planted annually, primarily in monoculture systems (TASS, 2002). Furthermore, about 5.5 million stocker Agriculture in the Texas High Plains depends heavily on irrigation cattle (about 25% of U.S. total; TASS, 2002; USDA with water withdrawn from the Ogallala aquifer at nonsustainable rates. Our hypothesis was that integrating crop and livestock systems Natl. Agric. Stat. Serv., 2003) are shipped into the High would reduce irrigation water use, maintain profitability, and diversify profitability, and impact on water use of (i) a cotton mono-Texas Tech Univ., Lubbock, TX 79409-2122; R. Kellison, Silver Creek culture system managed by best management practices Farm, Lockney, TX 79241; E. Segarra, Dep. of Agric. and Appl. Econ., Texas Tech Univ., Lubbock, TX 79409; T. Wheeler, Texas A&M for the area and (ii) an integrated cotton-forage-livestock Agric. Exp. Stn., Lubbock, TX 79403; P. Dotray, Dep. of Plant and system. This long-term project continues, but results of Soil Sci., Texas Tech Univ., Lubbock, TX 79409-2122 and Texas Coopthe first 5 yr are presented here. erative Extension, Lubbock, TX 79403; J.C. Conkwright, High Plains Underground Water Conservation District No. 1, Lubbock, TX 79411-2499; and V. Acosta-Martinez, USDA-ARS, Lubbock, TX 79415.

Arid and semiarid landscapes are often fragile and, thus, vulnerable to both natural weather extremes and human activities. Climate change and increasing demands for food to meet needs of a growing global population will place greater stress on these environments. Cropping and livestock systems have generally succeeded in these regions to the extent that the environment could be altered through water development and irrigation. Water sources for irrigation, including surface and groundwater, are declining in quality and quantity. Improvements in irrigation use efficiency now exceed 95% but often have led to increased water use, instead of water savings, as more systems have been installed. Also, as groundwater becomes scarce, more energy is required to extract water from greater depths. Increasing demands for alternative water uses and depletion of historic water sources make many irrigated systems in dry climates nonsustainable. The Texas High Plains exemplifies these challenges, where agriculture depends heavily on irrigation at nonsustainable rates of water extraction from the Ogallala aquifer. Today, agriculture uses about 95% of total water withdrawn from the aquifer. Crop rotations and integrating crop and livestock systems could reduce irrigation water use and diversify income compared with a monoculture. This region was historically a grazing land ecosystem offering opportunities for pastoral systems and benefits from diversification. Long-term comparisons of two irrigated systems [a cotton (Gossypium hirsutum L.) monoculture and an integrated cotton-forage-beef cattle system] in the Texas High Plains have demonstrated water savings of about 25% achieved through integration, while remaining economically viable and diversifying income sources. Additional benefits included reduced soil erosion, lower chemical inputs including a 40% reduction in N fertilizer, improved soil microbial and enzymatic activities, enhanced C sequestration, and greater rainfall infiltration than the monoculture system. Greater annual crop yields can shift short-term profitability to the monoculture system, but long-term sustainability is likely to depend on environmental benefits and water savings achieved by integrated systems. Challenges include existing large investments in local infrastructure focused on monoculture systems, producer adoption of alternative strategies, enhanced knowledge and management skill required, and a need for more research. Dryland agriculture will increase with remaining water diverted to other uses including livestock, municipalities, manufacturing, and energy generation. Technological advances can increase water savings but can also decrease system resilience with dependence on nonsustainable external buffers. Regional resource and economic stability will likely depend more on internal resilience of appropriately integrated plant and animal agricultural systems.

Texas High Plains agriculture, largely dependent on water from the Ogallala aquifer for irrigation, exemplifies semiarid agricultural regions where irrigation is used at nonsustainable rates of extraction. Integrating crop and livestock systems has been suggested to conserve water and to achieve other environmental and economic goals compared with monoculture systems. From 1998 to 2008, two large‐scale systems, with three blocks in a randomized block design, compared irrigation water, productivity, chemical inputs, and specific pests of (i) a cotton (Gossypium hirsutum L.) monoculture, and (ii) an integrated three‐paddock system that included cotton in a two‐paddock rotation with grazed wheat (Triticum aestivum L.) and rye (Secale cereale L.) and the perennial variety WW‐B. Dahl old world bluestem (OWB) [Bothriochloa bladhii (Retz) S.T. Blake] in a third paddock for grazing and seed production. All paddocks were irrigated by subsurface drip. Angus crossbred beef steers (Bos taurus; initial BW 229 kg; SD = 33 kg) grazed 185 d from January to mid‐July each year. During the 10 yr following the establishment year, cotton lint yield was similar and averaged 1370 kg ha−1 for both systems. Bluestem seed yield averaged 25 kg pure live seed (PLS) ha−1. Steers gained 139 kg on pasture and 0.79 kg d−1. Per hectare, the integrated system used 25% less (P < 0.001) irrigation water, 36% less N fertilizer, and fewer other chemical inputs than monoculture cotton. Integrated production systems that are less dependent on irrigation and chemical inputs appear possible while achieving goals of sustainability, fiber production, and food security.

Effects of factors influencing spatial and temporal variability of crop yields are usually expressed in crop growth patterns. Consequently, monitoring crop growth can form the basis for managing site‐specific farming (SSF). This experiment was conducted to determine whether crop growth patterns forecast grain yields. Effects of irrigation rates (50 and 80% evapotranspiration, ET), elevation, soil texture, soil NO3‐N, arthropods, and diseases on corn (Zea mays L.) growth and grain yield were evaluated at Halfway, TX, in 1998 and 1999. Data on plant height, leaf area index, leaf dry matter, stem dry matter, and ear dry matter were collected from geo‐referenced locations (DGPS). These data were used to derive total dry matter, crop growth rate, and net assimilation rate (NAR). Grain yields at DGPS locations were classified into four distinct clusters. In 1998, a dry season, clusters were strongly influenced by elevation and soil texture. Grain yields were higher at high elevations where water use was high and soil texture was heavy compared with low elevations. Grain yields at low elevations also were reduced by common smut [Ustilago zeae (Beckm.) Unger] that preferred dry conditions. In 1999, a relatively wet season, clusters included areas with different elevation and soil texture combinations. Measured parameters forecast grain yields better in 1998 than in 1999. Differences in NAR were evident before the 12‐leaf stage, making NAR a potentially useful measurement for early in‐season management decisions. Biomass measurements, for which differences were observed after the 12‐leaf stage, also may be used to formulate decisions for both in‐season and the following season's management.

PurposeThe purpose of this study is to explore the relationship between migration, remittances and agricultural productivity by applying the new economics of labor migration model in the context of north‐west China. The specific objectives are to examine the impacts of rural out‐migration on agricultural productivity in various farming systems, and whether remittances have been reinvested in agriculture.Design/methodology/approachCross‐sectional household survey data from three townships were analyzed with the three‐stage least squares (3SLS) regression model.FindingsIn multi‐cropping small farming systems, at least in the short run, the loss resulting from losing family labour on lower‐return grain crop production is likely to be offset by the gain from investing in capital‐intensive and profitable cash crop production.Originality/valueThis study provides empirical evidence for the MELM theory. It expands Taylor et al.'s studies by comparing investment behavior and production choices among multiple farm activities, and enriches previous studies by showing that the relation between remittances and agricultural investment depends on the farm activities' profitability.