Fertilized croplands unintentionally export large amounts of reactive nitrogen (N), which degrades water and air quality and contributes to climate change. In this dissertation, I focus on how these reactive N losses are likely to change in the near future as agriculture intensifies in the tropics, and ecological intensification strategies to mitigate N losses are more widely adopted. I use a combination of empirical field measurements in Mato Grosso, Brazil and Skåne, Sweden, literature review, and statistical models to quantify trends. In chapter one, I quantified emissions of nitric oxide (N₂O) and nitrous oxide (N₂O) in forest, single cropped soybean, and N-fertilized double-cropped soybean-maize at three nitrogen fertilizer levels within the largest area of recent cropland expansion on earth, in the Amazon and Cerrado biomes in Mato Grosso, Brazil. I found that NO emissions do not increase when forests are converted to croplands under current fertilization levels, and that NO will respond more strongly than N₂O fluxes to increases in fertilizer applications. In chapter two, I investigated anion exchange capacity and soil nitrate (NO₃¯) pools in deep soils in Mato Grosso, Brazil in the southern Amazon. I found that soil NO₃¯ pools in the top 8 m increased from 143 kg N ha¯¹ in forest to 1,052 and 1,161 kg N ha¯¹ in soybean and soybean-maize croplands. This NO₃¯ accumulation in croplands compared with forest soils matched the estimated amount of surplus N from the croplands, and could be explained by the soil’s positive charge through its anion exchange capacity. In chapter three, I conducted a meta-analysis of the effects of fertilization amount on of NO₃¯ leaching, N₂O emissions, NO emissions, and ammonia (NH₃) volatilization, totaling over 1,000 observations. I found that the relationship between N inputs and losses differed little between temperate and tropical croplands, although total NO losses were higher in the tropics. Among the potential drivers I studied, the N input rate controlled all N losses, but soil texture and water inputs also controlled NO₃¯ leaching losses. In chapter four, I explored the differences in NO₃¯ leaching, fertilizer N use efficiency, and soil N cycling in perennial wheat, which is being domesticated as a more sustainable alternative to annual crops, and annual wheat at a long-term experimental site in Skåne, Sweden. I found that NO₃ leaching was more than two orders of magnitude lower in perennial wheat, overall ecosystem recovery of fertilizer was quite high and not significantly different between perennial and annual wheat after the first growing season, and that measures of soil N cycling were largely the same between both crops. Together, these chapters highlight that reactive N losses will remain a critical global challenge in the coming decades, but that there are also key opportunities to reduce N losses by increasing the use of perennial crops and focusing tropical agricultural intensification on Oxisol soils which buffer against NO₃¯ leaching.