Piling up reactive nitrogen and declining nitrogen use efficiency in Pakistan: a challenge not challenged (1961–2013)

Pakistan is situated between the latitudes and longitudes of 24°–37° North and 61°–75° East, covering an area of 796 095 km² with a coastline of 1046 km. The climate of Pakistan is arid to semi-arid (around 24% of the area gets rainfall between 250–500 mm per year, and around 60% gets less than 250 mm) with a mean annual temperature of 21^oC, which has increased by 0.35 °C since 1960, with an average increase of 0.08 °C per decade (Gadiwala and Burke 2013).

Out of the 79.6 Mha land area of Pakistan, only about 22 Mha (23%) is available for cultivation: 18 Mha of which is irrigated, and 4 Mha is rainfed. Forests, both natural and man-made, cover about 4% of Pakistan. The country has one of the world's largest contiguous irrigation systems (Kahlown et al 2005).

During the last five decades, Pakistan has experienced numerous changes in its agro-food system (table S1). A massive increase in total population by 259%, and decrease in rural population (18%) was recorded. There was a slight decrease in total agricultural area (1%) and arable land (4%). Pakistan's agriculture became intensive during the past 53 years, with an increase in cropping intensity, use of machinery, energy consumption, fertilizer application, and irrigation. An increase in the yield of almost all crops took place during this period, particularly for sugarcane, maize, and citrus (table S1). Apart from these changes, significant changes have also been observed in soil (soil salinization, soil erosion) and water resources, which have degraded with time (Qureshi et al 2008, Tahir and Rasheed 2008).

Diet patterns have also changed with time with the increase in the consumption of protein from animal sources. During the past five decades, Pakistan has experienced a 14% increase in total protein consumption, with the share of animal protein increased by 48% (table S1). Combined with the increase in population, this increase in the share of animal protein has caused a substantial increase in the consumption of animal products.

Recently, a considerable increase in poultry and livestock populations was observed in all animal categories, particularly for goats and buffalo. To maintain supply of feed to this increasing livestock population, the portion of grain crops designated to livestock feed has been increased; however, this was insufficient, and the deficit was fulfilled through imported feed products of maize, barley, and soybean derivatives. The total import and export of agricultural products increased significantly, but in terms of weight, the increase in import was around three-fold compared with export being 11-fold. Meanwhile, in terms of finance the increase in the value of imports was 34-fold compared with 25-fold for exports (table S1). The share of the agricultural sector in Pakistan's economy (GDP) decreased by 40% during the past 53 years. The energy production in Pakistan has always been insufficient to fulfill the country's demand, and deficit was met through import. The increase in energy production, consumption, and import during the same period was thirteen-fold, eight-fold, and six-fold, respectively (table S1).

The total N input to cropland in Pakistan has experienced a ten-foldincrease during the last five decades from 408 GgNyr⁻¹ (1961–1965) to 4636 GgNyr⁻¹ (2009–2013) (figure 2), which means an increase from 30–222 kgNha⁻¹yr⁻¹ in terms of area under agriculture and from 512–5823 kgNkm⁻²yr⁻¹ throughout the country.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

All the components of N input (synthetic N, manure, N fixation, atmospheric deposition) increased with the passage of time. Among these inputs, manure N (44%) was the highest contributor in total N inputs, followed by N fixation (23%), atmospheric deposition (17%), and synthetic N (16%) during 1961–1965. However, with time, rapid increase in the use of synthetic N caused it to surpass all other inputs: recently (2009–2013), synthetic N (68%) was the biggest contributor in total N input to croplands, followed by manure (23%), atmospheric deposition (5%), and N fixation (4%) (figure 2).

The urea production in Pakistan progressed rapidly with the installation of various urea formulation plants to fulfill the country's demand. During the past 53 years, 83% of the N fertilizer consumed in Pakistan came from local production, and 17% from imports. However, only about 1.2% of the N produced in the country was exported during the study period (figure S1).

A major portion of the N fertilizer is applied to cereal crops (59%) in Pakistan; out of which 92% is applied on wheat and rice only. The share of N applied to fruits (3.5%), vegetables (2%), sugar crops (8%), oilseeds (1%), and root and tuber crops (0.6%) is minimal; fiber crops (20%) are an exception. On the whole, Pakistan consumes about 3% of the total N used in the world (table S2).

The rapid increase in the production of agricultural products (crops and livestock) during the last five decades made it possible to meet the challenge of feeding the rapidly growing population in Pakistan. Overall production of N by different crop and livestock products during 1961–1965 was 302 GgNyr⁻¹ (crops: 78%; livestock: 22%), which increased up to 1436 GgNyr⁻¹ (crops: 73%; livestock: 24%) during 2009–2013 (table 2). Cereals, oilseeds, and milk products were the significant contributors in the total N production used for food and feed in the country during the study period.

During the last five decades, Pakistan traded for 319 different crop and livestock commodities, and the volume of trade increased with time in order to fulfill the requirements of the growing population. The total N trade increase from 32 GgNyr⁻¹ (1961–1965) to 194 GgNyr⁻¹ (2009–2013) has an important share in the N cycle of Pakistan (table 2). During 1961–1965, the share of N import and export in total N trade in Pakistan was 63% and 37%, respectively, with a major contribution by crop products in both total import (98%) and export (81%) (table 2). With the passage of time the indigenous production of crop and livestock products increased, and dependence on import decreased to almost an even share of import (52%) and export (48%) during 2009–2013, with more than a 95% contribution in both by crop products (figure S2). The net N import (import − export) during the study period was variable: it was positive during the 1960s, remained negative during the 1970s–1990s, and mostly remained positive from 1990–2013 because of large dependence on imports.

Pakistan traded with 212 countries during the past 25 years (1989–2013). The top five countries in terms of N export were the United Arab Emirates, Afghanistan, Kenya, Iran, and Saudi Arabia, with the total N exported during the same period being 239, 181, 77, 64, and 46 GgN, respectively (figure S3). Similarly, the United States of America (417 GgN), India (311 GgN), Australia (262 GgN), Canada (206 GgN), and China (48 GgN) were the top five countries for Pakistan in terms of N import (figure S3).

There has not been much change in the total protein consumption in Pakistan during 1961–1965 (3.29 kgNcap⁻¹yr⁻¹) to 2009–2013 (3.76 kgNcap⁻¹yr⁻¹) (figure 3), with only a 12.5% increase during the last five decades. Diet patterns changed with time, with a gradual decrease in the share of protein derived from crop sources (from 73%–59%) and an increase in the share of animal protein (from 27%–40%).

During 1961–1965, only 58% of the Nr applied to the cropland was recovered upon harvesting, and the remaining 42% was surplus N (171 Gg), the majority of which was lost to the environment. The NUE continuously declined in Pakistan, decreasing gradually until only 23% NUE was recorded during 2009–2013, which means that a massive 77% of N (3581 Gg surplus N) was either retained in the soil or lost to the environment (figure 4). Such massive accumulation of Nr into the soil or environment is a direct threat to the air and water quality in Pakistan.

Zoom In Zoom Out Reset image size

The gap between the N input to cropland and N used by the crop widened with time, and resulted in huge amounts of surplus N lost to the environment in several ways. Gaseous emission of N both from manure and synthetic fertilizer is one way of N loss. Gaseous emission of N (N₂O, NO, and NH₃) increased with time and with the increase in the application of manure and synthetic fertilizer. Loss of NH₃, N₂O, and NO emission increased from 70, 10, and 1 GgNyr⁻¹ (1961–1965) to 1023, 155, and 46 GgNyr⁻¹ (2009–2013), respectively (figure 5). Overall, the total gaseous N emission of these gases was 20% (82 Gg N) of the total N input applied on cropland during 1961–1965, which increased to 26% (1224 Gg N) during 2009–2013.

Zoom In Zoom Out Reset image size

Major changes in the N flux of Pakistan's agro-food system are summarized in the form of a cycle (figure 6).

Zoom In Zoom Out Reset image size

The considerable increase (397%) in the population of Pakistan between 1950 (38 million) to 2015 (189 million) (FAOSTAT 2016), which is further expected to rise to 350 million by 2050 (Nizami 2010), is a main driver of the enormous agricultural expansion within the country during the last five decades. Fulfilling agricultural and livestock demands of this growing population is not possible without intensive agricultural practices, which rely heavily on excessive inputs, particularly of fertilizers.

Mineral fertilizers, predominantly N, played an important role in enhancing production of crops on Pakistani soils, which are alkaline and calcareous in nature (Alam et al 2005). The increase in fertilizer use was tremendous in the era of the green revolution, with an average 1751% increase in fertilizer use between 1961–1965 and 1981–1985. However, this increase was mainly for nitrogenous fertilizer (65 GgNyr⁻¹ to 953 GgNyr⁻¹) with a minimum amount of phosphorus (3 GgNyr⁻¹ to 280 GgNyr⁻¹) and no or negligible potassium (0 GgNyr⁻¹ to 27 GgNyr⁻¹) (IFA 2016). The consumption of N fertilizer in Pakistan increased around 50-fold during the past five decades (figure 2), and was about 90%–95% of the total mineral fertilizer application until the last decade. An increase in the N fertilizer input during 1961–2013 was also observed in other developing countries (although less than Pakistan) adjacent to Pakistan, like Bangladesh (3567%), China (2627%), India (3932%), and Sri Lanka (370%) (IFA 2016).

Moreover, the changing diet pattern in Pakistan, with a gradual increase in the share of animal protein (from 27%–40%) and decrease in the share of crop protein (from 73%–59%) (figure 3), during the past 53 years further suggests that Pakistan will require higher poultry and livestock populations to fulfill protein demands, which already considerably increased (2597% and 268%) during 1961–2013 (FAOSTAT 2016). Although the livestock sector itself consumes a lot of N, it also supplies N to crops in the form of manure. Significant increase in N application from manure (180 GgNyr⁻¹ to 1041 GgNyr⁻¹) (figure 2) took place in Pakistan, although much less in comparison with synthetic N. Our findings further revealed that out of the total N excreted by livestock in Pakistan around 25% actually reaches the field (figure 6). Improper storage and poor management cause the majority of the N loss from this valuable resource to the environment. The livestock sector is the biggest contributor of greenhouse gas emissions in Pakistan's agriculture sector (Khan et al 2011a). Loss of N from the manure management system in Pakistan (∼75%) was also higher than in other neighboring developing countries (Bangladesh, China, India, and Sri Lanka) (data not shown). In comparison with synthetic N and manure, the increase in N input from biological N fixation was merely 89% during the study period (figure 2).

The cropping intensity, which has also increased tremendously during the study period, has left ill effects on soil health because there is no time for recovery and/or for a fallow season. Introduction of high yielding varieties, particularly maize (hybrids), potato, wheat, rice, and cotton, has urged farmers to apply more fertilizers. The fertilizer requirement of every crop changes with time based on its genetic makeup and yield potential. However, in Pakistan, fertilizer requirements of different crops are not updated with time. Most of the farmers lack information regarding the soil nutrients of their farm lands (Hamid and Ahmad 2001). Therefore, farmers generally use fertilizer based on their own experiences, which results in highly imbalanced fertilization. The ratio of N, P₂O₅, and K₂O nutrients in Pakistan during 1999–2000 and 2001–02 was, on average, 1:0.28:0.01 (FAO 2004). This imbalanced fertilizer use has depleted the soils of other essential nutrients, mainly P and micronutrients. As per recent estimates, about 30%–40% of Pakistan's soils are deficient in K. Zia et al (2000) revealed that soil fertility in Pakistan has deteriorated so much that even if chemical fertilizers are used in balanced amounts the soil will be unable to sustain crop productivity. Applying fertilizers following inappropriate application methods is also a common practice among the farming community. All these problems result in inefficient use of fertilizer, and therefore the per hectare yield in Pakistan has not increased proportionate to the increase in the use of fertilizer (Tahir et al 2008).

Recovery of N in soil–plant systems seldom go above 50%, which means that around more than half is being lost (Abbasi et al 2003). This increasing gap between the N applied and taken up by crops has resulted in a significant reduction in NUE in Pakistan from 58% during 1961–1965 to 23% in 2009–2013 (figure 4), which led to an increase in surplus N during the same period (figure 4). Significant changes (at the per hectare level) in total N input to crops, N surplus, and NUE were not only seen in Pakistan (569%, 1112%, and −58%) (figure 4), but were also observed in neighboring countries like Bangladesh (282%, 384%, and −22%), China (704%, 2985%, and −64%), India (308%, 465%, and −36%), Nepal (55%, −16%, and 31%), and Sri Lanka (124%, 133%, and −8%) during 1961–1965 and 2005–2009 (Lassaletta et al 2014). China and Pakistan are the maximum N expanding countries, with minimum NUE in the region. These changes reflect that the developing countries require proper attention and better management regarding N in order to control the outburst of N-related environmental issues in the future.

Extensive use of urea as an N source with very low P and low or no K, is one of the major reasons for low NUE. The share of urea ranges from 72%–85% of total N consumed in Pakistan during 1981 to 2013 (IFA 2016). Urea has been reported to have a lower fertilizer use efficiency compared with other N sources (Zaman et al 2008). Urea is highly susceptible to be lost in gaseous emissions, particularly as NH₃, as soon as it is applied to croplands, especially under climatic conditions like in Pakistan. A field experiment on maize showed maximum grain yield in calcium ammonium nitrate, followed by ammonium sulfate; meanwhile, urea exhibited the lowest yield (Abbasi et al 2013). An arid to semi-arid climate, high temperature, calcareous nature of soils, reduced soil moisture, and lack of awareness among farmers regarding proper application of N fertilizer further adds to the problem of higher N losses.

The low productivity of Pakistan's agriculture is another major area of concern, and the inputs including N fertilizer does not yield as per their potential. The average yields obtained at farmers' fields hardly exceed 30% of their potential. Reports show that the yield gaps of wheat, rice, maize, sugarcane, and cotton are 72%, 83%, 88%, 78%, and 72%, respectively, as compared to their potential yield under experimental conditions (Khan et al 2007). Furthermore, the research potential yields obtained in Pakistan are less than those demonstrated in many other developing and developed countries (Iqbal and Ahmad 2005).

Shortage of agricultural scientists and lack of research funding have also played a major role in the mismanagement of N fertilizer. During 1988, the number of agricultural scientists working in the country was only 44 per million people, much less compared with countries like Egypt (300), UK (1400), and USA (2360) (John Mellor Associates 1994). Moreover, the financial incentives and opportunities for career growth offered to highly skilled scientists were very limited (FAO-GOP 2002). Although the scenario has changed a bit from 2000, there is still a need for further improvement.

Regions with more requirement of N in food and feed fulfills their demands through imports from countries that are either consuming less N or are producing a lot of N in the form of food and feed products. This imbalance between countries for livestock and crop production and consumption is also one of the reasons behind inefficient N use, and results in a surplus of N because of incompetence in closing the nutrient cycles (Bai et al 2014, Lassaletta et al 2014, Billen et al 2015, Leip et al 2015, Strokal et al 2016).

Pakistan is neither an importer nor an exporter of N, with almost equal share of both import (52%) and export (48%). N trade in Pakistan is heavily dominated by crop products (>90%), with livestock products sharing below 10% (table 2).

The increase in surplus N at such a large scale in Pakistan is leaving air and water resources at greater risks of being contaminated with N. Numerous studies have already reported nitrate contamination in groundwater resources throughout Pakistan, with synthetic N fertilizer been highlighted as the main cause. During 2007–2008, results of 747 samples (from surface water and groundwater), taken throughout the country reveals that 19% of the samples had nitrate concentrations (23% in Balochistan and Punjab provinces) beyond safe limits (Tahir and Rasheed 2008). However, this number has increased to 23% in 2017 (Podgorski et al 2017) (figure S4), with an additional 12% on the margin of nitrate contamination (7–10 mg l⁻¹), which can further make the number go beyond 35% in the near future. Various other studies have reported nitrate contamination in big cities like Kasur (Farooqi et al 2007), Islamabad, Rawalpindi (Kazmi and Khan 2005), Quetta, Faisalabad (PCRWR 2005), and Lahore (Naeem et al 2007). Moreover, the considerable increase in air pollution in Pakistan, particularly of NH₃, N₂O, and NO, was another important consequence that scientists in Pakistan have related to the excessive use of N (Iqbal and Goheer 2008, Khan et al 2011b). Our estimates have also mentioned significant increase in gaseous N emissions (from 82 GgNyr⁻¹ to 1224 GgNyr⁻¹). Emissions on the national scale can be correlated with the total use of nitrogenous fertilizers on arable crops.

Improving NUE is the most cost-effective and efficient way to reduce N losses from agricultural sources to the environment. Progress in NUE requires better management practices at all levels. The application of synthetic N fertilizers should be according to the crop demand and nutritional status of the soil. Following proper fertilizer application methods, like deep placement or applying irrigation immediately after application of N fertilizer, is also very effective in reducing N losses (Maqsood et al 2016). Another way to accomplish this is to use new types of N fertilizers, such as controlled-release fertilizers (polymer coated fertilizers), and urease and nitrification inhibitors. Creation of new cereal crop cultivars, which are capable of fixing N as legumes do, based on advances in molecular biology and biological technology is needed (Zheng et al 2002). Moreover, increasing the use of cultivation-induced BNF (Roy et al 2002), escalating recycling of Nr within agroecosystems (Smil 2002), redistributing Nr to areas where Nr is needed for food production from areas where Nr production is high (Erisman et al 2001), and providing incentives to farmers for reduced N over-fertilization (Howarth et al 2002b) are some of the other ways that could significantly help solve the problem, if followed consistently.

Manure N is the second biggest contributor in total N input in Pakistan. Despite the fact that the amount of N excreted by livestock in Pakistan is more than the N applied through synthetic fertilizers (figure 6), and despite its countless benefits on soil health, the use of this N resource is not given much importance. Segregation of livestock farms from arable farms, and the laborious work to apply manure compared to using synthetic N fertilizer makes its use in agriculture more difficult. Therefore, focusing more on efficient utilization of manure as a N source in balance with synthetic N according to the crop demand can also contribute a great deal in alleviating N-related environmental issues. This can also effectively reduce dependence on synthetic N fertilizers.

Improvement in the level of extension services to farmers, initiation of various educational programs, and formulation of adequate and appropriate guidelines toward balanced fertilizer use is required in Pakistan.

Several developed countries have already announced various policies (for example, the European Nitrate Directive, The Fertiliser Ordinance, and the Groundwater Directive) toward efficient N management (Salomon et al 2016, Bouraoui and Grizzetti 2014). Denmark has increased its NUE from 20%–30% to 40%–45% during the past 30 years as an effect of the policies and actions taken to mitigate the effects of reactive N (Dalgaard et al 2014). China, the biggest developing country, which consumes about 30% of N fertilizer in the world, has also started a series of policies to tackle the excessive fertilization in agriculture; these policies include having a 'zero increase of fertilizer by 2020', and the 'Clean Water Act' (Zhou et al 2016). Therefore, the Pakistan government should also review its policy toward fertilizer use in the country; proper legislation focusing on balanced fertilization is needed.

The success of these approaches is not possible unless there exists a level of strong collaboration and cooperation among soil scientists, ecologists, agronomists, agricultural economists, and politicians (Galloway et al 2002).