Sindh’s water volatility

During 1980-2000, the rural landscape of Punjab saw a revolution that led to over 25 percent increase in area under cultivation by the end of century. During that same period, crop acreage in southern Sindh remained flat, effectively declining by two percent. If there were one reason to sum up the divergent paths taken by the two provinces that would be tubewell-subsidy.

While Punjab doubled its groundwater abstraction from 25 million-acre feet (MAF) to over 50MAF, farmers from Sindh were less lucky as salinity of water tables in the province meant that irrigation in the region would remain dependent on availability of surface water.

Until mid-2000s, this did not pose a serious challenge as availability of surface water during both cropping seasons remained on an upward trajectory. A host of natural factors (such as regular precipitation cycles) and human-led interventions (such as lining of canals) ensured that resources in the canal system were adequate to meet the demands of lower riparian, especially in the aftermath of water accords of 1990 that had set a mechanism for resource sharing in the river basin.

Over the last 10 years, that arrangement has begun to buckle under the pressure of climatic vagaries. Using water availability during kharif season (whole of Indus Basin Irrigation System (IBIS) numbers used as proxy), BR Research ran a comparison against kharif crop cultivation in Sindh over the past three decades.

Why acreage? Because total crop output is equally a function of yield, which in turn is also affected by such variables as seed variety and fertilizer application – hard to distinguish in a rudimentary analysis. Aggregate crop acreage, on the other hand, is an isolated variable - although not entirely but in some part dependent on water availability.

Between 1993 and 2000, Sindh recorded three periods of double-digit decline in kharif crop acreage, even as canal water to farmgate (in all of IBIS) peaked at about 51MAF, supply-increase inching slowly at average 0.85 percent per annum.

In comparison, performance of provincial agriculture in 20 years since – as measured by kharif crop acreage - has been impressive, as it only saw two seasons of double-digit decline, whereas growth exceeded five percent in at least four years.

However, the graph plots of two variables reveal an interesting pattern – that of volatility inversion. Before 2005-06, it was crop acreage that saw serious swings in year on year percent change into and out of negative growth territory. Since then, while annual change in kharif season water availability has exceeded five percent at least as many times but has been immediately followed by a period of water falling short of previous year levels.

The increased volatility, as the graph shows, means that water availability compared to previous years has seen double-digit changes in either direction in 9 out of last 13 years. While this has led to years of – what Economic Survey proudly labels “bumper crop” – it also means more frequent periods when farmers are worse-off nominally compared to previous years.

What does this mean for Sindh’s agriculture performance scorecard? On one hand, farm output for all major crops such as rice, wheat and sugarcane (except for cotton) have recorded peak acreage and yield in the past five years alone. Even Sindh’s cotton has seen its yield overshoot thousand kg per hectare in the last decade.

Yet, the last ten years alone saw two instances where crop output declined by over one-fifth, once in the 2010-11 flood year followed by the semi-drought conditions witnessed in 2018-19. Such extreme events are not only accompanied by obvious human suffering. They also set back those associated with the sector by depressing incomes – often permanently as ever greater number of families move to slums in urban centres and out of farming to seek more stable modes of earning.

Consider that agriculture takes up over 90 percent of Pakistan’s average water availability of 130MAF. Of this, close to 45 percent is groundwater exclusively to non-coastal provinces of Punjab and KP (and some parts of Balochistan). Last year, surface water in IBIS (for both seasons) recorded its lowest level at 72MAF since 1988, which remember, has to be equitably shared between all provinces.

Given this context, the wholescale dependence of Sindh’s agriculture on canal irrigation system is highly risky. Whether the solution requires moving to desalination technology, reducing economic dependence on the sector, and/or efficient agronomic techniques needs a whole-of-government approach.

But if there is any lesson from last year’s kharif performance, it’s that time of cleverly declaring “bumper crop” seasons should be over. If anything, those should give policymakers moment to take stock and plan for vagaries of climate in the coming years.

Sources used: Agriculture Census 2010; Agriculture Statistics of Pakistan (annual volumes); SBP State of the Economy Reports; UNDP – Development Advocate, Pakistan 2016; kharif season water availability between 1982 and 1994 interpolated based on total water in IBIS and long-term average share of kharif season in water distribution; water volume is based on surface water at farmgate and not canal-head. Definition of kharif crop acreage for Sindh includes rice, cotton and sugarcane only.