The Scale and Complexity of Salinity Impacts on Sri Lankan Rice Farming Systems: Actionable Insights

Abstract

Saline-affected rice (Oryza sativa L.) production environments in Sri Lanka can be divided into three categories including: late-season salinity in irrigated mega-cultivation environments during the minor cultivation season where soil EC ≥7 dSm−1, late season salinity in rain-fed farming systems in the west, southwest, and eastern coastal line during the minor cultivation season where soil EC ≥20 dSm−1, and early season salinity in selected irrigated and rain fed sites during major and minor cultivation seasons as a result of residual overload of salts that was not washed off due to inadequate rain. In the west and southern coast early season salinity salinity can exceed EC ≥12 dSm−1. The proposed zones of saline-afflicted production environments permit designing of target ideotypes and locally adapted rice varieties. Accordingly, high yielding, 3 to 3.5 months duration varieties that are tolerant at >7 dSm−1 are recommended for intensive irrigated farming systems affected due to late season salinity (panicle initiation stage of the crop, PI); high yielding, 2.5 to 3 months duration varieties can avoid late season salinity in intensive irrigated farming, and varieties tolerant up to EC = 12 to 20 dSm−1 throughout the crop life including seedling and PI stages can target saline affected, semi-subsistence rice cultivation in rain-fed systems. In fact, secondary salinization in local rice farming environments is resulting from interaction among multiple factors; therefore, system-level interventions are necessary to manage the impacts.

1. Introduction

1.1 Background

Rice (Oryza sativa) is the staple food of an estimated 3.5 billion people from East and South Asia, the Middle East, the West Indies, and Latin America, accounting for 50–80% of their daily calorie intake [1]. It is also the primary source of income and employment for more than 200 million households in developing countries [2]. Rice is the main staple providing over 45% of the total caloric and 40% of the protein requirements for the Sri Lankan population [3, 4]. In the country, rice cultivation occupies 16% of the total land area and 34% of the total cultivated area whereby 11.6% of the total local population and 32% of the total labor force are directly engaged in the rice sector [5]. The rice sector contributes 7% to the national agricultural GDP [6], and therefore is associated with food and nutrition security and is critical for generating livelihood income for the Sri Lankan population.

Salinization of arable soils is an increasing challenge in global agriculture. Worldwide about 20% of total cultivated and about 33% of irrigated agricultural lands are afflicted due to soil salinity [7, 8]. Soil salinization is aggravated by the adverse effects of climate change causing abandonment globally of 0.3–1.5 million hectare year⁻¹ [9]. Consequently, an estimated 50% of the arable land would be salinized by the year 2050 [10]. The cost of global crop loss owing to salinization is estimated to be USD 27.3 billion [11]. Rice, with a threshold of 3 dSm⁻¹ for most cultivated varieties, is very sensitive to salinity [12], and rice cultivation is extremely vulnerable to soil salinization. In moderately saline areas, the yield loss of rice is 10–15%, whereas in highly saline areas, the yield was reduced by 30–45% [13]. Soil salinity has become a major limiting factor for local rice production, especially in the irrigated farming systems in the north-central and eastern plains, and in the rain-fed systems in the west coast and the Jaffna peninsula [14]. The net income of rice production is reduced up to 22% and 43% in moderate and highly saline areas, respectively in Sri Lanka [15], and therefore salinity has a significant impact on the economic sustainability of local rice production systems.

1.2 The concept of soil salinity

Saline soils have an excess accumulation of Na⁺, K⁺, Ca²⁺, Mg²⁺, HCO₃⁻, Cl⁻, NO₃⁻, SO₄⁻², and CO₃⁻² or mixtures of these ions [16], thereby affecting the normal functions of plant growth. Based on the physiochemical properties: electrical conductivity (EC), exchangeable sodium percentage (ESP), sodium adsorption ratio (SAR), and pH, saline soils can be categorized as saline, sodic (alkali), and saline-sodic soils. Optimal soil conditions for crop growth are EC <4 dSm⁻¹, pH = 6.5–7.0, SAR <13, and ESP < 13 [17]. Saline soils are characterized by EC of saturated extract >4 dSm⁻¹, pH <8.5, SAR <13%, and ESP <15% of the exchangeable cations. However, saline soils often are in normal physical conditions with good structure and permeability, and therefore with proper management measures can be used for crop cultivation. In contrast, sodic soils are low in total salts but are characterized by high exchangeable Na⁺. Sodic soils record high ESP >15%, pH > 8.5, and SAR >13%, but EC is often <4 dSm⁻¹ [18, 19]. High levels of sodium and low total salts result in dispersed soil particles, and poor physical properties and sodic soils are sticky when wet, but hard, cloddy, crusty, and nearly impermeable to water when dry. Saline-sodic soils contain large amounts of total soluble salts but greater than 15% of exchangeable Na⁺. The pH is <8.5, SAR >15%, and the EC is >4 dSm⁻¹. The physical properties of saline-sodic soils are good as long as an excess of soluble salts is present.

Soil salinity can arise due to natural processes causing primary salinity or due to manmade secondary salinity. In field conditions, low salinity viz. EC 2–4 dSm⁻¹ can arise from natural salinity and/or irrigation salinity. Species with low to moderate salt tolerance can be grown successfully in low saline soils. Moderate to high salt-tolerant plant species can be grown when the EC is between 4 and 8 dSm⁻¹ [20] usually present in water logged irrigated conditions. Under high salinity with an EC value of >9 dSm⁻¹ only halophytes can be grown [21], and therefore the choice of crops is a useful measure of salinity management. Rice expresses low tolerance and based on current guidelines rice yields decrease by 12% for every unit of (dSm⁻¹) salinity increase above 3.0 dSm⁻¹ [22, 23]. Detailed analysis of the salinity problem including associated intricate mechanisms, therefore, can help in developing effective management practices including the selection of tolerant rice varieties.

1.3 Major factors underlying salinity tolerance

Salinity tolerance in plants involves a number of traits that act in isolation (independent) or in combination (less independent) [24, 25]. The adverse effects of salinity on plant growth are generally associated with the osmotic potential of the soil solution and the high level of toxicity of sodium (and chloride for some species) that causes multiple disturbances in crop metabolism, growth, and development at the molecular, biochemical and physiological levels [26]. Plant response to salinity is expressed in two major phases: the initial, rapid osmotic phase that inhibits growth and, a later, slower ionic phase that accelerates tissue and organ senescence [27]. Salinity-induced osmotic effects reduce plant biomass and yields; however, selected ions, such as Na⁺, Cl⁻, Ca⁺², and Ba⁺, cause additional injury and crop damage [28]. Accordingly, three distinct salinity response mechanisms were described in tolerant germplasm, including osmotic stress tolerance, Na⁺ exclusion from photosynthetic and other sensitive tissues, and tissue tolerance against accumulated Na⁺ and possibly also accumulated Cl⁻ [27].

Rice is relatively tolerant to salinity at germination and late vegetative growth, compared to the early seedling stage (3-leaf stage) and reproductive stage (pollination and fertilization). Therefore, apparently, plants at different developmental stages may express one or more of the different tolerance mechanisms. However, there is a poor correlation between the tolerance mechanisms expressed at the two most salt-sensitive stages, the early seedling and reproductive stages [27]. Multiple tolerance phenotypes, associated mechanisms, and genes were identified from different salinity-tolerant germplasm at different development stages [3, 29]. Therefore, a thorough understanding of the molecular mechanisms associated with tolerance traits, and advanced technological innovations to incorporate the traits into elite varieties can accelerate breeding programs targeting saline tolerance.

This chapter aims to review the literature available on the nature and the scale of soil salinization problems in local rice farming environments. The analysis, thereby, would provide actionable insights and potential targets to reduce the genetic vulnerability of elite germplasm and to improve soil salinity management strategies in local rice farming and production environments.

2. Rice farming systems in Sri Lanka

Rice is cultivated island-wide except at elevations 2000 m above sea level, and 75% of the rice lands in Sri Lanka are located in inland valley systems. The remaining 25% are in the terraced slopes of uplands, coastal plains, associated floodplains, and alluvial plains [30]. Rice cultivation is practiced following the “northeast” and “southwest” monsoons. The northeast monsoon is from December to February in the following year, and the southwest is from May to August. The spatial distribution of rainfall brought by the monsoons defines two distinct climatic zones namely the “wet” and “dry” zones. The wet zone is in the southwestern sector of the island, whereas the dry zone is in the north, northeastern, and eastern parts of the country. The marginal areas between the two major zones are in the intermediate zone (Figure 1). The wet zone receives rain from the southwest monsoon and the amount varies from about 100 mm to over 3000 mm. In addition, from March to April wet zone receives the “first inter-monsoon” rain due to convectional influence, which can be over 250 mm to >700 mm. The dry zone receives rain from the northeast monsoon and the peak rainy season is from December to February, where the rainfall can be up to 177 mm to 1281 mm. The influence of depression and cyclone weather systems in the Bay of Bengal cause a “second inter-monsoon” starting from late September [33] or October to November [33, 34, 35]. The entire island receives in excess of 400 mm of rain from the second inter-monsoon. The two inter-monsoon periods contribute 37% of the total annual rainfall [36] and are critically important for the local crop calendars.

Figure 1.

Land extends of the different farming systems: Major irrigated, minor irrigated, and rain-fed based on data from 2011 to 2021. Lands affected due to coastal and inland salinity and the border lines of wet, dry, and intermediate zones are indicated (data obtained from [31, 32]).

The special variability in rainfall along with soil, elevation, and hydrological regimes identify diverse rice farming environments in the country. However, hydrological regimes are the principal determinant of rice production in the country, and three major rice farming systems are identified based on the source, supply, and use of surface water. These include rice lands under major and medium irrigation schemes, minor irrigation schemes, and rain-fed systems (Figure 1). Major irrigation systems are those that have a command area of >1000 ha, whereas the command area of medium schemes is between <80 ha and 1000 ha. Small tanks or minor irrigation systems are those having an irrigated command area of 80 ha or less [37]. The different rice production systems are geographically distributed (Figure 1). Rice land irrigated by major and medium irrigation schemes extends in the dry zone in an area of 730,000 ha in the north-central and the eastern parts of the island [38], where the water supply is from major tanks, rivers, and major stream diversion systems. Rice farming in major irrigation systems contributes 53% of the total annual rice production. Rice farming systems in the central dry and intermediate zones mostly depend on minor irrigation. Minor irrigation schemes operate under village tanks that are natural or man-made water reservoirs. Out of the total extent of the irrigated lands, 37% is in this category, and the minor irrigated systems contribute 20% to the total annual rice production in the country [39]. In the southwest, southern coastal region of the wet zone, and in the Jaffna peninsula, rice is cultivated under rain-fed systems. Of the total area of rice lands, 34.7% is rain-fed [31], and rain-fed systems contribute only 27% to the total rice production. All three systems, therefore, are critical in providing food supply and livelihoods for the local community.

Rice cultivation in irrigated and rain-fed systems is practiced in two major cropping seasons namely “major cultivation season” and “minor cultivation season,” coinciding with the two monsoons (Figure 2). The “major cultivation season”, which is from September to February coincides with the second inter-monsoon rain followed by the northeast monsoon. In the major season, cultivation starts in late September [33] with the onset of the second inter-monsoon, and the crop is harvested at the end of February. In the minor cultivation season cultivation starts during the end of March to mid-April with the first inter-monsoon rain, and the crop is harvested at the end of August [33]. In the major cultivation season with sufficient rainfall, rice is cultivated throughout the country. However, during the minor cultivation season, 21% of the rice lands are abandoned or used for the cultivation of other field crops due to water scarcity [31]. Rice lands under major irrigated farming systems are cultivated in two seasons, viz. major cultivation season and minor cultivation season per year. However, in minor irrigation systems, rice is cultivated in one season, the major cultivation season, with a few exceptions where irrigation is sufficient to cultivate two seasons. Wet zone rain-fed systems in the southwest are cultivated in two seasons depending on water availability, but rainfed systems in Jaffna peninsula are cultivated only during the major cultivation season [31].

Figure 2.

Rainfall patterns and cultivation seasons in rice farming systems in Sri Lanka (adapted from [33, 34, 35]).

Average annual rice production from major cultivation season, major irrigation schemes during the period 2011 to 2022, was 1,592,588 MT accounting for 54% of the total annual rice production. In the minor irrigation and rain-fed rice farming systems, the average rice production from major season over the same period was 702, 953 MT (31%) and 653, 288 MT (29%), respectively [31]. Rice production was significantly reduced in the minor cultivation season due to the limitation of water in the minor irrigation and rain-fed systems whereby the average production from the minor season during the period from 2011 to 2022 was 340,634 MT and 136,014 MT, respectively [40]. Therefore, 72% of the total production from the minor cultivation season is from rice under major irrigation schemes, and the average annual production over the period from 2011 to 2022 was 1,222,106 MT [31]. Accordingly, altogether, >70% of the national annual rice production is from major irrigated systems (Figure 3). Therefore, major season cultivation in the major irrigated fields is the most productive rice farming environment or the “mega-rice cultivation system” in the country. Although the national contributions are less from rain-fed and minor irrigated systems where smallholder farmers cultivate rice on a semi-subsistence scale, those systems are critically important for national food security.

Figure 3.

Percentage of rice production in different districts in the two cropping seasons. Bar graphs indicate total seasonal rice production from the district as a percentage of the total national production from each season based on data from 2011 to 2021 (data obtained from [31]).

3. Spatial variation of soil salinity in Sri Lanka

Soil salinity varies across the country and the highest salinity levels are reported in the north-central plains and along the coastal belt (Figure 1, Table 1) [15, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53]. Primary salinity due to natural weathering of salt-containing bedrock is restricted to isolated patches in the island. However, secondary salinity is widespread and found in inland and coastal rice farming systems. There are three main processes associated with secondary salinity in local rice farming systems viz, irrigation salinity, dryland salinity, and coastal salinity. Irrigation salinity arises in irrigated areas as a result of rising groundwater tables from excessive irrigation or due to the use of poor-quality water. In dryland non-irrigated landscapes, dryland salinity occurs due to the rise of the water table that drives soluble ions deposited by primary salinity to the soil surface. Coastal salinity is a result of the accumulation of salts in soil and water due to seawater intrusion from surface flow or seepage of seawater through sea level rise and tidal activity. The incidence of tides and dikes is the main reason for the intrusion of seawater into inland areas. Primary salinity, found in a few isolated sites such as the Nawagathegama soils series in the northwest part of the country [54], has a minimal economic impact on local agriculture; however, secondary salinity resulting from human activities causes significant economic losses in the local rice farming systems.

Type of salinity	Location	Salinity level (Electrical conductivity)	Temporal variation	Other significant findings	Reference/s
Coastal salinity	Jaffna Peninsula	Soil EC range: 0.05–34 dSm^−1	August 2019 (Peak of the dry season and fallow season of the paddy cultivation)	13% and 46% of the paddy cultivation lands are extremely saline and saline respectively,; Salinity increased by 1.4 folds in 30 years; Salinity increment-; Slightly saline area by 51%; Moderately saline area by 24%; Extremely saline area by 65%	[41, 42, 43]
	Jaffna	EC of soils collected from Tsunami affected areas high (7–19 dSm^−1) in January and low (1–5 dSm^−1) in September	High salinity in January and low in September	EC of water and soil was high just after Tsunami and decreased to low levels with time.	[44]
	Bentota river basin; Bentota DSD- the lower part of the river basin	Soil EC range-; 0.14–20 dSm^−1; 0.0112 to 0.2443μScm^−1	August 2016 to March 2017; Groundwater salinity is high in September and from January-March	Land in the Bentota River basin; Non-saline- 29.79%; Slightly saline- 40.58%; Moderately saline- 19.47%; Saline- 10.16%; The EC distribution pattern was changed with the depth of the soil layer. 40 cm and 60 cm reported high salinity levels; Land in Bentota DSD; Moderately saline- 35.6%; Slightly saline- 64.4%	[45, 46]
	Hambantota district	Soil EC range > 16 to 0 dSm^−1	October-December (Dry period) only	Non- saline- 40.4% (<2); Slightly saline- 38.0% (2–4); Moderately saline- 20.5% (4–8); Strongly saline- 1.2% (8–16); 24.8 sq.km (>16)	[47]
	Colombo; Kahapola	Soil EC (1:5); 0–2.8 dSm^−1	Higher salinity >2 dSm^−1 in the dry season; August-September	EC values greater than 0.6 dSm^−1 in 1:5 Extract.	[48]
	Puttalam- Madampe	Soil EC; 3–4 dSm^−1	Highest salinity August-September
	Mannar	Soil EC; 3–4 dSm^−1	—
	Mathara	Soil EC; 100–422 μscm^−1	Highest salinity August-September
	Mathara (Tsunami affected and unaffected areas)	Soil EC in the tsunami-affected area varied from 100 μScm^−1 to 422 μScm^−1 and groundwater salinity changed between 400 μScm^−1 to 2000 μScm^−1	—	Soil depth; soil salinity, pH, and EC variations were directly related to groundwater salinity	[49]
	Puttalum district- Arachchikattuwa	Maximum E C values; 1.9, 1.8 and 1.7 dSm^−1	Maximum EC- February, July, and August; Minimum EC- April, September, October, and November	Percentage land extent EC >4 dS/m was 10% in September and April. April (secondary cultivation season) and September (Main cultivation season) were the suitable months to establish paddy crop since EC levels were at a minimum	[50]
	Nilwala River area	Groundwater EC 38.48 μScm^−1	—	Due to sand mining	[51]
Inland salinity	Kurunegala District; Mahananeriya; Ibbagamuwa; Alawwa	Major part of the areas was <0.15 dSm^−1; In certain areas EC (1:5) rise up to 0.8 dSm^−1	—	Soil chemical parameters in the regions were suitable for paddy cultivation.	[52]
	Mahaweli river system H Irrigation scheme; Nochchiyagama; Madatugama	Soil EC range-; 1.3–7.90 dSm^−1; Soil EC range-; 1.1–8.80 dSm^−1	September to February; March to August; (The main inflow seasons from Mahaweli)	Mahaweli H area has less than 10% of the total irrigable area with significant; soil salinity problems; 40% of the farmer fields are affected by moderate salinity; The yield loss ranged from 10–15%; high and severe salinity reduced yield by about one third	[15]

Table 1.

Spatial and temporal variation in salinity in Sri Lanka.

The inland salinity is a combination of both irrigation salinity and dryland salinity and has a mosaic distribution island-wide [53]. The main processes that contribute to irrigation salinity are the rise of groundwater tables resulting from poor drainage, excess irrigation, and perennial irrigation systems, which result in shallow water tables that bring salts to the upper layers of the soil profile. The problem is aggravated due to blocked drainage canals that prevent draining of salts [55]. A case study, in the largest irrigation scheme in the country the “Mahaweli,” observed that soil salinity is a principal determinant of rice production in affected irrigated areas whereby yield loss of 10–15% is recorded in moderately saline areas; however, the yield was reduced by third in high saline areas [15]. In the segment “Mahaweli H” > 10% of the irrigable area is affected due to high salinity and an additional 40% is affected by moderate salinity providing a relative measure of the scale of the problem [15]. When salinity is >5dSm⁻¹, rice yield is reduced by more than 20% in affected compared to unaffected cultivations [56]. Inland salinity, therefore, has a significant impact on mega-rice cultivation environments, causing reduced production and direct economic losses.

Multiple studies have shown high concentrations of inorganic ions and EC >2.5 dSm⁻¹ in irrigated water in the dry zone, significantly exceeding the recommended EC range in irrigation water, which is 0.7–0.75 dSm⁻¹ [18]. Despite the high EC values, natural Na⁺ concentrations in inland soils, including soils in the mega-cultivation environments, and except in a few isolated pockets, were low [55]. A significant portion of the northern half of the country recorded soil Na⁺ concentration > 200 mg/kg, which is below the toxic limits [57]. However, high CO₃⁻², HCO₃⁻, Ca⁺², and Mg⁺² concentrations in the dry zone soils increase EC [58, 59]. Anuradhapura, Puttalam, Polonnaruwa, and Kurunegala districts record soil Mg²⁺ and Ca²⁺ concentrations as high as 730 ppm (optimum level for rice production, 120 ppm, [60] and 509–3452 ppm [61], respectively). Ca²⁺ saturation of cation exchange capacity (CEC) of soils in the dry zone is Ca 52.5% of CEC (5 cmolc/kg), whereas the recommended Ca CEC is <20% [62]. Further, high HCO₃⁻² and SO₄⁻² are reported in water in irrigation wells in north-central areas, which can contribute to high soil EC values. In such conditions, although not subjected to Na⁺ toxicity, the crops are subjected to osmotic stress owing to the high ion concentration in the root zone.

Coastal salinity has a significant impact on the rain-fed farming systems in the southwest coastal belt [48, 63] and in the Jaffna peninsula [41, 42, 43]. Due to the low coastal slope rice farming systems in the southwest coast are frequently affected by inundation and irrigation with salty water [48, 52, 64]. An area of 0.112 million ha is affected due to coastal salinity in Sri Lanka [30], where the EC of the soil extraction can exceed 4 dSm⁻¹ [65]. In affected sites, tidal waves can range between 45 cm and 60 cm during spring tide and between 10 cm and 25 cm during neap. In dry conditions, the reduction of water levels in rivers causes backflow of seawater along rivers and intrudes the rice fields with salt water, thereby saltwater intrudes into lands up to 50 cm above mean sea level [32] in dry weather. Further, blockage of drainage canals can also cause salinization in rice fields in the coastal line. The EC range in the southwest coast varied from 0.1 to 0.4 dsm⁻¹ based on soil: water 1:5 extracts. Unlike in inland salinity, soils affected due to coastal salinity contain high concentrations of Na⁺ and Cl⁻ ions [66, 67]. In the southwest coastal belt (Kaluthara to Matara), Na⁺ levels vary from 50 to 100 ppm; however, in Jaffna and on the eastern coastline the Na⁺ levels are higher than 200 ppm [53]. Detailed observations were made in the Bentota river basin, where 70% of the land is affected due to coastal salinity, ranging from 8 dSm⁻¹ to 16 dSm⁻¹ along with >10% of land affected by high salinity (>16 dSm⁻¹) [45, 46, 53]. Agricultural production loss per year due to salinity in the area is over 3.6 million USD [45]. Rice lands are gradually abandoned in the Jaffna peninsula owing to the high levels of seawater intrusion causing salinization of soils and water in wells used for irrigation. In addition, the major irrigated farming systems in Mannar, Hambantota, and Trincomalee coastal lines are also affected due to coastal salinity, where EC values can vary between 3 to 30 dSm⁻¹ [42, 47, 48, 68]. Coastal salinity has a significant impact on semi-subsistence rice farming systems in the southwest, northwest, and eastern coastal line limiting food availability and livelihoods for the local farmers, and thereby destabilizing local food systems and increase poverty.

4. Temporal variation of salinity in Sri Lanka

Soil salinity varies seasonally whereby both coastal and inland salinity is high during the dry season. During the wet season with high rainfall salts are washed off from the soils, but in dry periods, salts accumulate causing salinity. Furthermore, the impact of coastal salinity is reduced during the wet months due to high rainfall causing a pushing effect and runoff inflow toward the sea reducing sea water backflow and saline water intrusion into the land [69]. With the heavy rain, salts are washed off in the main cultivation season, but with less water availability and with high water evaporation, the minor cultivation season is more salinity prone especially toward the end of the season. The phenomena create a seasonal cycle in salinity in the irrigated and rain-fed systems whereby salinity deposited from primary and secondary sources (seawater and irrigation water) during the dry season is washed off during the wet periods (Figures 4 and 5) [48, 50, 69, 70, 71]. The amount of rainfall and water availability, therefore, are primary factors that determine soil salt concentration and the variability in rainfall causes seasonal and annual fluctuations in salinity levels.

Figure 4.

Temporal variation in salinity and rainfall in Northwestern [A], Southwestern [B], and eastern [C] coastal line of Sri Lanka. The vertical axis on the left indicates the monthly average rainfall (mm) and on the right indicates soil electrical conductivity (dSm⁻¹). Gray boxes indicate monthly rainfall (mm), whereas the line electrical conductivity (dSm⁻¹). The green boxes from left to right present crop stages of 2.5, 3-, and 3.5-months rice varieties, including seedling (susceptible), tillering/vegetative, reproductive/PI, and mature/ripening crop. The crop-salinity susceptibility window is indicated in red boxes (based on [45, 48, 50, 63]).

Figure 5.

Temporal variation of salinity and rainfall in irrigated “mega production areas” of Sri Lanka. The vertical axis on the left indicates monthly average rainfall (mm) and on the right indicates the electrical conductivity (dSm⁻¹) in irrigation water. Gray boxes indicate the rainfall, whereas the line indicates electrical conductivity (dSm⁻¹) in irrigated water. The green boxes from left to right present growth stages of crops of 2.5, 3-, and 3.5-months, including seedling (susceptible), tillering/vegetative, reproductive and PI (susceptible), and mature/ripening stages. The crop/salinity susceptibility window is indicated in red boxes (based on [69, 70, 71]).

The west coast where the rainy (March to August: first inter monsoon followed by southwest monsoon; September to February: second inter-monsoon followed by northeast monsoon) and dry periods observe bimodal patterns. Two clear salinity peaks were recognized. In time series salinity estimates of soil and water in selected sites in the rain-fed and irrigated rice lands in the west coast the highest salinity levels are recorded from August to September at the end of the minor cultivation season (EC > 24 dSm⁻¹). A second minor salinity peak was observed at the end of the major cultivation season during February to March, where the EC values in soil >20 dSm⁻¹ (Figure 4) [45, 48, 50, 63]. A similar pattern is reported in the Jaffna peninsula [72]. When salts were not washed off due to insufficient rain, the seedling crop of the both minor and major cultivation season is affected due to early season salinity. Unlike in the west coast, in the north-central and east parts of the country EC values gradually built over a period of six months during the dry season starting from the later part of the major cultivation season throughout the minor cultivation season [72] (Figure 5). Accordingly, the major irrigated rice farming systems are affected due to high salinity over the dry months from April to October, where EC value can be as high as 7 to 9 dSm⁻¹. Therefore, farmers experience transitional saline conditions with significant yearly (depending on the annual rainfall) and seasonal fluctuations in salinity level in both irrigated and rain-fed local farming systems.

5. Salt-tolerant germplasm in Sri Lanka

Sri Lanka was the first country to recommend rice varieties for salinity-prone areas whereby the first variety released was an improved land race “Pokkali” [73]. Presently, there are four elite locally improved rice varieties recommended for cultivation in salt-affected areas namely Bg369, Bg310, At401, and At354 [33]. Salt- tolerant germplasm Pokkali [74] and Nona Bokra [75] are extensively used in local salinity tolerance breeding programs. Pokkali is an immediate parent of Bg310, At401, and At354, whereas Nona Bokra is an immediate parent of Bg369. In addition, tolerant and moderately tolerant local rice germplasm that express Na⁺ exclusion at root surface, Na⁺ exclusion at root xylem parenchyma, and capacity to maintain shoot water status under saline stress were identified [76].

Pokkali and Nona Bokra are known to withstand salinity EC > 12 dSm⁻¹ from seedling to harvesting [68, 77, 78, 79] and were also reported to express high to moderate salinity tolerance throughout the crop life. Pokkali and Nona Bokra seedlings recorded salinity susceptibility scores of three compared to a score of seven in susceptible checks [80]. The concentration of NaCl inhibiting 50% of shoot growth (IC50) in 72 hours is 305 mM in Nona Bokra and 272 mM in Pokkali but significantly low in susceptible IR64 (188 mM) and IR29 (243 mM). Shoot Na⁺ accumulation after 2 hours of 200 mM salinity stress at the seedling stage was 50% in Pokkali (2629.6 ppm) compared to that in the susceptible variety IR64 (5674.3 ppm). Similarly, plant Na⁺ concentration and Na⁺/K⁺ ratio were 2.4 and 2.5 folds lower in Pokkali seedlings compared to that in the salt-sensitive, IR64 [81]. Nona Bokra and Pokkali report low shoot Na⁺ content of 5 mg/g by shoot dry weight compared to IR64 (13 mg/g) and IR29 (12 mg/g) [82]. Nona Bokra, Pokkali, and IR28 reported 0.6 mg, 4.0 mg, and 5.8 mg reduction of dry weight of a fertile spikelet at 15 dSm⁻¹, indicating the presence of tolerance mechanisms that are effective over the growth period. [83]

Pokkali and Nona Bokra express “Na⁺ exclusion,” and therefore the phenotypes are “high Na⁺ in the culm and low Na⁺ in the leaf” [76]. Evidence found on additional resistance traits such as efficient photosynthesis, adaptive defense mechanisms, osmolyte accumulation, membrane integrity, and optimum biochemical environment in the cells of Pokkali may provide whole-life resistance [84, 85, 86, 87]. Furthermore, multiple Quantitative Trait Loci (QTLs) associated with salinity tolerance were predicted in Pokkali [88, 89, 90, 91, 92], including the Saltol locus [93, 94, 95, 96, 97], which is collocated with a major QTL, SKC1 from the salt-tolerant landrace, Nona Bokra [98]. Saltol/SKC1 is a major QTL associated with low Na⁺ uptake, high K⁺ uptake, and Na⁺/K⁺ homeostasis in shoots in the seedling stage. Saltol/SKC1 was collocated with an important candidate gene OsHKT1;5, a high-affinity K transporter gene [99, 100, 101] that is expressed in xylem parenchyma cells, and retrieves Na⁺ ions from xylem sap under salinity. Furthermore, root QTLs associated with the total Na⁺ in the root (qRNTQ-1) and K⁺ concentration in root (qRKC-4) that contribute to salt tolerance were reported in a cross between Nona Bokra/Koshihikari [102]. The physiological, genetic, and genomic studies, therefore, revealed multiple tolerance mechanisms expressed at different growth stages in Pokkali, Nona Bokra, and in the local varieties derived from those parents.

Although the four recommended local varieties expressed comparable saline tolerance at specific crop growth stages, overall tolerance levels during the crop life were not comparable with that of Pokkali and Nona Bokra. Based on the percentage germination of seeds soaked at 45 dSm⁻¹ salt solution for 9 days [103], At 354 (21–30% germination) was moderately susceptible for salinity compared to Pokkali (51–70% germination); however, Bg369 was highly susceptible (>10% germination) [68, 78, 104]. Evidence from multiple studies found that based on susceptibility scores at 12 dSm⁻¹ At354 and Bg369 were tolerant and comparable with Pokkali at seedling stage [68, 76]. Nona Bokra and Pokkali seedlings survived (100% survival rate) up to 8 dSm⁻¹; however, above 4 dSm⁻¹ survival rates declined by 50% in IR28 [105]. Based on the survival percentage of plants treated at 12 dSm⁻¹ from 3 weeks to flowering, Bg369 and At354 were comparable with Pokkali [68]. Additionally, the yield reduction at 8 dSm⁻¹ based on grain yield per panicle, At354 was low compared to Pokkali and Nona Bokra (per panicle grain yield reduction was 0.72 g, 0.43 g, and 0.38 g respectively) [78]. A F5, RILs population of At354 × Bg352 (susceptible check), predicted six QTLs on chromosomes 1 and 4 from At 354, contributing to 10–16% of the total phenotypic variability. The six QTLS were collocated with 10 tolerant haplotypes of Os01g0581400, Os10g0107000, Os11g0655900, Os12g0622500, and Os12g0624200, highlighting the genetic potential of At354 as a salinity tolerant variety [106]. Therefore, it is clear that the full potential of Pokkali, Nona Bokra, and the related germplasm is yet to be explored and incorporated into local elite rice varieties.

Several reports have identified unique local rice germplasm that can be used in base broadening for salinity tolerance [68, 78, 104]. In a study, using 102 rice accessions from local germplasm including both traditional and improved varieties five susceptibility/tolerance categories were identified based on growth parameters during the initial phase (Phase I) of osmotic stress caused by reduced water uptake due to excess salts in the external soil solution at 100 mM Na⁺(Figure 6) [65]. Interestingly, in this study Pokkali was categorized as “tolerant,” whereas six phenotypically superior germplasm were categorized in the “highly tolerant” category. Multiple tolerance traits were expressed in 7% of the germplasm [76]. Interestingly, unique germplasm superior to Pokkali at different development stages were also reported. Bg406 records 70% germination under 12 dSm⁻¹ EC, expressing similar or more tolerance than Pokkali (51–70%); At402, performed equal or better at the seedling stage and reproductive stages [68]. Similarly, At303, Bg350, and Bg450 have a high tolerance compared to Pokkali at the osmotic phase of salinity stress [76] owing to their ability to maintain high relative growth rate at early stages of salinity (24 hours after final salinization at 10 dSm-1). The unique varieties namely Bw400, At402, and Bg406 expressed tolerance compatible with Pokkali at varying degrees throughout the crop life [107]. As these traits and associated alleles are distinct from Pokkali and Nona Bokra parental germplasm, these provide unique germplasm base for increasing salinity tolerance in elite rice genotypes without compromising crop vigor. Therefore, further analysis of the local germplasm will identify novel physiological and genetic tolerance mechanisms, and the associated genotypes can be used in base-broadening for salinity tolerance breeding in rice.

Figure 6.

Salinity tolerance mechanisms in 54 selected traditional and improved rice germplasms from Sri Lanka at the seedling stage (based on [76]).

6. Discussion

Secondary salinization is increasingly becoming a productivity constrain in local rice farming systems in the “mega-cultivation environments” in north central and eastern inland fields and rain-fed systems in the west, southwest and eastern coastal line. The spatial and temporal dynamics in climatic, edaphic, and hydrological regimes, and associated variations in the farming practices create complex and diverse saline stress environments in local rice farming systems. Furthermore, salinity stress coincides with the susceptible 20-day period, the 3-leaf stage and PI of the rice crop life whereby early season salinity coincides with the 3-leaf seedling stage and late season salinity coincides with the PI stage. Secondary salinization of cultivation environments, therefore, is the result of the interaction among multiple factors, including sea intrusion, sub-optimal quality of the irrigation water, injudicious use of irrigation water, suboptimal farming practices, and unavailability of local elite germplasm, and hence is a problem that operates at system level; therefore, system-level interventions are necessary to mitigate and/or to manage the impacts.

Analyzing the variability in physical and chemical field characteristics and crop growth stage enabled zoning of the saline-afflicted production environments into three major categories: (1) late-season salinity in irrigated mega-cultivation environments during the minor cultivation season, where the soil EC values can be higher in selected locations but on average ≥ 7 dSm⁻¹, (2) late season salinity in rain-fed farming systems in the west, southwest, and eastern coastal line during the minor cultivation season, where the soil EC values can be higher in selected locations but on average ≥ 20 dSm⁻¹, and (3) early season salinity in rain-fed farming systems in west, southwest. and eastern coastal line during the minor cultivation season, where the soil EC values can be higher in selected locations but on average ≥ 12 dSm⁻¹. Although, detailed studies were not available, early season salinity during the major season is reported in the intermediate and dry zone rice fields as a result of residual salts from the previous season that was not washed off due to inadequate rainfall from north-west monsoon (personal observation).

Mega-cultivation environments that contribute nearly 70% of the average annual rice production [108] is an important economic enterprise and critical for the local economy and food availability. However, the intensive irrigated farming system is destabilized as a result of secondary salinization. Irrigation practices in the absence of proper drainage management trigger accumulation of salts in the root zone negatively affecting crop productivity. Poor drainage, undulating topography, reuse of irrigation water, and mono-cropping for long periods are recognized as the main causes of salinity developments in inland areas [109]. The seasonal soil salinization and desalinization cycles are driven by the deposition of salt by irrigation and evaporation and the removal of salts from the root zone by washing off due to rain and by gradual leaching. However, imbalances in this seasonal cycle can create residual effects leading to the accumulation of salts in the system, increasing residual salinity levels, and gradual salinization of the soils. As late-season salinity is largely an irrigation-limited issue, optimizing farm irrigation management strategies and robust water governance can reduce salinity impacts significantly and enhance productivity, while minimizing long-term land degradation in rice farming systems. Standards for irrigation and irrigation water quality based on rainfall, edaphic, and crop parameters [110, 111, 112, 113] can be useful for efficient use of irrigation water and also sustainable land use. Actionable guidelines for irrigation and management of water quality are, therefore, an immediate need to prevent or minimize rapid soil salinization in mega-cultivation environments.

Furthermore, late season crop in mega-rice cultivation environments is subjected to simultaneous occurrence of multiple abiotic stresses, including water stress, high temperature, and salinity. Sterility of panicles is very common in saline-afflicted rice environments possibly as a consequence of a decreased pollen viability or receptivity of the stigmatic surface or both [114, 115, 116, 117]. Salinity stress at the reproductive stage, especially after the booting stage, cause a reduction of grain yield due to the reduced number of filled grains, total number of grains and total grain weight per panicle, 1000-grain weight, and total grain weight per plant [118, 119, 120]. Evidence found that the reduction in grain number and grain weight in salinized panicles was not only due to reduced pollen viability but also due to higher accumulation of photosynthates (sugars) in panicle branches and panicle stem coupled with reduced activity of starch synthetase in developing grains. Accumulation of Na⁺ in floral parts impedes starch synthase activity that transfers glucose to starch primer in developing grains resulting in the failure of the seed set [121]. However, early grain initiation, loss of pollen fertility, failure of pollination, spikelet death or zygotic abortion, changes in carbohydrate availability, and changes in the kernel sink potential are also the consequences of water deficit [122, 123]. The optimum temperature range for normal growth of rice is 22 to 32°C [124] and the threshold temperature of rice during anthesis is 33.7°C [125]. Under local open field conditions, pollen sterility was recorded at 31 ± 0.8°C [126]. Temperature more than 35°C under high relative humidity (near 85–90%) has an effect on the evaporative cooling of spikelet and makes complete spikelet sterility, and thus subsequent yield losses [127]. Therefore, designing mitigation measures would require a holistic approach whereby management practices and technology interventions need to be integrated effectively to avoid, overcome, or prevent the parallel occurrence of multiple abiotic stresses during the late season.

Early minor cultivation season, and occasionally in selected fields during major season, salinity can affect the seedling crops as a result of residual salinity from the previous season that is not washed off owing to insufficient rainfall or due to continuous seawater intrusion causing inundation of coastal low-lying rice lands. Early season salinity devastates seedling fields causing total loss of the crops. Transplanting aged seedlings is commonly practiced in coastal rice fields to avoid early-season salinity whereby farmers transplant 18 days old seedlings instead of 12 days and practice gap filling in case of mortality. Old seedlings with an established root and shoot system can tolerate salinity stress [128]. In addition, at manageable levels, salinity-afflicted coastal rice lands can produce rice by the cultivation of tolerant cultivars such as At354, At401, and Bw400 and by practicing recommended good agricultural practices (GAP) [113]. These include proper management of saltwater drainage canals, proper management of soil, maintenance of satisfactory fertility levels, pH and structure of soils by maximization of soil surface cover, application of organic manure, using crop rotation, minimum tillage, proper leveling, and adding gypsum (CaSO₄), which can cause leaching out of excess sodium [55].

Despite being one of the most susceptible cereal crops, rice cultivation was traditionally practiced in coastal floodplains as an effective way of land utilization in South and East Asia [129, 130], including Sri Lanka. In experimental plots, salinity of soil was controlled below 3% by growing rice. Soil desalination rate was increased from 65 to 74% by combining physical remediation measures with planting of rice [131]. The reason for this counterintuitive practice is that rice thrives well in standing water, which helps in leaching salts from the root zone to lower layers. Persistence of multiple abiotic stresses, including waterlogging in the wet season and soil and water salinity in both wet and dry seasons make managing these rain-fed production systems challenging. Therefore, salt-tolerant varieties, while increasing yields, enable effective use of saline lands. Integrated approaches based on a detailed understanding of the land potential can help in effective land diversification and crop diversification for developing sustainable land use strategies, and cultivation of tolerant rice varieties can be part of the solution.

Zoning of saline-afflicted production environments enables identifying genetic vulnerabilities, and thereby the unique targeted phenotypes (ideotypes) and locally adapted varieties. Ideotype breeding can create multi-trait genotypes to increase genetic yield potential by modifying selected individual traits to improve crop performance in saline production environments. Since salinity levels in local farming systems are transitional with significant yearly and seasonal fluctuations, maximizing resource use and profits in farms would require a clear understanding of the tradeoffs between performance of varieties under benign conditions and under stress. The vigor/ stress response tradeoff of the varieties, therefore, is an important selection criterion for varietal selections. Knowing the vigor/stress response tradeoff and the traits associated with the tradeoff enables designing potential ideotypes. Accordingly, three major ideotypes can be described. Two different ideotypes can target late-season salinity in irrigated mega-cultivation environments viz. (1) high yielding, 3 to 3.5 months varieties that are tolerant at >7 dSm-1, for intensive irrigated farming systems affected due to late season salinity (PI stage) and (2) high yielding, 2.5 to 3 months varieties for intensive irrigated farming that can avoid late season salinity. (3) the third, ideotypes targeting the rain-fed system in the west, with southwest, and the eastern coastal line will have to express tolerance throughout the crop life including seedling and PI stages. Varieties with average yield under salinity up to 12 to 20 dSm⁻¹ would be a practical selection criterion for breeding varieties for the semi-subsistence rain-fed cultivation systems in these sites.

Climate change and rising temperatures have a strong positive correlation with increasing soil salinity [132]. Sri Lanka being a tropical island is highly vulnerable to climate change whereby the average temperature is increasing at a rate of 0.01–0.03°C per year [133, 134, 135] along with significant fluctuations in rainfall resulting in frequent droughts and floods [34, 136, 137, 138]. Higher temperature intensifies the excessive deposition of salt on the surface due to evaporation and increased capillary action. Once reached the surface leaching salts below the rooting zone is extremely difficult, especially under water-limited conditions in the dry zone. Moreover, as an indirect effect of increased temperature sea level rises resulting in increased salinity encroachment in coastal and deltaic areas that have previously been favorable for rice production [138]. The sea level rise in Sri Lanka is 0.3 m by 2010 with a predicted increase of sea level up to 1 m by 2070 [139]. Natural land degradation in coastal areas through sea level rise and increased coastal salinity has changed land use patterns [140]. In the western coast, in addition to rising sea level a combined impact of multiple factors, including groundwater extraction, river damming for hydropower, and riverbed mining has caused sinking of the shorelines and drawing seawater inland.

7. Conclusion and prospects

The temporal and spatial impact of salinity signifies a wide variation in different rice farming systems in Sri Lanka. The immediate and future risk assessments can scale the increasing trends in secondary salinization of the local rice farming systems that are further aggravated by climate change. Such actionable data can help in designing efficient management strategies.

Acknowledgments

This work was carried out with the aid of a grant from UNESCO and the International Development Research Centre, Ottawa, Canada. The views expressed herein do not necessarily represent those of UNESCO, IDRC or its Board of Governors.