² ≈ 0.96 for green, 0.99 for blue on average across products). The estimated elasticities are strikingly close to the multiplicative identity one would expect from first principles (i.e., product WF ≈ FCR × feed WF), with an additional positive effect of concentrate share: Feed conversion (FCR): A 1% increase in FCR is associated with a ≈ 1.02% increase in green unit WF and a ≈ 0.98% increase in blue unit WF on average across livestock types (mean =1.02, =0.98). These near‑unity elasticities confirm that, all else equal, feed use per unit output is the dominant lever on product WF, consistent with the multiplicative structure of livestock production systems. (Table S4 – summary of factor analysis) Weighted feed WF: A 1% increase in the weighted feed WF (m³ t⁻¹ feed) is associated with a ≈ 0.34% increase in green unit WF and a ≈ 0.76% increase in blue unit WF (mean =0.34, =0.76). This asymmetric response indicates that changes in feed sourcing and crop water intensity transmit much more strongly to blue water footprints, reflecting the dominant role of irrigated feed crops in shaping blue WF outcomes. (Table S4 – summary of factor analysis) Concentrate fraction: Holding FCR and feed WF constant, increasing the concentrate share in the ration raises product WFs, with semi‑elasticities ≈1.04 for green WF and ≈1.12 for blue WF. Interpreted at the mean, a +0.10 increase in concentrate share (i.e., +10 percentage points) is associated with roughly 10-11% higher unit WFs, consistent with concentrate‑rich rations sourcing disproportionately from water-intensive and irrigated feed crops. The magnitude of this effect varies by livestock type but is consistently positive across systems (Table S4 – summary of factor analysis) Per‑product regressions display the same pattern: both FCR and feed WF carry elasticities ~1, and the concentrate fraction is positively associated with unit WFs for every product (high R² in all cases). Full coefficients appear in Supplementary Table S4. These estimated elasticities provide the structural basis for attributing long‑run changes in unit WFs. In the sections below, we combine the estimated factor sensitivities with observed historical changes in feed conversion, feed water intensity, and concentrate share to decompose global unit WF trends between 1972 and 2023. Interpretation and linkage to prior work These results are fully consistent with the biophysical logic and previous global assessments: Scaling with feed demand : Because animals must consume feed to produce output, FCR reductions (through genetics, health, management) translate almost one‑for‑one into lower unit WFs. This is the same mechanism underlying long‑run improvements observed in the United States (1960–2016) (Mekonnen et al., 2019) and elsewhere, where better feed conversion contributed materially to lower unit WFs despite rising total production. (Table S4 – summary of factor analysis) Feed sourcing and composition : Weighted feed WF captures both crop choice and geography. Concentrate feeds (e.g., maize, soy), particularly when irrigated, increase blue WFs even when FCR improves, consistent with prior findings on the water cost of diet intensification. System‑level implications : The positive concentrate coefficient does not imply that intensification is inherently water‑inefficient; rather, it highlights that efficiency gains must be paired with water‑aware feed sourcing to avoid shifting pressure from green to blue water resources Attribution of Long‑Run Changes (1972–2023) We decomposed global changes in unit water footprints (WFs) between 1972 and 2023 using the log‑additive decomposition shown in Eq. S8. Fig. 6 displays the global contributions by component, and the exact values are provided in Table S5. Fig. S1 displays contributions by components for major countries, and Table S6 provides exact values for these countries. Monogastrics (Pig, Chicken, Egg) Substantial reductions in both green and blue unit WFs are observed: Chicken: −67% (green), −54% (blue), driven by FCR improvements (~−35 pp) and declining feed WFs (~−24 pp green; ~−7.3 pp blue). Concentrate changes contribute an additional −2% to −5%. Egg: −68% (green), −66% (blue), driven by FCR improvements (~−30 pp green; ~−31 pp blue) and declining feed WFs (~−23 pp green; ~−24 pp blue). Concentrate changes contribute an additional −2%. Pig: −70% (green) and −71% (blue), with large contributions from FCR (~−35 pp green; ~−28 pp blue) and feed WF (~−36 pp green; ~−48 pp blue). Ruminants (Cattle, Dairy, Sheep, Goat) Ruminant trends are mixed: Cattle: Green WFs decline (~−31%), and blue WFs decline (~−18%), driven by FCR improvement (~-12 pp) and rising feed WF (~+1.2 pp green; ~+10 pp blue). Dairy: Green WF declines (~−54%), but blue WF increases markedly (+35.8%), almost entirely due to rising feed blue WF (+51 pp), overwhelming FCR improvements (−30 pp). Sheep: Similar pattern: small green decline (~−19.7%) but increase in blue (~+11.8%) from positive feed blue contributions. Across livestock systems, three robust conclusions emerge. First, improvements in feed conversion ratios (FCRs) continue to be the most powerful mechanism driving reductions in unit water footprints. Second, the sourcing of feed—particularly the crop types used and the geographic regions in which they are produced—plays the dominant role in shaping long‑term trends in blue water footprints. Third, changes in the share of concentrate feeds can either offset or amplify these improvements, depending on the relative water intensity associated with their production. Discussion Comparison with previous studies A comparison of our updated consumptive water footprint (WF) estimates with major prior global assessments reveals both broad agreement at the global scale and substantial differences at the country and feed‑component levels. Earlier studies, particularly Mekonnen and Hoekstra (2012), provided the first comprehensive global WF benchmarks for livestock, combining country‑level feed composition, production systems, and feed conversion efficiencies. More recent work (Heinke et al., 2020) applied updated feed and crop‑water models to reassess livestock water use, estimating that approximately 3,374 km³ yr⁻¹ of green + blue water is required for global feed production. Our global estimates of total consumptive WF for livestock feed fall within the same order of magnitude as earlier work (Table 1). For the year 2000, the total global feed WF from this study (2,368 km³ yr⁻¹) closely matches Mekonnen and Hoekstra (2012) estimate for 1996–2005 (2,217 km³ yr⁻¹), Zimmer and Renault (2003) estimate for 2000 (2,340 km³ yr⁻¹), and Molden (2007) estimate for 2000 (2,152 km³ yr⁻¹), differing by only 6%, 1%, and 11%, respectively. This shows strong consistency in global‐aggregate WF magnitudes across methods and datasets. Table 1 . Comparison of global consumptive water footprint (WF) estimates (green + blue, km³ yr⁻¹) for livestock feed across major global studies. Values are shown for feed crops, pasture, and total consumptive WF for comparable reference years. Percent differences reflect the deviation of each study from the Current Study’s estimate. Study Period Feed crop Pasture Total % difference Current Study 2000 1249 1119 2368 Mekonnen and Hoekstra (2012) 1996-2005 1304 913 2217 6% Weindl et al. (2017) 2000 2170 2590 4760 -101% Molden (2007) 2000 1312 840 2152 9% Heinke et al. (2020) 1998-2002 1686 1688 3374 -42% de Fraiture et al. (2007) 2000 840 2152 2992 -26% Zimmer and Renault (2003) 2000 2340 1% Despite similar global totals, the partitioning between feed crops and pasture differs substantially across studies (Table 1). For 2000, our estimate of feed‑crop WF (1,249 km³ yr⁻¹) is -4% lower than Mekonnen and Hoekstra (2012), while pasture WF (1,119 km³ yr⁻¹) is 18% higher. These shifts reflect updated feed volumes, improved livestock productivity data, and revised pasture mapping. Other studies diverge more sharply, reflecting differences in data sources and modeling assumptions. For example, Weindl et al. (2017) reported total livestock‑feed consumptive WF of 4,760 km³ yr⁻¹, more than double our estimate, due in part to broader inclusion of extensive systems and differences in crop‑water modeling. Heinke et al. (2020) reported an estimate of 3,374 km³ yr⁻¹ for 2000 conditions, 58% higher than ours. This discrepancy likely stems from their use of the LPJmL4 global hydrological model, which simulates agriculture using broad crop functional types rather than explicit crop-specific representations, potentially reducing fidelity in modeling crop evapotranspiration relative to the crop model used here. Additionally, uncertainty in harvested area inputs—known to be a dominant source of variability in large‑scale agricultural water use assessments—further contributes to differences across studies (Demeke et al., 2026a). Comparison of our updated unit consumptive WF (green + blue) with those reported by Mekonnen and Hoekstra (2012) for major livestock products are shown in Fig.7. The 1:1 line illustrates near‑agreement for most products at the global mean. At the country scale (Fig. S2), deviations between our results and Mekonnen and Hoekstra (2012) become substantial, reflecting updated feed‑mix data, revised harvested areas, and national differences in feed supply chains. The most important contributors to these discrepancies are: Differences in annual feed volumes and diet composition, especially for monogastrics, where rapid intensification after 2000 altered concentrate use. Revised feed‑crop WFs, stemming from updated harvested areas, crop‑water requirements, and spatially explicit crop modeling. These are known to introduce uncertainty (see Demeke et al. (2026a)). Updated concentrate accounting, where FAO FBS data, now used to replace model‑based grain estimates, shifts the relative roles of grains versus pasture or crop residues. In sum, while our global‑aggregate WF estimates fall well within the envelope of prior studies, large differences emerge at the feed‑category and country levels. These differences reflect improvements in livestock productivity data, updated feed composition and FCR trajectories, and advances in crop‑water modeling. The overall picture aligns with recent analyses indicating that method selection, input data quality, and assumptions about feed sourcing heavily influence livestock WF outcomes. Policy relevance Understanding long‑term dynamics of feed use and livestock water footprints is not only scientifically important but also central to emerging policy frameworks aimed at improving water governance and food‑system sustainability. Because feed production accounts for the majority of livestock water withdrawals and is tightly linked to basin‑level blue water stress, robust national time‑series of green and blue water footprints provide critical evidence for designing water‑allocation caps, evaluating agricultural intensification strategies, and improving the long‑run resilience of livestock systems. Recent FAO LEAP guidelines emphasize aligning water‑productivity improvements with assessments of blue water scarcity, highlighting that mitigation strategies must consider both efficiency gains and their local hydrologic context (Boulay et al., 2021). Spatially detailed global analyses further show that reducing livestock water use alone will not eliminate water stress in many basins, underscoring the need for integrated land–water planning, crop choice optimization, and feed‑trade strategies tailored to local hydrologic limits (Gerbens-Leenes et al., 2013; Herrero and Thornton, 2013; Herrero et al., 2009; Hoekstra, 2014; Mekonnen and Hoekstra, 2012; Wisser et al., 2024). Additionally, because livestock products constitute a large share of agricultural water use and demand is projected to increase substantially toward mid‑century, countries must reconcile dietary shifts, productivity goals, and environmental constraints in ways that protect freshwater ecosystems while sustaining food and nutrition security. By generating consistent, multi‑decadal, country‑level estimates of feed use and livestock water footprints, this study offers actionable information for national governments, basin authorities, and international agencies seeking to implement evidence‑based policies that enhance water sustainability, strengthen climate resilience, and support equitable transitions in the livestock sector. Limitations Despite the use of harmonized global datasets and a consistent methodological framework to enable long‑term and cross‑country comparisons, several limitations should be acknowledged when interpreting the results. Scaling a global average feed conversion ratio (FCR) for 2015 across the 1972–2023 period using annual yield trends assumes a stable, linear relationship between yield and feed efficiency, which does not fully capture temporal and regional variability driven by changes in feed composition and quality, livestock genetics, animal health, management practices, and environmental conditions. As a result, historical improvements or declines in feed efficiency that are decoupled from yield trends may not be fully represented. In addition, livestock production systems (grazing, mixed, and industrial) are not explicitly distinguished, unlike in previous studies, and livestock production is instead represented in an aggregated, country‑level manner. While this approach ensures internal consistency across countries and time, it masks important intra‑country heterogeneity in feed sources, water use intensity, and production efficiency, potentially smoothing system‑specific extremes. The national‑scale aggregation limits the representation of sub‑national variability, particularly in large and climatically diverse countries, where spatial mismatches between feed production and livestock consumption can influence water‑use patterns. Finally, the study inherits the limitations of the global model used. Specifically, the limitation in groundwater surface water partitioning is affected by the specific groundwater representation in the model used. Together, these limitations highlight the need for future assessments that integrate system‑specific feed efficiencies and finer spatial resolution to better capture structural and technological changes in livestock production. Conclusion This study provides a long‑term, globally consistent assessment of livestock feed requirements and associated resource implications over the 1972–2023 period, offering new insights into how changes in agricultural productivity and feed conversion efficiency shape the water footprint of animal production. By harmonizing multi‑decadal yield data with feed conversion metrics, the analysis reveals pronounced temporal and spatial contrasts in feed demand and efficiency gains across countries, highlighting the uneven pace at which technological and structural changes in livestock systems have unfolded. The results underscore the importance of accounting for historical dynamics when evaluating contemporary livestock sustainability, as present‑day efficiencies often reflect cumulative investments in crop and livestock productivity, management, and trade rather than uniform global progress. At the same time, the observed divergence among regions suggests that aggregate improvements can coexist with persistent inefficiencies and localized pressures on water resources. By quantifying these long‑term patterns within a unified framework, this work complements existing snapshot‑based assessments and strengthens the empirical basis for linking livestock production, food security, and water sustainability (Herrero et al., 2009; Herrero et al., 2020; Steinfeld and Gerber, 2010). Overall, the study advances understanding of the systemic drivers underlying global livestock feed demand and its water implications, providing a transparent reference point for future efforts that seek to incorporate production‑system detail, sub‑national variability, and dynamic technological changes into integrated food–water sustainability assessments. Methods Water footprint accounting framework: We quantified the green and blue water footprints of livestock production using a harmonized accounting framework consistent with the Water Footprint Assessment methodology (Hoekstra et al., 2011). The analysis covers seven major livestock categories—beef cattle, dairy cattle, pigs, sheep, goats, broiler chickens, and layer chickens—and spans the period 1972–2023 for all countries. Total livestock water footprints were decomposed into feed‑related, drinking, and service water components, with feed production accounting for the dominant share of consumptive water use (Chapagain and Hoekstra, 2003; Hoekstra et al., 2011; Mekonnen and Hoekstra, 2012). Feed requirements and feed composition: Livestock feed requirements were reconstructed using feed conversion ratios anchored to FAO‑based estimates (FAO, 2022) and extended over time using observed changes in per‑animal productivity. Feed baskets were allocated across ingredient groups using livestock feed composition data and reconciled with FAO food balance statistics (FAO, 2025) to ensure mass balance and consistency over time. This approach captures long‑term changes in total feed demand while reflecting shifts between concentrate and non‑concentrate feed sources at the country level. Water footprints of feed and livestock products: Country‑specific green and blue water footprints of feed crops from Demeke et al. (2026b) were combined with reconstructed feed quantities to estimate feed‑related water use. For traded feed commodities, water footprints were calculated as weighted averages of domestic production and imports. Unit water footprints of livestock products were then derived by linking feed quantities, feed water intensities, and animal productivity in a consistent accounting framework. Partitioning sustainable and unsustainable blue water footprints of feed crops: We partitioned the blue water footprint (WF) of each feed crop into contributions from surface water and groundwater, using CWatM (Burek et al., 2020) outputs that attribute irrigation supply by source and identify groundwater depletion/overdraft (Wolkeba et al., 2024). We then classified the surface‑water component as unsustainable where blue water scarcity > 1, indicating that consumptive blue water use exceeds locally available water (natural runoff minus environmental flow requirements). These steps yield spatially and temporally consistent estimates of sustainable versus unsustainable blue WF for each feed crop, which are then aggregated to national feed baskets and livestock products. Attribution of temporal changes: To identify the main drivers of changes in unit water footprints over time, we applied country‑year panel regression models separately for each livestock product. The analysis attributed variation in unit water footprints to changes in feed conversion efficiency, the water intensity of feed baskets, and the share of concentrate feeds. This framework enables a consistent decomposition of long‑term trends across countries and livestock categories. Data sources and supplementary methods: All data were obtained from harmonized global sources, including FAOSTAT (FAO, 2022), FAO livestock and feed datasets (FAO, 2022), and recent global feed water footprint products (Demeke et al., 2026b). Water scarcity maps were obtained from Wolkeba et al. (2024). All equations and detailed methodological descriptions are provided in the Supplementary Information and are numbered independently as Eqs. (S1–S8). Declarations Data availability The output data is provided as a Supplementary Table. All input datasets used in the current study are publicly available from their respective original sources, as referenced in Methods and Supplementary Information. Code availability There are no special codes used in the current work. Any Python scripts used for handling large data in the current work are available upon request from the corresponding author. Acknowledgements We acknowledge the financial support received from the Global Water Security Center, Alabama Water Institute, the University of Alabama. Additionally, we would like to recognize the contributions of FAO's GLEAM team, particularly Dr. Dominik Wisser and Dr. Giuseppe Tempio, for providing the GLEAM dashboard data, which includes feed intake per livestock production system for the year 2015. Authors Contributions M.M.M. conceived the ideas, conducted the analysis, and led the writing of the paper; B.W.D. and F.T.W. provided the input data, reviewed, and edited the manuscript. References Alexandratos, N., Bruinsma, J., (2012) World agriculture towards 2030/2050: the 2012 revision, ESA Working paper No. 12-03. Food and Agriculture Organization, Rome. Boulay, A.-M., Drastig, K., Amanullah, Chapagain, A., Charlon, V., Civit, B., DeCamillis, C., De Souza, M., Hess, T., Hoekstra, A.Y., Ibidhi, R., Lathuillière, M.J., Manzardo, A., McAllister, T., Morales, R.A., Motoshita, M., Palhares, J.C.P., Pirlo, G., Ridoutt, B., Russo, V., Salmoral, G., Singh, R., Vanham, D., Wiedemann, S., Zheng, W., Pfister, S. (2021) Building consensus on water use assessment of livestock production systems and supply chains: Outcome and recommendations from the FAO LEAP Partnership. Ecological Indicators 124, 107391. Burek, P., Satoh, Y., Kahil, T., Tang, T., Greve, P., Smilovic, M., Guillaumot, L., Zhao, F., Wada, Y. (2020) Development of the Community Water Model (CWatM v1.04) – a high-resolution hydrological model for global and regional assessment of integrated water resources management. Geosci. Model Dev. 13, 3267-3298. Chapagain, A.K., Hoekstra, A.Y. (2003) Virtual water flows between nations in relation to trade in livestock and livestock products. Water Res. Rep. Ser. 13. de Fraiture, C. (2007) Integrated water and food analysis at the global and basin level. An application of WATERSIM. Water Resources Management 21, 185-198. de Fraiture, C., Wichelns, D., Rockström, J., Kemp-Benedict, E., Eriyagama, N., Gordon, L.J., Hanjra, M.A., Hoogeveen, J., Huber-Lee, A., Karlberg, L., (2007) Looking ahead to 2050: scenarios of alternative investment approaches. Demeke, B.W., Mekonnen, M.M., Brauman, K.A., Magliocca, N. (2026a) Uncertainty of water footprint estimates caused by harvested area inputs. Environmental Research: Water 2, 015006. Demeke, B.W., Mekonnen, M.M., Brauman, K.A., Magliocca, N., Moradkhani, H. (2026b) Global spatially detailed water footprint of crop production over five decades (accepted). Scientific Reports. FAO (2019) Water use in livestock production systems and supply chains -Guidelines for assessment (Version 1). Livestock Environmental Assessment and Performance (LEAP) Partnership. FAO, (2022) GLEAM 3 Dashboard. In: Shiny Apps. FAO, Rome, Italy. FAO, (2025) FAOSTAT online database. FAO, Rome. Gerbens-Leenes, P.W., Mekonnen, M.M., Hoekstra, A.Y. (2013) The water footprint of poultry, pork and beef: A comparative study in different countries and production systems. Water Resources and Industry 1–2, 25-36. Govoni, C., Chiarelli, D.D., Rulli, M.C. (2024) A global dataset of the national green and blue water footprint of livestock feeds. Scientific Data 11, 1419. Heinke, J., Lannerstad, M., Gerten, D., Havlík, P., Herrero, M., Notenbaert, A.M.O., Hoff, H., Müller, C. (2020) Water Use in Global Livestock Production—Opportunities and Constraints for Increasing Water Productivity. Water Resources Research 56, e2019WR026995. Herrero, M., Havlík, P., Valin, H., Notenbaert, A., Rufino, M.C., Thornton, P.K., Blümmel, M., Weiss, F., Grace, D., Obersteiner, M. (2013) Biomass use, production, feed efficiencies, and greenhouse gas emissions from global livestock systems. Proceedings of the National Academy of Sciences 110, 20888-20893. Herrero, M., Thornton, P.K. (2013) Livestock and global change: Emerging issues for sustainable food systems. Proceedings of the National Academy of Sciences 110, 20878-20881. Herrero, M., Thornton, P.K., Gerber, P., Reid, R.S. (2009) Livestock, livelihoods and the environment: understanding the trade-offs. Current Opinion in Environmental Sustainability 1, 111-120. Herrero, M., Thornton, P.K., Mason-D’Croz, D., Palmer, J., Benton, T.G., Bodirsky, B.L., Bogard, J.R., Hall, A., Lee, B., Nyborg, K., Pradhan, P., Bonnett, G.D., Bryan, B.A., Campbell, B.M., Christensen, S., Clark, M., Cook, M.T., de Boer, I.J.M., Downs, C., Dizyee, K., Folberth, C., Godde, C.M., Gerber, J.S., Grundy, M., Havlik, P., Jarvis, A., King, R., Loboguerrero, A.M., Lopes, M.A., McIntyre, C.L., Naylor, R., Navarro, J., Obersteiner, M., Parodi, A., Peoples, M.B., Pikaar, I., Popp, A., Rockström, J., Robertson, M.J., Smith, P., Stehfest, E., Swain, S.M., Valin, H., van Wijk, M., van Zanten, H.H.E., Vermeulen, S., Vervoort, J., West, P.C. (2020) Innovation can accelerate the transition towards a sustainable food system. Nature Food 1, 266-272. Hoekstra, A.Y. (2014) Water for animal products: a blind spot in water policy. Environmental Research Letters 9, 091003. Hoekstra, A.Y., Chapagain, A.K., Aldaya, M.M., Mekonnen, M.M. (2011) The water footprint assessment manual: Setting the global standard. Earthscan, London, UK. Mekonnen, M.M., Hoekstra, A.Y. (2012) A Global Assessment of the Water Footprint of Farm Animal Products. Ecosystems 15, 401-415. Mekonnen, M.M., Neale, C.M.U., Ray, C., Erickson, G.E., Hoekstra, A.Y. (2019) Water productivity in meat and milk production in the US from 1960 to 2016. Environment International 132, 105084. Molden, D., (2007) Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture. Earthscan/International Water Management Institute, London, UK / Colombo, Sri Lanka. Poore, J., Nemecek, T. (2018) Reducing food’s environmental impacts through producers and consumers. Science 360, 987. Steinfeld, H., Gerber, P. (2010) Livestock production and the global environment: Consume less or produce better? Proceedings of the National Academy of Sciences 107, 18237-18238. Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., Haan, C. (2006) Livestock’s long shadow: environmental issues and options. Food and Agriculture Organization, Rome. Weindl, I., Bodirsky, B.L., Rolinski, S., Biewald, A., Lotze-Campen, H., Müller, C., Dietrich, J.P., Humpenöder, F., Stevanović, M., Schaphoff, S., Popp, A. (2017) Livestock production and the water challenge of future food supply: Implications of agricultural management and dietary choices. Global Environmental Change 47, 121-132. Wisser, D., Grogan, D.S., Lanzoni, L., Tempio, G., Cinardi, G., Prusevich, A., Glidden, S. (2024) Water Use in Livestock Agri-Food Systems and Its Contribution to Local Water Scarcity: A Spatially Distributed Global Analysis. Water 16, 1681. Wolkeba, F.T., Mekonnen, M.M., Brauman, K.A., Kumar, M. (2024) Indicator metrics and temporal aggregations introduce ambiguities in water scarcity estimates. Scientific Reports 14, 15182. Zimmer, D.B., Renault, D., (2003) Virtual water in food production and global trade : Review of methodological issues and preliminary results, in: Hoekstra, A.Y. (Ed.), Virtual water trade: Proceedings of the International Expert Meeting on Virtual Water Trade. UNESCO-IHE, Delft, The Netherlands. Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryInformationv2.docx Supplementary Information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9360861","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":620515713,"identity":"955550f3-e16b-486d-b4c0-66f67e01fac1","order_by":0,"name":"Mesfin Mekonnen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7ElEQVRIiWNgGAWjYJCCDwkgkgfCkWNA4uACjDOQtRgTpwVZVWIDIS267e0PGx7uYIjm5zl87HFBxZ30DccPMD5424Zbi9mZM4YNiWcYcmf2tqUbzzjzLHfDmQRmw7n4tNzIYX+Q2MaQu+E8j5k0b9vh3A0HEtiADDxa7j9/2ADSsh+s5d/hdIPzD9h/49Vyg8EQrGUDbw9QS8PhBIMbCWzMeLWcyQFpkcidceZYmjTPscOGM288bJaccw6PluPHHzb+bLPJ7e9JPibNU3NYnu988sEPb8pwa4ECCWQOYwNB9aNgFIyCUTAK8AMAxrhXSGCeaa8AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-3573-9759","institution":"The University of Alabama","correspondingAuthor":true,"prefix":"","firstName":"Mesfin","middleName":"","lastName":"Mekonnen","suffix":""},{"id":620515714,"identity":"458cb04c-ec9e-4121-8be7-2be72f703798","order_by":1,"name":"Betelhem Demeke","email":"","orcid":"","institution":"The University of Alabama","correspondingAuthor":false,"prefix":"","firstName":"Betelhem","middleName":"","lastName":"Demeke","suffix":""},{"id":620515715,"identity":"0c2d64e6-fd6b-45f9-9cc8-632fbedd3080","order_by":2,"name":"Fitsume Wolkeba","email":"","orcid":"","institution":"University of Texas at Austin","correspondingAuthor":false,"prefix":"","firstName":"Fitsume","middleName":"","lastName":"Wolkeba","suffix":""}],"badges":[],"createdAt":"2026-04-08 20:10:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9360861/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9360861/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109147841,"identity":"b3aba915-c5b8-42fc-a9dd-543a6d01b30f","added_by":"auto","created_at":"2026-05-13 04:47:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":314975,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTotal water footprint of global livestock production (1972–2023). a)\u003c/strong\u003e Time‑series trends in the total WF (sum of green and blue WFs) of major livestock categories (beef cattle, dairy cattle, pig, broiler, layer, sheep, goat) showing absolute values and percent contribution to total WF change over the 51‑year period. \u003cstrong\u003eb)\u003c/strong\u003e Sankey diagram for 2023 illustrating the distribution of total WF (in km\u003csup\u003e3\u003c/sup\u003e) across major feed sources, direct water use, livestock categories, and the partitioning between green and blue WF, with blue water further subdivided into sustainable surface, unsustainable surface, sustainable groundwater, and groundwater depletion/overdraft components. This figure visualizes the structural drivers behind livestock water demand and identifies dominant feed and species contributions.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/18dd1b8a1b61dd484e6fed6d.png"},{"id":109147843,"identity":"a081f36e-a97e-455b-b691-c978757b2f37","added_by":"auto","created_at":"2026-05-13 04:47:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":76446,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGreen and blue water footprints of livestock production by country in 1972 and 2023\u003c/strong\u003e. Stacked bars show national livestock water footprints (km³ yr⁻¹) separated into green and blue components for 1972 (hatched) and 2023 (solid). Percent values above each bar indicate the country’s share of the global livestock water footprint in the corresponding year. Growth in each country’s contribution to the global total livestock water footprint between 1972 and 2023 is shown by a side‑offset arrow and label adjacent to the 2023 bar (negative values denote declines). Countries shown individually are the largest contributors in 2023; all remaining countries are aggregated as “Other countries.” The figure highlights major shifts in the geographical distribution of global livestock water use, including rapid increases in China, India, Brazil, and Pakistan, and a decline in the United States.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/0624b56d51ac23775c5552e4.png"},{"id":109205345,"identity":"92e519a9-305b-4b08-989d-65cd152bb3ec","added_by":"auto","created_at":"2026-05-13 15:04:20","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":82276,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eWater footprints of livestock feed sources and direct water use in 1972 and 2023.\u003cbr\u003e\n \u003c/strong\u003eStacked bars show the green and blue water footprints (km³ yr⁻¹) attributed to major feed sources and direct on‑farm water use for 1972 (hatched) and 2023 (solid). Percent values above each bar indicate the proportional contribution of each component to the total livestock water footprint in the corresponding year, while labels adjacent to the 2023 bars show the percentage change since 1972. The figure highlights key structural drivers of long‑term change, including the almost doubling of pasture‑related water footprints, large increases associated with maize and wheat, and the pronounced growth in soybean‑related water footprints linked to increasingly concentrate‑intensive livestock systems.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/74d883dcbf658164a6e0b4f3.png"},{"id":109205246,"identity":"6771c939-3ece-4843-8134-37219422808a","added_by":"auto","created_at":"2026-05-13 15:03:53","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":57538,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eUnsustainable blue water footprint of livestock production (1972 vs. 2023), by country and water source category. \u003c/strong\u003eStacked bars show the blue WF partitioned into sustainable surface, sustainable groundwater, unsustainable surface, and groundwater depletion/overdraft for 1972 and 2023 for selected countries. Mid‑bar labels indicate the share of each water‑source category in that country’s total national blue WF for the respective year. Top labels show each country’s contribution to the global total unsustainable blue WF in 1972 and 2023. The figure highlights both the shift in national water‑source dependence and the changing distribution of unsustainable blue water use at the global scale.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/98221fb1834e9022890fb0db.png"},{"id":109147845,"identity":"a2c5c621-fd2a-4f35-b7b4-9ef0ed954c16","added_by":"auto","created_at":"2026-05-13 04:47:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":59946,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eUnit water footprint of livestock products (L per kg of product) in 1972 and 2023. \u003c/strong\u003eDecrease in each livestock product’s unit water footprint between 1972 and 2023 is shown by a side‑offset arrow and label adjacent to the 2023 bar (negative values denote declines). The figure demonstrates substantial reductions in unit WF across all products, especially egg, pork, and chicken meat, reflecting productivity improvements, feed efficiency gains, and system-level changes in livestock production.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/ce52bfad1cfb07f51f4a6ef4.png"},{"id":109147847,"identity":"854f09a9-77e0-4544-999e-0a01f3a18334","added_by":"auto","created_at":"2026-05-13 04:47:38","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":97562,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal decomposition of unit water footprint (WF) change from 1972 to 2023, showing contributions from feed conversion ratio (FCR), weighted feed water footprint (FeedWF), concentrate share (Conc), and the residual term. Green WF contributions are shown with solid bars and Blue WF with hatched bars. Exact numerical log‑point contributions corresponding to each bar appear in Table S5.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/054c4f716d5594155dae612d.png"},{"id":109206032,"identity":"59974dfd-fcc8-428d-9b22-d74247e6a144","added_by":"auto","created_at":"2026-05-13 15:10:38","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":75697,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal consumptive water footprint (green + blue, m³ t\u003csup\u003e-1\u003c/sup\u003e) of major livestock products, comparing Mekonnen and Hoekstra (2012) with values from the Current Study. Each point represents a livestock product (beef, chicken meat, eggs, milk, pork). The dashed line denotes the 1:1 relationship (R² shown on the plot). Results illustrate the systematic differences between the original global estimates of Mekonnen \u0026amp; Hoekstra and the updated values produced by the Current Study.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/82246025ba6a388fc508d905.png"},{"id":109207266,"identity":"75fa3ee6-de65-4653-94f6-e5cc580cebfd","added_by":"auto","created_at":"2026-05-13 15:19:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":954941,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/62b6e733-ada7-40c6-9660-d891fe5de94c.pdf"},{"id":109147842,"identity":"b6352979-e7e6-4ce5-9684-30a25537d698","added_by":"auto","created_at":"2026-05-13 04:47:37","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2047376,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformationv2.docx","url":"https://assets-eu.researchsquare.com/files/rs-9360861/v1/facf4885733951ef695c7355.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Disentangling feed conversion efficiency and feed sourcing in global livestock water footprint trends, 1972–2023","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLivestock systems are central to global food security and rural livelihoods, supplying approximately 18% of global calories and 42% of protein while supporting the livelihoods of more than one billion people worldwide \u0026nbsp;(Alexandratos and Bruinsma, 2012; FAO, 2025). At the same time, livestock production is among the most resource‑intensive components of food systems, placing substantial demands on land and freshwater resources, much of which are embodied in animal feed (Alexandratos and Bruinsma, 2012; Heinke et al., 2020; Mekonnen and Hoekstra, 2012; Poore and Nemecek, 2018; Steinfeld et al., 2006). Livestock occupies roughly 30% of the Earth’s ice‑free land surface and relies on grasslands and croplands to supply animal biomass intake, with pronounced regional and system‑level variation in feed efficiency and production practices (Herrero et al., 2013; Steinfeld et al., 2006). These biophysical differences translate into large disparities in environmental burdens and improvement potential across species and production systems.\u003c/p\u003e\n\u003cp\u003eWater is a particularly salient constraint. Agriculture accounts for roughly 70% of global blue water withdrawals, and livestock‑related water use is dominated by the water required to produce feed, with drinking and service water playing comparatively minor roles (Mekonnen and Hoekstra, 2012). Global assessments estimate that livestock feed production consumes approximately 2,100–4,400 km³ yr⁻¹ of green and blue water, corresponding to up to about 40% of agricultural consumptive water use (de Fraiture, 2007; Heinke et al., 2020; Mekonnen and Hoekstra, 2012; Weindl et al., 2017). These estimates underscore the tight coupling between livestock water use, feed sourcing, and productivity. At the same time, methodological differences in accounting for green, blue, and grey water, and in reporting volumes versus impact‑based metrics, complicate comparisons across studies and can yield divergent results for similar products (Boulay et al., 2021; FAO, 2019).\u003c/p\u003e\n\u003cp\u003eA substantial body of literature has quantified the water footprints of animal products and identified feed conversion efficiency, feed composition, and feed origin as dominant drivers of unit water footprints across species and production systems (Gerbens-Leenes et al., 2013; Heinke et al., 2020; Mekonnen and Hoekstra, 2012; Mekonnen et al., 2019; Poore and Nemecek, 2018; Wisser et al., 2024). Seminal work showed that animal products generally exhibit higher unit water footprints than nutritionally comparable plant‑based foods (Mekonnen and Hoekstra, 2012). Subsequent analyses documented strong gradients from beef to pork to poultry. They demonstrated how transitions from grazing to mixed and industrial systems simultaneously improve feed efficiency while often increasing reliance on irrigated concentrate feeds, thereby shifting the balance between green and blue water use (Gerbens-Leenes et al., 2013; Heinke et al., 2020).\u003c/p\u003e\n\u003cp\u003eDespite these advances, the feed side of livestock water use, the amount of feed required, its composition, and source, remains incompletely resolved in a long‑term, country‑specific context. Global datasets have quantified biomass use and feed efficiencies across regions and systems, highlighting the central role of grasslands and mixed crop–livestock systems (Herrero and Thornton, 2013). More recent efforts provide national‑scale water footprint estimates for feed items in the late 2010s and increasingly link feed production, hydrology, and trade (Govoni et al., 2024; Wisser et al., 2024). However, these datasets do not resolve country‑ and animal‑specific feed composition as a consistent time series spanning multiple decades, limiting the ability to reconstruct the historical co‑evolution of feed baskets, feed sourcing (rainfed versus irrigated), and livestock productivity.\u003c/p\u003e\n\u003cp\u003eEvidence from regional case studies illustrates the importance of this historical perspective. In the United States, for example, the water footprint of meat and milk production declined markedly between 1960 and 2016 due to improvements in animal productivity, feed conversion efficiency, and crop yields that reduced the water footprint of feed ingredients (Mekonnen et al., 2019). Comparable, globally consistent reconstructions remain scarce, constraining efforts to disentangle structural changes in livestock systems from yield‑driven reductions in feed water requirements and to benchmark trajectories across countries.\u003c/p\u003e\n\u003cp\u003eAgainst this backdrop, three key gaps persist: (i) limited temporal coverage of country‑level, animal‑specific feed use and composition; (ii) the absence of harmonized, long‑run green and blue water footprint estimates for livestock production; and (iii) insufficient attribution of changes in unit water footprints to evolving feed baskets, feed conversion efficiencies, and crop water requirements.\u003c/p\u003e\n\u003cp\u003eThe current study addresses these gaps by assembling a global, country‑level, animal‑specific time series of feed use and composition spanning 1961–2023, harmonized with FAOSTAT production statistics and informed by the livestock systems literature. We use this framework to (i) trace the evolution of livestock feed baskets across species and countries; (ii) estimate green and blue consumptive water use attributable to livestock production consistently over 1972–2023; and (iii) derive unit water footprints that decompose temporal changes into contributions from feed composition, feed conversion efficiency, and crop water requirements. By providing a unified six‑decade perspective, this analysis complements existing global snapshots and recent feed water footprint datasets, offering new insights into how nutrition, technology, and agronomy have jointly shaped the water intensity of animal‑source foods.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eGlobal water footprint of livestock production (1972\u0026ndash;2023)\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe total consumptive water footprint (WF) of global livestock production increased by 54% over the past five decades, increasing from 1,979 km\u0026sup3; yr⁻\u0026sup1; in 1972 to 3,044 km\u0026sup3; yr⁻\u0026sup1; in 2023 (Fig. 1). This long‑term growth trend and the relative increases across livestock categories are depicted in Fig. 1a, while the 2023 structural composition of WF by feed sources and water types is summarized in Fig. 1b. Detailed WFs by feed sources, livestock type, and water sources are provided in Table-S1 (Supplementary Information).\u003c/p\u003e\n\u003cp\u003eBreaking down by species, ruminants (beef cattle, dairy cattle) remain the dominant contributors to global livestock WF, while broilers show the fastest relative growth. The evolution of species‑level totals and percentage changes are visualized in Fig. 1a; exact species totals and shares for the period 1972 to 2023 are reported in Table S1.2.\u003c/p\u003e\n\u003cp\u003eAt the national level, growth patterns were highly heterogeneous (Fig. 2). China exhibited the strongest absolute increase, rising from 147 km\u0026sup3; yr⁻\u0026sup1; to 586 km\u0026sup3; yr⁻\u0026sup1; (+299%), and overtaking the United States to become the world\u0026rsquo;s largest contributor (19% of global WF in 2023). India (+259%) and Brazil (+136%) also showed large increases consistent with strong expansion in dairy and meat demand. In contrast, the United States reduced its livestock WF by 13%, reflecting efficiency gains, shifts toward more productive production systems, and improvements in feed conversion (Mekonnen et al., 2019). Pakistan showed the largest relative growth (+605%), though from a low baseline. Together, China, India, Brazil, the United States, and Pakistan accounted for 44% of global livestock WF in 2023 (Fig. 2). Detailed WFs per country, livestock type, and water sources are provided in Table-S2 (Supplementary Information).\u003c/p\u003e\n\u003cp\u003eThe feed‑source decomposition reveals the structural drivers behind these changes (Fig. 3). Pasture remains the single largest component (48% of total WF in 2023), growing by 83% since 1972, while maize now contributes the largest WF among feed crops (495 km\u0026sup3; yr⁻\u0026sup1; in 2023). The most dramatic relative increase occurred for soybean (\u0026asymp;20‑fold since 1972), reflecting global shifts toward concentrate‑rich rations. The 2023 partitioning of green versus blue water, and the sustainability categories within blue water, is illustrated in Fig. 1b, which links water sources to feed streams and livestock outputs.\u003c/p\u003e\n\u003cp\u003eThe results confirm that (i) livestock scaling, not just yield or feed efficiency, remains the dominant driver of global WF growth, (ii) rapid shifts toward concentrate‑heavy diets amplify pressures on cropland water resources, and (iii) large increases in maize and soybean WFs mirror the global rise in mixed and industrial production systems.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eUnsustainable blue water footprint of livestock production\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe blue component of the livestock water footprint (WF) shows a marked shift toward unsustainable sources over the past five decades. Aggregated across countries, total blue WF rose from 158 km\u0026sup3; yr⁻\u0026sup1; in 1972 (45% surface\u0026ndash;sustainable, 23% groundwater\u0026ndash;sustainable, 8% surface\u0026ndash;unsustainable, 23% groundwater depletion/overdraft) to 252 km\u0026sup3; yr⁻\u0026sup1; in 2023 (41% surface\u0026ndash;sustainable, 6% groundwater\u0026ndash;sustainable, 12% surface\u0026ndash;unsustainable, 41% groundwater depletion/overdraft), indicating that unsustainable blue WF increased by 2.7 times (from 50 to 133 km\u0026sup3; yr⁻\u0026sup1;, +170%).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCountry contributions to the global unsustainable blue total evolved unevenly (Fig. 4). China\u0026rsquo;s absolute unsustainable blue volume more than doubled (from 15 to 33 km\u0026sup3; yr⁻\u0026sup1;), yet its share of the global unsustainable total fell from 30% (1972) to 24% (2023) because other countries, especially those with rapidly expanding groundwater dependence, grew faster. India moved in the opposite direction: its unsustainable blue volume rose from 2 to 25 km\u0026sup3; yr⁻\u0026sup1;, and its contribution to the global unsustainable total almost quadrupled from 5% to 19%, reflecting strong growth in groundwater depletion/overdraft for livestock feed and production. The US maintained almost similar absolute magnitudes but reduced its global contribution from 24% to 10%, indicating that, relative to worldwide trends, US livestock systems now account for a smaller share of global unsustainable blue water use. Among other notable contributors, Spain increased unsustainable blue volumes (\u0026asymp;1\u0026rarr;3 km\u0026sup3; yr⁻\u0026sup1;) while its share of the global unsustainable total reduced from 3% to 2%; Egypt rose from ~2 to 5 km\u0026sup3; yr⁻\u0026sup1;, with its global contribution decreasing from5% to 4%. Table S1 and Table S2 provide the sustainable and unsustainable surface and groundwater per feed and country, respectively.\u003c/p\u003e\n\u003cp\u003eTaken together, the country‑level trajectories in Fig. 4 show rising groundwater depletion/overdraft dependence alongside differing national contributions to the global unsustainable blue total. Even where some large producers reduced their share of the global unsustainable blue total (e.g., China, United States), absolute unsustainable blue volumes generally increased; meanwhile, countries such as India account for an expanding portion of the global unsustainable blue WF, with growth driven especially by groundwater depletion/overdraft. This underscores that efficiency gains must be paired with basin‑level limits and source‑shifting in regions where livestock and feed production depend on unsustainable blue water.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eUnit water footprint of livestock products\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnit water footprints declined substantially across most livestock products between 1972 and 2023 (Fig. 5). Notable reductions include pork (\u0026minus;70% in L kg⁻\u0026sup1;), eggs (\u0026minus;68%), and chicken meat (\u0026minus;66%), consistent with improvements in feed conversion efficiency, genetics, and management in intensive systems. Milk declined by ~52% across units, while cattle, sheep, and goat meat showed more modest decreases. The unit WFs of the major livestock products are provided in Table S3.\u003c/p\u003e\n\u003cp\u003eUnit WF reductions are widespread and consistent with known global improvements in livestock productivity observed in the literature (e.g., Mekonnen and Hoekstra (2012) for US productivity trends). The fact that total WFs increased despite lower unit WFs underscores the overwhelming role of production scale compared to efficiency gains.\u003c/p\u003e\n\u003cp\u003eAcross Figs. 1, 2, 3, and 5, three consistent patterns emerge. First, scale effects dominate changes in total water footprints, as global livestock WF increased despite widespread declines in unit WF, indicating that growth in production outpaced efficiency gains. Second, changes in feed composition strongly shape water outcomes: increasing reliance on maize and soybean feeds, combined with the continued dominance of pasture, explains much of the observed spatial and temporal variation in livestock WFs (Figs. 1b and 3). Third, efficiency gains differ markedly across species, with monogastric systems achieving the largest reductions in unit WFs (Fig. 5), while ruminant production continues to account for the majority of global livestock WF due to its scale and dependence on pasture-based feeds (Fig. 1a).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eDrivers of Unit Water Footprints of Livestock Products\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAcross all livestock types, the model explains the vast majority of observed variation in unit WFs (R\u0026sup2; \u0026asymp; 0.96 for green, 0.99 for blue on average across products). The estimated elasticities are strikingly close to the multiplicative identity one would expect from first principles (i.e., product WF \u0026asymp; FCR \u0026times; feed WF), with an additional positive effect of concentrate share:\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eFeed conversion (FCR): A 1% increase in FCR is associated with a \u0026asymp; 1.02% increase in green unit WF and a \u0026asymp; 0.98% increase in blue unit WF on average across livestock types (mean \u003cimg width=\"17\" height=\"21\" src=\"data:image/png;base64,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\" v:shapes=\"_x0000_i1025\" alt=\"image\"\u003e=1.02, \u003cimg width=\"17\" height=\"19\" src=\"data:image/png;base64,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\" v:shapes=\"_x0000_i1025\" alt=\"image\"\u003e=0.98). These near‑unity elasticities confirm that, all else equal, feed use per unit output is the dominant lever on product WF, consistent with the multiplicative structure of livestock production systems. (Table S4 \u0026ndash; summary of factor analysis)\u003c/li\u003e\n \u003cli\u003eWeighted feed WF: A 1% increase in the weighted feed WF (m\u0026sup3; t⁻\u0026sup1; feed) is associated with a \u0026asymp; 0.34% increase in green unit WF and a \u0026asymp; 0.76% increase in blue unit WF (mean \u003cimg width=\"17\" height=\"21\" src=\"data:image/png;base64,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\" v:shapes=\"_x0000_i1025\" alt=\"image\"\u003e=0.34, \u003cimg width=\"17\" height=\"19\" src=\"data:image/png;base64,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\" v:shapes=\"_x0000_i1025\" alt=\"image\"\u003e=0.76). This asymmetric response indicates that changes in feed sourcing and crop water intensity transmit much more strongly to blue water footprints, reflecting the dominant role of irrigated feed crops in shaping blue WF outcomes. (Table S4 \u0026ndash; summary of factor analysis)\u003c/li\u003e\n \u003cli\u003eConcentrate fraction: Holding FCR and feed WF constant, increasing the concentrate share in the ration raises product WFs, with semi‑elasticities \u003cimg width=\"17\" height=\"21\" src=\"data:image/png;base64,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\" v:shapes=\"_x0000_i1025\" alt=\"image\"\u003e\u0026asymp;1.04 for green WF and \u003cimg width=\"17\" height=\"20\" src=\"data:image/png;base64,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\" v:shapes=\"_x0000_i1025\" alt=\"image\"\u003e\u0026asymp;1.12 for blue WF. Interpreted at the mean, a +0.10 increase in concentrate share (i.e., +10 percentage points) is associated with roughly 10-11% higher unit WFs, consistent with concentrate‑rich rations sourcing disproportionately from water-intensive and irrigated feed crops. The magnitude of this effect varies by livestock type but is consistently positive across systems (Table S4 \u0026ndash; summary of factor analysis)\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ePer‑product regressions display the same pattern: both FCR and feed WF carry elasticities ~1, and the concentrate fraction is positively associated with unit WFs for every product (high R\u0026sup2; in all cases). Full coefficients appear in Supplementary Table S4.\u003c/p\u003e\n\u003cp\u003eThese estimated elasticities provide the structural basis for attributing long‑run changes in unit WFs. In the sections below, we combine the estimated factor sensitivities with observed historical changes in feed conversion, feed water intensity, and concentrate share to decompose global unit WF trends between 1972 and 2023.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eInterpretation and linkage to prior work\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThese results are fully consistent with the biophysical logic and previous global assessments:\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003e\u003cem\u003eScaling with feed demand\u003c/em\u003e: Because animals must consume feed to produce output, FCR reductions (through genetics, health, management) translate almost one‑for‑one into lower unit WFs. This is the same mechanism underlying long‑run improvements observed in the United States (1960\u0026ndash;2016) (Mekonnen et al., 2019) and elsewhere, where better feed conversion contributed materially to lower unit WFs despite rising total production. (Table S4 \u0026ndash; summary of factor analysis)\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eFeed sourcing and composition\u003c/em\u003e: Weighted feed WF captures both crop choice and geography. Concentrate feeds (e.g., maize, soy), particularly when irrigated, increase blue WFs even when FCR improves, consistent with prior findings on the water cost of diet intensification.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eSystem‑level implications\u003c/em\u003e: The positive concentrate coefficient does not imply that intensification is inherently water‑inefficient; rather, it highlights that efficiency gains must be paired with water‑aware feed sourcing to avoid shifting pressure from green to blue water resources\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAttribution of Long‑Run Changes (1972\u0026ndash;2023)\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe decomposed global changes in unit water footprints (WFs) between 1972 and 2023 using the log‑additive decomposition shown in Eq. S8. Fig. 6 displays the global contributions by component, and the exact values are provided in Table S5. Fig. S1 displays contributions by components for major countries, and Table S6 provides exact values for these countries.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMonogastrics (Pig, Chicken, Egg)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSubstantial reductions in both green and blue unit WFs are observed:\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eChicken: \u0026minus;67% (green), \u0026minus;54% (blue), driven by FCR improvements (~\u0026minus;35 pp) and declining feed WFs (~\u0026minus;24 pp green; ~\u0026minus;7.3 pp blue). Concentrate changes contribute an additional \u0026minus;2% to \u0026minus;5%.\u003c/li\u003e\n \u003cli\u003eEgg: \u0026minus;68% (green), \u0026minus;66% (blue), driven by FCR improvements (~\u0026minus;30 pp green; ~\u0026minus;31 pp blue) and declining feed WFs (~\u0026minus;23 pp green; ~\u0026minus;24 pp blue). Concentrate changes contribute an additional \u0026minus;2%.\u003c/li\u003e\n \u003cli\u003ePig: \u0026minus;70% (green) and \u0026minus;71% (blue), with large contributions from FCR (~\u0026minus;35 pp green; ~\u0026minus;28 pp blue) and feed WF (~\u0026minus;36 pp green; ~\u0026minus;48 pp blue).\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cem\u003eRuminants (Cattle, Dairy, Sheep, Goat)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eRuminant trends are mixed:\u003c/em\u003e\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eCattle: Green WFs decline (~\u0026minus;31%), and blue WFs decline (~\u0026minus;18%), driven by FCR improvement (~-12 pp) and rising feed WF (~+1.2 pp green; ~+10 pp blue).\u003c/li\u003e\n \u003cli\u003eDairy: Green WF declines (~\u0026minus;54%), but blue WF increases markedly (+35.8%), almost entirely due to rising feed blue WF (+51 pp), overwhelming FCR improvements (\u0026minus;30 pp).\u003c/li\u003e\n \u003cli\u003eSheep: Similar pattern: small green decline (~\u0026minus;19.7%) but increase in blue (~+11.8%) from positive feed blue contributions.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eAcross livestock systems, three robust conclusions emerge. First, improvements in feed conversion ratios (FCRs) continue to be the most powerful mechanism driving reductions in unit water footprints. Second, the sourcing of feed\u0026mdash;particularly the crop types used and the geographic regions in which they are produced\u0026mdash;plays the dominant role in shaping long‑term trends in blue water footprints. Third, changes in the share of concentrate feeds can either offset or amplify these improvements, depending on the relative water intensity associated with their production.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eComparison with previous studies\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA comparison of our updated consumptive water footprint (WF) estimates with major prior global assessments reveals both broad agreement at the global scale and substantial differences at the country and feed‑component levels. Earlier studies, particularly Mekonnen and Hoekstra (2012), provided the first comprehensive global WF benchmarks for livestock, combining country‑level feed composition, production systems, and feed conversion efficiencies. \u0026nbsp;More recent work (Heinke et al., 2020) applied updated feed and crop‑water models to reassess livestock water use, estimating that approximately 3,374 km\u0026sup3; yr⁻\u0026sup1; of green + blue water is required for global feed production.\u003c/p\u003e\n\u003cp\u003eOur global estimates of total consumptive WF for livestock feed fall within the same order of magnitude as earlier work (Table 1). For the year 2000, the total global feed WF from this study (2,368 km\u0026sup3; yr⁻\u0026sup1;) closely matches Mekonnen and Hoekstra (2012) estimate for 1996\u0026ndash;2005 (2,217 km\u0026sup3; yr⁻\u0026sup1;), Zimmer and Renault (2003) estimate for 2000 (2,340 km\u0026sup3; yr⁻\u0026sup1;), and Molden (2007) estimate for 2000 (2,152 km\u0026sup3; yr⁻\u0026sup1;), differing by only 6%, 1%, \u0026nbsp;and 11%, respectively. This shows strong consistency in global‐aggregate WF magnitudes across methods and datasets.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1\u003c/strong\u003e. \u003cstrong\u003eComparison of global consumptive water footprint (WF) estimates (green + blue, km\u0026sup3; yr⁻\u0026sup1;) for livestock feed across major global studies.\u0026nbsp;\u003c/strong\u003eValues are shown for feed crops, pasture, and total consumptive WF for comparable reference years. Percent differences reflect the deviation of each study from the Current Study\u0026rsquo;s estimate.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"577\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e\u0026nbsp;Study\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003ePeriod\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eFeed crop\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003ePasture\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eTotal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e% difference\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eCurrent Study\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1249\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1119\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2368\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eMekonnen and Hoekstra (2012)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1996-2005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1304\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e913\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2217\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e6%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eWeindl et al. (2017)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2170\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2590\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e4760\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e-101%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eMolden (2007)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e840\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2152\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e9%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eHeinke et al. (2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1998-2002\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1686\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1688\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e3374\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e-42%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003ede Fraiture et al. (2007)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e840\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2152\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2992\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e-26%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003eZimmer and Renault (2003)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e2340\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e1%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eDespite similar global totals, the partitioning between feed crops and pasture differs substantially across studies (Table 1). For 2000, our estimate of feed‑crop WF (1,249 km\u0026sup3; yr⁻\u0026sup1;) is -4% lower than Mekonnen and Hoekstra (2012), while pasture WF (1,119 km\u0026sup3; yr⁻\u0026sup1;) is 18% higher. These shifts reflect updated feed volumes, improved livestock productivity data, and revised pasture mapping.\u003c/p\u003e\n\u003cp\u003eOther studies diverge more sharply, reflecting differences in data sources and modeling assumptions. For example, Weindl et al. (2017) reported total livestock‑feed consumptive WF of 4,760 km\u0026sup3; yr⁻\u0026sup1;, more than double our estimate, due in part to broader inclusion of extensive systems and differences in crop‑water modeling. Heinke et al. (2020) reported an estimate of 3,374 km\u0026sup3; yr⁻\u0026sup1; for 2000 conditions, 58% higher than ours. This discrepancy likely stems from their use of the LPJmL4 global hydrological model, which simulates agriculture using broad crop functional types rather than explicit crop-specific representations, potentially reducing fidelity in modeling crop evapotranspiration relative to the crop model used here. Additionally, uncertainty in harvested area inputs\u0026mdash;known to be a dominant source of variability in large‑scale agricultural water use assessments\u0026mdash;further contributes to differences across studies (Demeke et al., 2026a).\u003c/p\u003e\n\u003cp\u003eComparison of \u0026nbsp;our updated unit consumptive WF (green + blue) with those reported by Mekonnen and Hoekstra (2012) for major livestock products are shown in Fig.7. The 1:1 line illustrates near‑agreement for most products at the global mean. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAt the country scale (Fig. S2), deviations between our results and Mekonnen and Hoekstra (2012) become substantial, reflecting updated feed‑mix data, revised harvested areas, and national differences in feed supply chains. The most important contributors to these discrepancies are:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eDifferences in annual feed volumes and diet composition, especially for monogastrics, where rapid intensification after 2000 altered concentrate use.\u003c/li\u003e\n \u003cli\u003eRevised feed‑crop WFs, stemming from updated harvested areas, crop‑water requirements, and spatially explicit crop modeling. These are known to introduce uncertainty (see Demeke et al. (2026a)).\u003c/li\u003e\n \u003cli\u003eUpdated concentrate accounting, where FAO FBS data, now used to replace model‑based grain estimates, shifts the relative roles of grains versus pasture or crop residues.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eIn sum, while our global‑aggregate WF estimates fall well within the envelope of prior studies, large differences emerge at the feed‑category and country levels. These differences reflect improvements in livestock productivity data, updated feed composition and FCR trajectories, and advances in crop‑water modeling. The overall picture aligns with recent analyses indicating that method selection, input data quality, and assumptions about feed sourcing heavily influence livestock WF outcomes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003ePolicy relevance\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnderstanding long‑term dynamics of feed use and livestock water footprints is not only scientifically important but also central to emerging policy frameworks aimed at improving water governance and food‑system sustainability. Because feed production accounts for the majority of livestock water withdrawals and is tightly linked to basin‑level blue water stress, robust national time‑series of green and blue water footprints provide critical evidence for designing water‑allocation caps, evaluating agricultural intensification strategies, and improving the long‑run resilience of livestock systems. Recent FAO LEAP guidelines emphasize aligning water‑productivity improvements with assessments of blue water scarcity, highlighting that mitigation strategies must consider both efficiency gains and their local hydrologic context (Boulay et al., 2021). Spatially detailed global analyses further show that reducing livestock water use alone will not eliminate water stress in many basins, underscoring the need for integrated land\u0026ndash;water planning, crop choice optimization, and feed‑trade strategies tailored to local hydrologic limits (Gerbens-Leenes et al., 2013; Herrero and Thornton, 2013; Herrero et al., 2009; Hoekstra, 2014; Mekonnen and Hoekstra, 2012; Wisser et al., 2024). Additionally, because livestock products constitute a large share of agricultural water use and demand is projected to increase substantially toward mid‑century, countries must reconcile dietary shifts, productivity goals, and environmental constraints in ways that protect freshwater ecosystems while sustaining food and nutrition security. By generating consistent, multi‑decadal, country‑level estimates of feed use and livestock water footprints, this study offers actionable information for national governments, basin authorities, and international agencies seeking to implement evidence‑based policies that enhance water sustainability, strengthen climate resilience, and support equitable transitions in the livestock sector.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eLimitations\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDespite the use of harmonized global datasets and a consistent methodological framework to enable long‑term and cross‑country comparisons, several limitations should be acknowledged when interpreting the results. Scaling a global average feed conversion ratio (FCR) for 2015 across the 1972\u0026ndash;2023 period using annual yield trends assumes a stable, linear relationship between yield and feed efficiency, which does not fully capture temporal and regional variability driven by changes in feed composition and quality, livestock genetics, animal health, management practices, and environmental conditions. As a result, historical improvements or declines in feed efficiency that are decoupled from yield trends may not be fully represented. In addition, livestock production systems (grazing, mixed, and industrial) are not explicitly distinguished, unlike in previous studies, and livestock production is instead represented in an aggregated, country‑level manner. While this approach ensures internal consistency across countries and time, it masks important intra‑country heterogeneity in feed sources, water use intensity, and production efficiency, potentially smoothing system‑specific extremes. The national‑scale aggregation limits the representation of sub‑national variability, particularly in large and climatically diverse countries, where spatial mismatches between feed production and livestock consumption can influence water‑use patterns. Finally, the study inherits the limitations of the global model used. Specifically, the limitation in groundwater surface water partitioning is affected by the specific groundwater representation in the model used. Together, these limitations highlight the need for future assessments that integrate system‑specific feed efficiencies and finer spatial resolution to better capture structural and technological changes in livestock production.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study provides a long‑term, globally consistent assessment of livestock feed requirements and associated resource implications over the 1972–2023 period, offering new insights into how changes in agricultural productivity and feed conversion efficiency shape the water footprint of animal production. By harmonizing multi‑decadal yield data with feed conversion metrics, the analysis reveals pronounced temporal and spatial contrasts in feed demand and efficiency gains across countries, highlighting the uneven pace at which technological and structural changes in livestock systems have unfolded.\u003c/p\u003e\n\u003cp\u003eThe results underscore the importance of accounting for historical dynamics when evaluating contemporary livestock sustainability, as present‑day efficiencies often reflect cumulative investments in crop and livestock productivity, management, and trade rather than uniform global progress. At the same time, the observed divergence among regions suggests that aggregate improvements can coexist with persistent inefficiencies and localized pressures on water resources. By quantifying these long‑term patterns within a unified framework, this work complements existing snapshot‑based assessments and strengthens the empirical basis for linking livestock production, food security, and water sustainability (Herrero et al., 2009; Herrero et al., 2020; Steinfeld and Gerber, 2010).\u003c/p\u003e\n\u003cp\u003eOverall, the study advances understanding of the systemic drivers underlying global livestock feed demand and its water implications, providing a transparent reference point for future efforts that seek to incorporate production‑system detail, sub‑national variability, and dynamic technological changes into integrated food–water sustainability assessments.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eWater footprint accounting framework:\u0026nbsp;\u003c/strong\u003eWe quantified the green and blue water footprints of livestock production using a harmonized accounting framework consistent with the Water Footprint Assessment methodology\u0026nbsp;\u0026nbsp;(Hoekstra et al., 2011). The analysis covers seven major livestock categories—beef cattle, dairy cattle, pigs, sheep, goats, broiler chickens, and layer chickens—and spans the period 1972–2023 for all countries. Total livestock water footprints were decomposed into feed‑related, drinking, and service water components, with feed production accounting for the dominant share of consumptive water use\u0026nbsp;(Chapagain and Hoekstra, 2003; Hoekstra et al., 2011; Mekonnen and Hoekstra, 2012).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFeed requirements and feed composition:\u0026nbsp;\u003c/strong\u003eLivestock feed requirements were reconstructed using feed conversion ratios anchored to FAO‑based estimates (FAO, 2022) and extended over time using observed changes in per‑animal productivity. Feed baskets were allocated across ingredient groups using livestock feed composition data and reconciled with FAO food balance statistics (FAO, 2025) to ensure mass balance and consistency over time. This approach captures long‑term changes in total feed demand while reflecting shifts between concentrate and non‑concentrate feed sources at the country level.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWater footprints of feed and livestock products:\u0026nbsp;\u003c/strong\u003eCountry‑specific green and blue water footprints of feed crops from Demeke et al. (2026b) were combined with reconstructed feed quantities to estimate feed‑related water use. For traded feed commodities, water footprints were calculated as weighted averages of domestic production and imports. Unit water footprints of livestock products were then derived by linking feed quantities, feed water intensities, and animal productivity in a consistent accounting framework.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePartitioning sustainable and unsustainable blue water footprints of feed crops:\u0026nbsp;\u003c/strong\u003eWe partitioned the blue water footprint (WF) of each feed crop into contributions from surface water and groundwater, using CWatM (Burek et al., 2020) outputs that attribute irrigation supply by source and identify groundwater depletion/overdraft (Wolkeba et al., 2024). We then classified the surface‑water component as unsustainable where blue water scarcity \u0026gt; 1, indicating that consumptive blue water use exceeds locally available water (natural runoff minus environmental flow requirements). These steps yield spatially and temporally consistent estimates of sustainable versus unsustainable blue WF for each feed crop, which are then aggregated to national feed baskets and livestock products.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAttribution of temporal changes:\u0026nbsp;\u003c/strong\u003eTo identify the main drivers of changes in unit water footprints over time, we applied country‑year panel regression models separately for each livestock product. The analysis attributed variation in unit water footprints to changes in feed conversion efficiency, the water intensity of feed baskets, and the share of concentrate feeds. This framework enables a consistent decomposition of long‑term trends across countries and livestock categories.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData sources and supplementary methods:\u0026nbsp;\u003c/strong\u003eAll data were obtained from harmonized global sources, including FAOSTAT (FAO, 2022), FAO livestock and feed datasets (FAO, 2022), and recent global feed water footprint products (Demeke et al., 2026b). Water scarcity maps were obtained from Wolkeba et al. (2024). All equations and detailed methodological descriptions are provided in the Supplementary Information and are numbered independently as Eqs. (S1–S8).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe output data is provided as a Supplementary Table. All input datasets used in the current study are publicly available from their respective original sources, as referenced in Methods and Supplementary Information.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere are no special codes used in the current work. Any Python scripts used for handling large data in the current work are available upon request from the corresponding author.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe acknowledge the financial support received from the Global Water Security Center, Alabama Water Institute, the University of Alabama. Additionally, we would like to recognize the contributions of FAO's GLEAM team, particularly Dr. Dominik Wisser and Dr. Giuseppe Tempio, for providing the GLEAM dashboard data, which includes feed intake per livestock production system for the year 2015.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eM.M.M. conceived the ideas, conducted the analysis, and led the writing of the paper; B.W.D. and F.T.W. provided the input data, reviewed, and edited the manuscript.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAlexandratos, N., Bruinsma, J., (2012) World agriculture towards 2030/2050: the 2012 revision, ESA Working paper No. 12-03. Food and Agriculture Organization, Rome.\u003c/li\u003e\n \u003cli\u003eBoulay, A.-M., Drastig, K., Amanullah, Chapagain, A., Charlon, V., Civit, B., DeCamillis, C., De Souza, M., Hess, T., Hoekstra, A.Y., Ibidhi, R., Lathuilli\u0026egrave;re, M.J., Manzardo, A., McAllister, T., Morales, R.A., Motoshita, M., Palhares, J.C.P., Pirlo, G., Ridoutt, B., Russo, V., Salmoral, G., Singh, R., Vanham, D., Wiedemann, S., Zheng, W., Pfister, S. (2021) Building consensus on water use assessment of livestock production systems and supply chains: Outcome and recommendations from the FAO LEAP Partnership. Ecological Indicators 124, 107391.\u003c/li\u003e\n \u003cli\u003eBurek, P., Satoh, Y., Kahil, T., Tang, T., Greve, P., Smilovic, M., Guillaumot, L., Zhao, F., Wada, Y. (2020) Development of the Community Water Model (CWatM v1.04) \u0026ndash; a high-resolution hydrological model for global and regional assessment of integrated water resources management. Geosci. Model Dev. 13, 3267-3298.\u003c/li\u003e\n \u003cli\u003eChapagain, A.K., Hoekstra, A.Y. (2003) Virtual water flows between nations in relation to trade in livestock and livestock products. Water Res. Rep. Ser. 13.\u003c/li\u003e\n \u003cli\u003ede Fraiture, C. (2007) Integrated water and food analysis at the global and basin level. An application of WATERSIM. Water Resources Management 21, 185-198.\u003c/li\u003e\n \u003cli\u003ede Fraiture, C., Wichelns, D., Rockstr\u0026ouml;m, J., Kemp-Benedict, E., Eriyagama, N., Gordon, L.J., Hanjra, M.A., Hoogeveen, J., Huber-Lee, A., Karlberg, L., (2007) Looking ahead to 2050: scenarios of alternative investment approaches.\u003c/li\u003e\n \u003cli\u003eDemeke, B.W., Mekonnen, M.M., Brauman, K.A., Magliocca, N. (2026a) Uncertainty of water footprint estimates caused by harvested area inputs. Environmental Research: Water 2, 015006.\u003c/li\u003e\n \u003cli\u003eDemeke, B.W., Mekonnen, M.M., Brauman, K.A., Magliocca, N., Moradkhani, H. (2026b) Global spatially detailed water footprint of crop production over five decades (accepted). Scientific Reports.\u003c/li\u003e\n \u003cli\u003eFAO (2019) Water use in livestock production systems and supply chains -Guidelines for assessment (Version 1). Livestock Environmental Assessment and Performance (LEAP) Partnership.\u003c/li\u003e\n \u003cli\u003eFAO, (2022) GLEAM 3 Dashboard. In: Shiny Apps. FAO, Rome, Italy.\u003c/li\u003e\n \u003cli\u003eFAO, (2025) FAOSTAT online database. FAO, Rome.\u003c/li\u003e\n \u003cli\u003eGerbens-Leenes, P.W., Mekonnen, M.M., Hoekstra, A.Y. (2013) The water footprint of poultry, pork and beef: A comparative study in different countries and production systems. Water Resources and Industry 1\u0026ndash;2, 25-36.\u003c/li\u003e\n \u003cli\u003eGovoni, C., Chiarelli, D.D., Rulli, M.C. (2024) A global dataset of the national green and blue water footprint of livestock feeds. Scientific Data 11, 1419.\u003c/li\u003e\n \u003cli\u003eHeinke, J., Lannerstad, M., Gerten, D., Havl\u0026iacute;k, P., Herrero, M., Notenbaert, A.M.O., Hoff, H., M\u0026uuml;ller, C. (2020) Water Use in Global Livestock Production\u0026mdash;Opportunities and Constraints for Increasing Water Productivity. Water Resources Research 56, e2019WR026995.\u003c/li\u003e\n \u003cli\u003eHerrero, M., Havl\u0026iacute;k, P., Valin, H., Notenbaert, A., Rufino, M.C., Thornton, P.K., Bl\u0026uuml;mmel, M., Weiss, F., Grace, D., Obersteiner, M. (2013) Biomass use, production, feed efficiencies, and greenhouse gas emissions from global livestock systems. Proceedings of the National Academy of Sciences 110, 20888-20893.\u003c/li\u003e\n \u003cli\u003eHerrero, M., Thornton, P.K. (2013) Livestock and global change: Emerging issues for sustainable food systems. Proceedings of the National Academy of Sciences 110, 20878-20881.\u003c/li\u003e\n \u003cli\u003eHerrero, M., Thornton, P.K., Gerber, P., Reid, R.S. (2009) Livestock, livelihoods and the environment: understanding the trade-offs. Current Opinion in Environmental Sustainability 1, 111-120.\u003c/li\u003e\n \u003cli\u003eHerrero, M., Thornton, P.K., Mason-D\u0026rsquo;Croz, D., Palmer, J., Benton, T.G., Bodirsky, B.L., Bogard, J.R., Hall, A., Lee, B., Nyborg, K., Pradhan, P., Bonnett, G.D., Bryan, B.A., Campbell, B.M., Christensen, S., Clark, M., Cook, M.T., de Boer, I.J.M., Downs, C., Dizyee, K., Folberth, C., Godde, C.M., Gerber, J.S., Grundy, M., Havlik, P., Jarvis, A., King, R., Loboguerrero, A.M., Lopes, M.A., McIntyre, C.L., Naylor, R., Navarro, J., Obersteiner, M., Parodi, A., Peoples, M.B., Pikaar, I., Popp, A., Rockstr\u0026ouml;m, J., Robertson, M.J., Smith, P., Stehfest, E., Swain, S.M., Valin, H., van Wijk, M., van Zanten, H.H.E., Vermeulen, S., Vervoort, J., West, P.C. (2020) Innovation can accelerate the transition towards a sustainable food system. Nature Food 1, 266-272.\u003c/li\u003e\n \u003cli\u003eHoekstra, A.Y. (2014) Water for animal products: a blind spot in water policy. Environmental Research Letters 9, 091003.\u003c/li\u003e\n \u003cli\u003eHoekstra, A.Y., Chapagain, A.K., Aldaya, M.M., Mekonnen, M.M. (2011) The water footprint assessment manual: Setting the global standard. Earthscan, London, UK.\u003c/li\u003e\n \u003cli\u003eMekonnen, M.M., Hoekstra, A.Y. (2012) A Global Assessment of the Water Footprint of Farm Animal Products. Ecosystems 15, 401-415.\u003c/li\u003e\n \u003cli\u003eMekonnen, M.M., Neale, C.M.U., Ray, C., Erickson, G.E., Hoekstra, A.Y. (2019) Water productivity in meat and milk production in the US from 1960 to 2016. Environment International 132, 105084.\u003c/li\u003e\n \u003cli\u003eMolden, D., (2007) Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture. Earthscan/International Water Management Institute, London, UK / Colombo, Sri Lanka.\u003c/li\u003e\n \u003cli\u003ePoore, J., Nemecek, T. (2018) Reducing food\u0026rsquo;s environmental impacts through producers and consumers. Science 360, 987.\u003c/li\u003e\n \u003cli\u003eSteinfeld, H., Gerber, P. (2010) Livestock production and the global environment: Consume less or produce better? Proceedings of the National Academy of Sciences 107, 18237-18238.\u003c/li\u003e\n \u003cli\u003eSteinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., Haan, C. (2006) Livestock\u0026rsquo;s long shadow: environmental issues and options. Food and Agriculture Organization, Rome.\u003c/li\u003e\n \u003cli\u003eWeindl, I., Bodirsky, B.L., Rolinski, S., Biewald, A., Lotze-Campen, H., M\u0026uuml;ller, C., Dietrich, J.P., Humpen\u0026ouml;der, F., Stevanović, M., Schaphoff, S., Popp, A. (2017) Livestock production and the water challenge of future food supply: Implications of agricultural management and dietary choices. Global Environmental Change 47, 121-132.\u003c/li\u003e\n \u003cli\u003eWisser, D., Grogan, D.S., Lanzoni, L., Tempio, G., Cinardi, G., Prusevich, A., Glidden, S. (2024) Water Use in Livestock Agri-Food Systems and Its Contribution to Local Water Scarcity: A Spatially Distributed Global Analysis. Water 16, 1681.\u003c/li\u003e\n \u003cli\u003eWolkeba, F.T., Mekonnen, M.M., Brauman, K.A., Kumar, M. (2024) Indicator metrics and temporal aggregations introduce ambiguities in water scarcity estimates. Scientific Reports 14, 15182.\u003c/li\u003e\n \u003cli\u003eZimmer, D.B., Renault, D., (2003) Virtual water in food production and global trade : Review of methodological issues and preliminary results, in: Hoekstra, A.Y. (Ed.), Virtual water trade: Proceedings of the International Expert Meeting on Virtual Water Trade. UNESCO-IHE, Delft, The Netherlands.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9360861/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9360861/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Livestock water use is dominated by the water required to grow feed, yet long‑term, globally consistent reconstructions remain limited. We compile a country‑level, animal‑specific assessment of green and blue water footprints (WFs) for seven livestock categories across 1972–2023 using a harmonized accounting framework linking feed requirements, trade‑weighted feed WFs, and livestock production statistics. Global livestock WF increased from 1,979 km³ yr⁻¹ in 1972 to 3,044 km³ yr⁻¹ in 2023. Ruminants remain dominant water users, with beef cattle (34%) and dairy cattle (25%) contributing the most. In 2023, feed production accounted for 98% of total livestock WF, while direct water uses were 2%. China showed the largest increase (147→586 km³ yr⁻¹; +299%) and contributed 19% of the global total livestock WF; the top five countries accounted for 44%. Unit WFs declined markedly (e.g., pork −70%, eggs −68%, chicken −66%), yet unsustainable surface and groundwater use contributed 4% of the total and 52% of the blue WF.","manuscriptTitle":"Disentangling feed conversion efficiency and feed sourcing in global livestock water footprint trends, 1972–2023","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-13 04:47:19","doi":"10.21203/rs.3.rs-9360861/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-water","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natwater","sideBox":"Learn more about [Nature Water](https://www.nature.com/natwater/)","snPcode":"44221","submissionUrl":"https://mts-natwater.nature.com/cgi-bin/main.plex","title":"Nature Water","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"eb625945-6bc6-42af-910c-ac044938fb09","owner":[],"postedDate":"May 13th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-05-12T07:44:42+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"4","date":"2026-05-11T18:00:00+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":66039843,"name":"Earth and environmental sciences/Hydrology"},{"id":66039844,"name":"Physical sciences/Engineering/Civil engineering"}],"tags":[],"updatedAt":"2026-05-13T04:47:20+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-13 04:47:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9360861","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9360861","identity":"rs-9360861","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}