Environ Monit Assess (2015) 187:316 DOI 10.1007/s10661-015-4564-9
Paddy crop yield estimation in Kashmir Himalayan rice bowl using remote sensing and simulation model Mohammad Muslim & Shakil Ahmad Romshoo & A. Q. Rather
Received: 7 October 2014 / Accepted: 23 April 2015 # Springer International Publishing Switzerland 2015
Abstract The Kashmir Himalayan region of India is expected to be highly prone to the change in agricultural land use because of its geo-ecological fragility, strategic location vis-à-vis the Himalayan landscape, its transboundary river basins, and inherent socio-economic instabilities. Food security and sustainability of the region are thus greatly challenged by these impacts. The effect of future climate change, increased competition for land and water, labor from non-agricultural sectors, and increasing population adds to this complex problem. In current study, paddy rice yield at regional level was estimated using GIS-based environment policy integrated climate (GEPIC) model. The general approach of current study involved combining regional level crop database, regional soil data base, farm management data, and climatic data outputs with GEPIC model. The simulated yield showed that estimated production to be 4305.55 kg/ha (43.05 q h−1). The crop varieties like Jhelum, K-39, Chenab, China 1039, China-1007, and Shalimar rice-1 grown in plains recorded average
M. Muslim (*) Department of Ecology, Environment and Remote Sensing, SDA colony Bemina, Srinagar, Kashmir, India 190018 e-mail: [email protected] S. A. Romshoo Department of Earth Sciences, University of Kashmir, Hazratbal, Srinagar, Kashmir, India 190006 A. Q. Rather Division of Environmental Sciences, Sher-e-kashmir University of Agricultural Science and Technology, Shalimar, Kashmir, India 191121
yield of 4783.3 kg/ha (47.83 q ha−1). Meanwhile, high altitude areas with varieties like Kohsaar, K-78 (Barkat), and K -33 2 record ed yield of 41 02.2 kg /ha (41.02 q ha−1). The observed and simulated yield showed a good match with R 2 = 0.95, RMSE = 132.24 kg/ha, respectively. Keywords Food security . Sustainability . GEPIC . GIS . RMSE
Introduction Forecasting crop yields well before harvest is crucial especially in region characterized by climatic uncertainties (Mohammad et al. 2010). The prediction and real-time estimation of crop yield at provincial and national level are of great use in planning and policy making at regional and national level (Basso et al. 2007; Hayes and Decker 1996). Monitoring agricultural crop conditions during the growing season and estimating the potential crop yields are both important for the assessment of seasonal production (Doraiswamy et al. 2003). Crop models have been used to estimate crop yield potential at scales ranging from a specific field (Yang et al. 2006) to a region or country (Lobell et al. 2005; Aggarwal et al. 2006; Aggarwal and Kalra 1994). In addition to quantifying a potential yield, crop models have been used to understand the reasons for a yield gap, the difference between actual and potential yield Lobell (2013). Several studies have shown that assessment of potential yield and yield gaps can help in identifying the
316
Page 2 of 12
yield-limiting factors and in developing suitable strategies to improve the productivity of a crop (Aggarwal and Kalra 1994; Lansigan et al. 1996; Evenson et al. 1997; Naab et al. 2004). Integrating of satellite data and crop productivity models is one of the most important quantitative analysis methodologies for yield estimation at regional level (Yang et al. 2006). In case of crop yield modeling using satellite data, several studies have been undertaken to establish relationship between spectral parameters through vegetation indices and the crop yield (Singh et al. 2000). The first systematic attempt in India directed towards crop inventory through remote sensing technique was carried out under a joint ISRO-ICAR experimental project named Agricultural Resources Inventory and Survey Experiment (ARISE) during 1974–1975 and inventory and acreage under various crops were estimated (Dhanju and Shankaranarayana 1978; Sahai et al. 1977). The developments in use of remote sensing in India for crop acreage estimation have been summarized in the reviews by (Navalgund and Sahai 1985; Sahai 1985; Sahai and Dadhwal 1990; Navalgund et al. 1991; Singh et al. 1992, 2000; Singh and Goyal 1993; Sridhar et al. 1994; Singh and Ibrahim 1996; Dadhwal 1999). Crop simulation modeling (CSM) and remote sensing (RS) linkage has a number of applications in regional crop forecasting, agroecological zonation, crop suitability, and yield gap analysis (Dadhwal et al. 2003). Various researchers (Lal et al. 1993; Thornton et al. 1995; Carbone et al. 1996; Rosenthal et al. 1998) applied various crop models in regional estimation of crop yield and variability. A number of crop growth models have been developed and widely used, such as EPIC (Williams et al. 1989), DSSAT (IBSNAT 1989), WOFOST (Hijmans et al. 1994), CropSyst (Stockle et al. 1994), YIELD (Burt et al. 1981), CropWat (Clarke et al. 1998), and CENTRURY (Parton et al. 1992). Most existing crop growth models are mainly used for point or site specific applications (Priya and Shibasaki 2001; Liu et al. 2007b). Combining crop growth models with other techniques is a common way to extend the applicability of these models (Fischer et al. 2002). Applicability of these models can be extended to boarder spatial scales by combining them with a Geographic Information System (GIS). Several researchers have demonstrated the strength of linking GIS and crop simulation crop models on a farm and at regional level (Engel et al. 1999; Hartkamp et al. 1999; Thornton 1991. There have
Environ Monit Assess (2015) 187:316
been some preliminary attempts to integrate crop growth models with a GIS (Curry et al. 1990; Rao et al. 2000; Priya and Shibasaki 2001; Ines et al. 2002; Stockle et al. 2003), and these attempts mainly focus on scales no higher than national ones. Recent research has integrated the EPIC model with a GIS for global and regional scale studies (Tan and Shibasaki 2003; Liu et al. 2007a, b). Regional scale agricultural food security challenges requires accurate and detailed understanding of the management and biophysical conditions driving agricultural production systems vexed by spatial and temporal variability. Approaches to access this information varies from interpretation of real world census data to complex plant growth production models which often tend to be complex and/or computationally expensive (Jara et al. 2013). It is also not always straightforward which techniques one should use, or whether high-resolution information is even necessary for approaching certain types of impact problems (Mearns et al. 2003). However, when undertaken at a regional scale, their arise need for balanced quantification of biophysical and management drivers often available on varied scale (high and low resolution) otherwise resulting in tradeoffs and compromising on effectiveness of the results (Adam et al. 2011). Hence, an optimum scale of assessment needs to be established depending on the availability of various data sets at regional level. The present study employed mesoescale (25 km) model resolution for representing the spatial crop-yield variability in flat to complex topographic zones (CTZ) of Kashmir Himalayan region. Several works concluded on the efficient performance of mesoescale models in the complex topographic zones and flat zones (Cooley et al. 2005; Meza et al. 2008; Palusso et al. 2011). Database from mesoscale crop simulation models have been used to evaluate different management practices (irrigation protocols and dynamics of pest and diseases) (Cooley et al. 2005) and overall climate change impacts on agriculture (Jara et al. 2013; Tan and Shibasaki 2003; White et al. 2011). Till date, many of models have been applied to climate impact assessment at country or subcontinental scale to estimate potential productivity, not actual productivity, using climate data, and agricultural practices which cannot be evaluated effectively. However, there is no attempt to apply crop simulation models at regional scale under existing agricultural and management practices and on evaluation of how much accuracy can be achieved by extending the idea of GIS/ crop simulation model by integration of data, available
Environ Monit Assess (2015) 187:316
at regional scale. In Kashmir region, agriculture is the back bone of the economy. About 73 % of the population in the state derives its livelihood directly or indirectly from the agriculture sector. The proportion of area under agriculture in the state is 30.71 %, with rice as principal agricultural crops. The state faces a formidable challenge to feed the growing population, which is estimated to reach 12.4 and 14.28 million by year 2010 and 2020, respectively. The possible adverse bearing manifest directly from changes in land use, water regimes, and crop management practices on the output of various crop, including paddy which forms the chief staple crop in this Himalayan region has not yet been assessed. In this context, present study was undertaken to simulate rice yield under water non-limiting conditions in Kashmir Himalayan region using combination of satellite data and crop simulation model and demonstrate potential of the modeling application, at regional scale, by establishing the accuracy of simulated yield.
Study area The Kashmir valley is a longitudinal depression in the great northwestern complex of the Himalayan ranges. It constitutes an important relief feature of tremendous geographic significance. Carved out tectonically, the valley has a strong genetic relationship with the Himalayan complex, which exercises an all-pervading influence on its geographic entity. Territorially, it forms the interior part of the Jammu and Kashmir. The latitudinal extent of the state is 32.17° N to 37.6° N, whereas the longitudinal extent is 73.26° E to 80.30° E (Fig. 1). The valley occupies an area of 15,856 km2. The rice zone in valley corresponds to the irrigated temperate zone that varies in altitude from 1524–2100 m asl.
Materials and methods Materials The important input data for GEPIC model include crop level data, soil data, climate, and management data are discussed as follows: The regional crop level soil level database for the study was taken from the regional crop land mapping done by (Muslim et al. 2010) which showed an area of 139,363 ha. The model can accept up to 20 parameters
Page 3 of 12 316
for 10 soil layers. However, only a minimum of seven parameters is required: depth, percent sand, percent silt, bulk density, PH, percent organic carbon, and percent calcium carbonate. Other soil parameters can be estimated by EPIC itself. Therefore only these seven parameters in four layers are applied in this study. The soildepth intervals are 0–0.1, 0.1–10, 10–30, 30–50, and 50–80 cm. This study used the soil data base product generated by division of soil science, Shere Kashmir university of agricultural science and Technology, Kashmir (SKUAST-K) Shalimar for entire Kashmir division under soil health assessment project which was downloaded from the home page of SKUAST-K, Shalimar (available at: http://shak.skuastkashmir.ac.in/ SoilTest.aspx). Soil physical and chemical properties across various physiographic zones of study area are given in (Table 1). An Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) DEM having a spatial resolution of 30 m was used to incorporate the elevation information. GEPIC also requires detailed descriptions of management practices. These descriptions must specify the timing of individual operations like plantation dates, duration of growing season, LAI at various stages of the crop, fertilizer application in growing season, irrigation, pesticide, and tillage control options. In current study, a mean transplantation date of 15 June and growing season length of 145 days was given as input to model as per package to practices of paddy cultivation published by Division of Agricultural Extensions SKUAST-K, Shalimar. The current MODIS 1-km LAI-FPAR product is retrieved from the reflectance of two bands (648 and 858 nm) and on an 8-day compositing period. The product also includes extensive quality control (QC) information regarding cloud and data processing conditions. During each 8-day period, the highest-quality LAI and FPAR are selected. These data are further composited over 4 (or 3) consecutive 8day periods to produce monthly data (Tian et al. 2004). This study uses the MODIS LAI product from April to September 2010, which was downloaded from the home page of Myneni’s Climate and Vegetation research group in the Department of Geography at Boston University (available: ftp://crsa.bu.edu/pub/rmyneni/ myneniproducts/datasets/MODIS/MOD15_BU/C4/). Some other parameters of management practices including fertilizer limits are described in details in (Table 2). The daily maximum and minimum temperature data for year 2010 was procured from the Indian Meteorological Department (IMD) for six IMD stations
316
Environ Monit Assess (2015) 187:316
Page 4 of 12
Fig. 1 Satellite image of study area showing Kashmir valley Table 1 Soil physical and chemical properties across various physiographic zones of study area Physiographic regions
Soil parameters Percent sand Percent silt Bulk density pH (g/cm3)
% OC
CEC (C mol EC (dS/m) (p+)/kg)
High altitude (1850–2100 m asl, slope >5 %) 30.1– 6.5
32.5–52.5
1.32–1.46
7.4–7.9 0.63–0.97 9.8–29.3
0.20–0.31
Low altitude (1600–1850 m asl, slope, 30
20–29
8.44
20.78
12.83
26.77
T max maximum recorded temperature, T min minimum recorded temperature
316
Page 6 of 12
Environ Monit Assess (2015) 187:316
Fig. 2 The general frame work of overall methodology
statistical crop yield for the region. This makes grid-togrid comparison between the simulated and statistical yields impossible. Considering the scarcely available
Fig. 3 Distribution of validation points across study area
data, the GEPIC model was validated by extensive field observations of crop areas in each grid cell, and comparing it with the simulated yield. The field observation
Environ Monit Assess (2015) 187:316
Page 7 of 12 316
was carried through GPS and the points were plotted on the simulated grid to make grid-point comparison between simulated and observed yields. More than five sample points were collected in each grid and averaged to assign a single value to a point for comparisons (Fig. 3). The observed and simulated yield for various points across the study for 2010 is shown in (Table 4). The simulated yields and the observed yields were quite comparable as indicated by various statistical tests done on observed and simulated yields. The slope of the trend line was not significantly different from 1, while all the intercepts was not significantly different from 0. The coefficient of determination was higher than 0.6. With the R 2 value is almost 0.956, RMSE = 132.24 kg/ha (Fig. 4) the observed and simulated values showed the good agreement. The percent error for predicted yield does not exceed±5.53 % for all the six districts of the valley (Table 4). The RMSE obtained was found to be 3 % of the simulated mean yield over the study area. Adjusted R2 and root mean square deviation are usually adopted to assess the goodness-of-fit clearly indicated the close match between observed and simulated yield.
Results Estimating the crop production at regional level is always challenging due to the variety of cropping systems and levels of technology used. A number of simplifying assumptions were used to allow the complex interaction between the main environmental variables influencing crop yields to be understood. The GEPIC assumes by default raster datasets with a resolution of 0.5°×0.5°, an extent of −80 to 180° longitude and −90 to 90° latitude, and therefore 720×360 rows and columns. In current study, the model was adjusted for resolution of 0.25× 0.25°, and extent of −73.93 to 75.43° longitude and 33.47° to 34.72° latitude, and therefore 5×6 rows and columns. Crop model was simulated for physiological effects of CO2 at 390 ppm obtained from the carbon projections for 2010 from Mauna Loa Observatory in Hawaii, USA, and other management practices. The output obtained at mesoscale (25 km) model resolution representing the spatial crop-yield variability through descriptive maps is shown in Fig. 5. The average yield simulated by model for 2010 was 4305.55 kg/ha (43.05 q/ha). Significant difference was seen in yield of plains and high altitudes. The crop varieties like Jhelum, K-39, Chenab, China
Table 4 Simulated and observed yield from various points across study area 2010 S. No.
Village
Lat/long
Yield kg/ha
Percent error (±)
Simulated
Observed
1
Ganderbal
34.26/74.76
5260
5500
2
Telbal
34.18/74.18
3880
4000
5.53 2.76
3
Pathushshi
34.40/74.64
4660
4500
3.68
4
Warpora
34.37/74.47
4660
4500
3.68
5
Haigam
34.17/74.55
3980
4000
0.46
6
Mukaam
33.90/75.27
3690
3500
4.37
7
Wahipora
34.39/74.28
4170
4000
3.91
8
Kullangam
34.34/74.36
4170
4000
3.91
9
Gopalpora
33.96/74.80
5450
5500
1.15
10
HMT
34.09/74.75
4130
4000
2.99
11
Marhama
33.81/75.06
4580
4500
1.84
12
Awantipora
33.92/74.98
4580
4500
1.84
13
Magam
34.06/74.62
3980
4000
0.46
14
Kulgam
33.64/75.00
4580
4500
1.84
15
Shar
34.01/74.99
4130
4000
2.99
16
Tral
33.93/75.11
4580
4500
1.84
316
Environ Monit Assess (2015) 187:316
Page 8 of 12
Fig. 4 Scatter plot of observed and simulated yields
6000 5700 5400 5100
y = 1.093x - 474.6 R² = 0.956 RMSE= 132.24kg/ha
4800 4500
Observed
4200
Simulated
3900 3600 3300 3000 3000 3300 3600 3900 4200 4500 4800 5100 5400 5700 6000
1039, China-1007, and Shalimar rice-1 grown in plains recorded average yield of 4783.3 kg/ha (47.83 q ha−1). Meanwhile, high altitude areas with varieties like Kohsaar, K-78 (Barkat), and K-332 recorded yield of 4102.2 kg/ha (41.02 q ha−1).
Fig. 5 Simulated paddy rice yield in kg/ha for the study area
Discussions Agricultural recommendations for the sustainable management of land resources need accurate crop yield data on a global or regional scale. In present study, an attempt
Environ Monit Assess (2015) 187:316
was made to estimate the crop yield in Kashmir Himalayan region using GEPIC model. The average yield simulated by model for 2010 was 4305.55 kg/ha (43.05 qha−1). The average yield in Kashmir region varies from (35–45 qha−1) has been reported by (Bali and Uppal 1992). Significant variation in crop yield was reported from rice varieties grown in plains and high altitude regions of the valley. This can be attributed to the climatic variations, crop varieties grown, and crop management practices adopted in these physiographic regions. The crop management, and soil and climatic factors play an important role in biomass and grain yield has be verified by (Arora et al. 2006; Katsura et al. 2008; Rashid et al. 2009; Zhang et al. 2009) while working on paddy rice in other regional countries, i.e., India, Bangladesh, and China. Variation in grain yield between different varieties under different agro climatic conditions in Kashmir Himalayan region has also been reported by (Bali and Uppal 1992; Dhiman et al. 1999; Singh et al. 2000; Gosh 2001; Ganajaxi et al. 2001). Also the percent error for predicted yield was found to be greater for the high altitude regions (Gandharbal= ±5.53, Mukaam=±4.37) while those of plains the percent error was relatively less (Haigam=±0.46, Magam= ±0.46). Altitude profoundly affects the soil’s inherent fertility and behavior (Mani 1990; Bowman et al. 2002). Many soil fertility characteristics (including organic matter content, pH, cation exchange capacity, phosphate sorption, and phosphorus availability) show significant altitudinal variations (Jobbagy and Jackson 2000). The high altitude of Himalaya probably provides a unique climate (Mandal et al. 1999). In low temp nitrogen assimilation is effected that has profound effect on rice biomass production and yield of rice (Hasanuzzaman et al. 2009, 2010; Mandal et al. 1999). The effects of low temperatures on crop yield have also been well illustrated by findings Sasaki and Wada (1975) clearly indicating significant variation in yield altitudes to that of plains. The simulated crop yield showed consistency with the other works at regional and national level. Also, the validation of the simulated results with the observed data showed good performance of the model.
Conclusions In current study, GEPIC model provided a practical tool for simulating crop yield. The model facilitated the
Page 9 of 12 316
effective use of spatially distributed climatic, soil, and land use data to estimate yield for each grid with a regional coverage. Integrating crop models with spatial data at regional level improved the model performance and gave additional strength of finding the outputs at finer scale. The results showed variation in yield with crop varieties grown in plains and high altitudes with crop varieties at low altitude recording higher yield. The simulated yield corresponded reasonably with the observed data showing good performance of the model. The results showed that GEPIC model could simulate regional yield with acceptable accuracy. Thus crop growth simulation models can replace the need for years of costly multi location, on station and on farm trials to select rice varieties. As research tools, model development and application can contribute to identify gaps in our knowledge, thus enabling more efficient and targeted research planning. There is ample scope for application of the calibrated model to identify better cultivars and management practices for different rice growing environments. We note that significant limitations exist in the capacity to evaluate current agricultural productivity in Kashmir Himalayan region. This includes uncertainties in and limited access to crop production and farm management data available at the regional level. In current study crop estimates were done under non-limiting condition of water implying that with water as non-limiting factor that model yield simulation will be governed by weather, cultivar, and management practices. In current study thus solely focused on relative productivity under basic management practices to estimate crop yield. Despite some limitations, the modeling approach has evolved as the best means of not only yield estimation but to assess the effects of future global climate change, thus helping in the formulation of national policies for adaptation and mitigation purposes.
Acknowledgments I acknowledge assistance extended by the EPIC support team at Backland Research Center, Texas and Swiss Federal Institute for Aquatic Science and Technology (EAWAG), for providing the software support. I acknowledge Mr. Christian Fobes, Swiss Federal Institute for Aquatic Science and Technology (EAWAG) for technical support in down scaling the model. I also like to thank IMD Srinagar for the supply of meteorological data for the study. I acknowledge the anonymous reviewers for the time and effort devoted to review this manuscript.
316
Page 10 of 12
References Adam, M., Van Bussel, L. G. J., Leffelaar, P. A., Van Keulen, H., & Ewert, F. (2011). Effects of modelling detail on simulated potential crop yields under a wide range of climatic conditions. Ecological Modelling, 222, 131–143. Aggarwal, P. K., & Kalra, N. (1994). Simulating the effect of climatic factors, genotype and management on productivity of wheat in India (p. 156). New Delhi: Indian Agricultural Research Institute Publication. Aggarwal, P. K., Banerjee, B., Daryaei, M. G., Bhatia, A., & Bala, A. (2006). InfoCrop: a dynamic simulation model for the assessment of crop yields, losses due to pests, and environmental impact of agro-ecosystems in tropical environments. II. Performance of the model. Agricultural Systems, 89, 47– 67. Arora, V. K., Gajri, P. R., & Uppal, H. S. (2006). Puddling, irrigation and transplanting-time effects on productivity of rice-wheat system on a sandy loam soil of Punjab, India. Soil and Tillage Research, 85, 212–220. Bali, A. S., & Uppal, H. S. (1992). Effect of date of transplantation and water management practices on yield of Basmati rice (Oryza sativa). Indian Journal of Agronomy, 40, 186–192. Basso, B., McVicar, T. R., & Lee, B. (2007). Remote sensing and GIS applications in agrometeorology. In K. Stitger (Ed.), Chapter 12: Guidelines of agrometeorological practices. Geneva: World Meteorological Organization of the United Nation. Bowman, W. D., Cairns, D. M., Baron, J. S., & Seastedt, T. R. (2002). Islands in the sky: alpine and treeline ecosystems of the Rockies. In J. S. Baron (Ed.), Rocky mountain futures: an ecological perspective (pp. 183–202). Washington (DC): Island Press. Burt, J. E., Hayes, J. T., O’Rourke, P. A., Terjung, W. H., & Todhunter, P. E. (1981). A parametric crop water use model. Water Resources Research, 17, 1095–1108. Carbone, G. J., Narumalani, S., & King, M. (1996). Application of remote sensing and GIS technologies with physiological crop models. Photogrammetric Engineering and Remote Sensing, 62, 171–179. Clarke, D., Smith, M., & El-Askari, K. (1998). CropWat for windows: User guide, version 4.2. Rome: Food and Agriculture Organization of the United Nations. Cooley, H. S., Riley, W. J., Torn, M. S., & He, Y. (2005). Impact of agricultural practice on regional climate in a coupled land surface mesoscale model. Journal of Geophysical Research: Atmospheres, (1984–2012):110.D3. Curry, R. B., Peart, R. M., Jones, J. W., Boote, K. J., & Allen, L. H. (1990). Simulation as a tool for analyzing crop response to climate change. Transactions of the ASAE, 33(3), 981–990. Dadhwal, V. K. (1999). Remote sensing and GIS for agricultural crop acreage and yield estimation. International Archives of Photogrammetry and Remote Sensing, XXXII, 7-W9, (ISPRS Commission VII/ WG 2 Symposium on Application of Remote Sensing and GIS for Sustainable Development). Dehradun, India, 58–67. Dadhwal, V. K., Sehgal, V. K., Singh, R. P., & Rajak, D. R. (2003). Wheat yield modeling using satellite remote sensing with weather data: recent Indian experience. Mausum, 54, 253– 262.
Environ Monit Assess (2015) 187:316 Dhanju, M. S., & Shankaranarayana, H. S. (1978). Agricultural Resources Inventory and Survey Experiment (ARISE), ISROSAC-TR-11-78. Bangalore: ISRO. Dhiman, S. D., Nandal, D. P., & Hariom. (1999). Performance of dwarf rice (Oryza sativa) varieties under different time of plantation. Indian Journal of Agronomy, 42, 235–255. Doraiswamy, P. C., Moulin, S., Cook, P. W., & Stern, A. (2003). Crop yield assessment from remote sensing. PE&RS, 69, 665–674. Engel, T., Hoogenboom, G., Jones, J. W., & Wilkens, P. W. (1999). AEGIS/Win: a computer program for the application of crop simulation models across geographical areas. Agronomy Journal, 1997, 919–928. Evenson, R., Herdt, R., & Hossain, M. (1997). Rice research in Asia: progress and priorities. CAB International and IRRI, 418. Fischer, G., van Velthuizen, H. T., Shah, M., & Nachtergaele, F. O. (2002). Global Agro- Ecological Assessment for Agriculture in the 21st Century: Methodology and Results. IIASA Research Report RR-02–002. International Institute for Applied Systems Analysis, Laxenburg, Austria. Ganajaxi, J., Mohan kumar, H. D., Hedge, Y., & Angadi, V. V. (2001). Effect of plantation dates on N levels on the grain yield of aromatic rice genotypes under rain fed conditions. Karnataka Journal of Agricultural Sciences, 14, 758–759. Gosh, M. (2001). Performance of hybrid and high yielding varieties in Teri region of West Bengal. Journal of Interacedemica, 55, 78–581. Hartkamp, A. D., White, J. W., & Hoogenboom, G. (1999). Interfacing geographic information systems with agronomic modeling: a review. Agronomy Journal, 91, 761–772. Hasanuzzaman, M., Nahar, K., Roy, T. S., Rahman, M. L., Hossain, M. Z., & Ahmed, J. U. (2009). Tiller dynamics and dry matter production of transplanted rice as affected by plant spacing and number of seedling per hill. Academic Journal of Plant Sciences, 2, 162–168. Hasanuzzaman, M., Ahamed, K. U., Rahmatullah, N. M., Akhter, N., Nahar, K., & Rahman, M. L. (2010). Plant growth characters and productivity of wetland rice (Oryza sativa L.) as affected by application of different manures. Emirates Journal of Food and Agriculture, 22, 46–58. Hayes, M. J., & Decker, W. L. (1996). Using NOAA AVHRR data to estimate maize production in the United States Corn Belt. International Journal of Remote Sensing, 173, 189–3200. Hijmans, R. J., Guiking-Lens, I. M., & van Diepen, C. A. (1994). WOFOST 6.0. (user’s guide for the WOFOST 6.0 crop growth simulation model). Technical Document 12 DLO Winand Staring Centre, Wageningen. IBSNAT. (1989). Decision Support System for Agrotechnology Transfer V2.10 (DSSAT V2.10). Honolulu: Department of Agronomy and Soil Science. College of Tropical Agriculture and Human Resources: University of Hawaii, Hawaii. Ines, A. V. M., Gupta, A. D., & Loof, R. (2002). Application of GIS and crop growth models in estimating water productivity. Agricultural Water Management, 54(3), 205–225. Jara, V., Meza, F. J., Zaviezo, T., & Chorbadjian, R. (2013). Climate change impacts on invasive potential of pink hibiscus mealybug, Maconellicoccus hirsutus (Green), in Chile. Climatic Change, 117(1–2), 305–317.
Environ Monit Assess (2015) 187:316 Jobbagy, E. G., & Jackson, R. B. (2000). The vertical distribution of soil organic and its relation to climate and vegetation. Ecology Application, 10, 423–426. Katsura, K., Maeda, S., Lubis, I., Horie, T., Cao, W., & Shiraiwa, T. (2008). The high yield of irrigated rice in Yunnan, China ‘A cross-location analysis’. Field Crops Research, 107, 1–11. Lal, H., Hoogenboom, G., Calixte, J. P., Jones, J. W., & Beinroth, F. H. (1993). Using crop simulation models and GIS for regional productivity analyses. Transactions of ASAE, 36, 175–184. Lansigan, F. P., Bouman, B. A. M., & Aggarwal, P. K. (1996). Yield gaps in selected rice-producing areas in the Philippines. In P. K. Aggarwal, F. P. Lansigan, T. M. Thiyagarajan, & E. G. Rubia (Eds.), Towards integration of models in rice research (pp. 11–18). Wegeningen: SAARP Research Proceedings. Liu, J., Wiberg, D., Zehnder, A. J. B., & Yang, H. (2007a). Modelling the role of irrigation in winter wheat yield, crop water productivity, and production in China. Irrigation Science, 26, 21–33. Liu, J., Williams, J. R., Zehnder, A. J. B., & Yang, H. (2007b). GEPIC—modelling wheat yield and crop water productivity with high resolution on a global scale. Agricultural Systems, 94, 478–493. Lobell, D. (2013). The use of satellite data for crop yield gap analysis. Field Crops Research, 143, 56–64. Lobell, D. B., Ortiz-Monasterio, J. I., Asner, G. P., Matson, P. A., Naylor, R. L., & Falcon, W. P. (2005). Analysis of wheat yield and climatic trends in Mexico. Field Crops Research, 94, 250–256. Mandal, C., Mandal, D. K., Srinivas, C. V., Sehgal, J., & Velayutham., M. (1999). Soil climatic database for crop planning in India. Technical Bulletin No. 53. NBSS and LUP, 1014. Mani, M. S. (1990). Fundamentals of high altitude biology (2nd ed.). New Delhi: Oxford and IBM Publishing Co. Pvt. Ltd. Mearns, L. O., Carbone, G., Doherty, R. M., Tsvetsinskaya, E., Mccarl, B. A., Adams, R. M., & Mcdaniel, L. (2003). The uncertainty due to spatial scale of climate scenarios in integrated assessments: an example from U.S. agricultura. Integrated Assessment, 4, 225–235. Meza, F., Silva, D., & Vigil, H. (2008). Climate change impacts on irrigated maize in Mediterranean climates: evaluation of double cropping as an emerging adaptation. Agricultural Systems, 98, 21–30. Mohammad, H. I., Ahmad, A. R. M., & Abbas, O. M. (2010). Utilization of water budget model for early season forcasting of sorghum and optimum growing dates in Gadaref mechanized rainfed areas—Sudan. Agriculture and Biology Journal of North America, 1(4), 510–525. Monteith, J. L. (1977). Climate and the efficiency of crop production in Britain (pp. 277–294). London: Philosophical Transactions of the Royal Society. Muslim, M., Romshoo, S. A., & Rather, A. Q. (2010). Using RSProducts for Mapping and Studying Agricultural Land Use System. Proceedings of 6th JK Science Congress, University of Kashmir. Naab, J. B., Singh, P., Boote, K. J., Jones, J. W., & Marfo, K. O. (2004). Using CROPGRO-peanut model to quantify yield gaps in the Guinean Savanna zone of Ghana. Agronomy Journal, 96, 1231–1242.
Page 11 of 12 316 Navalgund, R. R., & Sahai, B. (1985). Remote sensing applications in agriculture—Indian experience and plans. In: Proceedings of 4th Asian Agricultural Symposium. Feb. 28-Mar. 1, Kyushu Tokai Univ., Kumamoto, Japan, 329–343. Navalgund, R. R., Parihar, J. S., Ajai, & NageshwarRao, P. P. (1991). Crop inventory using remotely sensed data. Current Science, 61, 162–171. Palusso, T., Kersebaum, K. C., Angulo, C., Hlavinka, P., Moriondo, M., Olesen, J. E., Patilf, R., Rugetg, F., Rumbaurc, C., Takách, J., Trnkad, M., Bindii, M., Çaldaðj, B., Ewertc, F., Ferrisee, R., Mirschelb, W., Şaylanj, L., Šiškak, B., Röttera, R., & Rötter, R. (2011). Simulation of winter wheat yield and its variability in different climates of Europe: a comparison of eight crop growth models. European Journal of Agronomy, 35, 103–114. Parton, W. J., McKeown, B., Kirchner, V., & Ojima, D. S. (1992). CENTURY users’ manual. Fort Collins: Colorado State University, NREL Publication. Priya, S., & Shibasaki, R. (2001). National spatial crop yield simulation using GIS-based crop production model. Ecological Modelling, 136(2–3), 113. Rao, M. N., Waits, D. A., & Neilsen, M. L. (2000). A GIS-based modeling approach for implementation of sustainable farm management practices. Environmental Modelling & Software, 15(8), 745–753. Rashid, M. H., Alam, M. M., Khan, M. A. H., & Ladha, J. K. (2009). Productivity and resource use of direct-(drum)-seeded and transplanted rice in puddled soils in rice-rice and ricewheat ecosystems. Field Crops Research, 113, 274–281. Rosenthal, W. D., Hammer, G. L., & Butler, D. (1998). Predicting regional grain sorghum production in Australia using spatial data and crop simulation modeling. Agricultural and Forest Meteorology, 91, 263–274. Sahai, B. (1985). Agricultural remote sensing in the Indian context. In Proceedings of US-INDIA Symposiumcum-workshop on Remote Sensing Fundamentals and Applications. Space Applications Centre, Ahmedabad, India. Sahai, B., & Dadhwal, V. K. (1990). Remote sensing in agriculture. In J. P. Verma & A. Verma (Eds.), Technology blending and agrarian prosperity (pp. 83–98). New Delhi: Malhotra Publishing House. Sahai, B., Chandrasekhar, S., Barde, N. K., & NagBhushna, S. R. (1977). Agricultural resources inventory and surveys experiment. In M. J. Rycroft & A. C. Stickland (Eds.), COSPAR space research (pp. 3–8). Oxford: Pergamon Press. Sasaki, K., & Wada, S. (1975). Effect of nitrogen, phosphoric acid and potassium apply on the incidence of sterility in rice plant. Proceedings of the Crop Science Society of Japan, 44, 250– 254. Singh, R., & Goyal, R. C. (1993). Use of remote sensing technology in crop yield estimation surveys. Project Report. New Delhi: IASRI. Singh, R., & Ibrahim, A. E. I. (1996). Use of spectral data in Markov Chain model for crop yield forecasting. Journal of the Indian Society of Remote Sensing, 24(3), 145–152. Singh, R., Goyal, R. C., Saha, S. K., & Chhikara, R. S. (1992). Use of satellite spectral data in crop yield estimation surveys. International Journal of Remote Sensing, 13(14), 2583– 2592.
316
Page 12 of 12
Singh, R., Semwal, D. P., Rai, A., & Chhikara, R. S. (2000). Small area estimation of crop yield using remote sensing satellite data. International Journal of Remote Sensing, 23(1), 49–56. Sridhar, V. N., Dadhwal, V. K., Sharma, R., Bairagi, G. D., Chaudhari, K. N., & Sharma, A. K. (1994). Wheat production forecasting for a predominantly unirrigated region in Madhya Pradesh (India). International Journal of Remote Sensing, 15, 1307–1316. Stockle, C. O., Martin, S. A., & Campbell, G. S. (1994). Cropsyst, a cropping systems simulation-model—water nitrogen budgets and crop yield. Agricultural Systems, 46(3), 335–359. Stockle, C. O., Donatelli, M., & Nelson, R. (2003). CropSyst, a cropping systems simulation model. European Journal of Agronomy, 18(3–4), 289–307. Tan, G., & Shibasaki, R. (2003). Global estimation of crop productivity and the impacts of global warming by GIS and EPIC integration. Ecological Modelling, 168(3), 357. Thornton, P. K. (1991). Application of crop simulation models in agricultural research and development in the tropics and subtropics. Paper series P-15. Muscle Shoals: International Fertilizer Development Center. Thornton, P. K., Saka, A. R., Singh, U., Kumwenda, J. D. T., Brink, J. E., & Dent, J. B. (1995). Application of a maize crop simulation model in the central region of Malawi. Experimental Agriculture, 31, 213–226.
Environ Monit Assess (2015) 187:316 Tian, Y., Dickson, R. E., Zhou, I., Zeng, X., Dai, Y., Myneni, R. B., Knyazikhin, Y., Zhang, X., Freidl, M., Yu, H., Wu, W., & Shaikh, M. (2004). Comparison of seasonal and spatial variations of leaf area index and fraction of absorbed photosynthetically active radiation from moderate resolution imaging spectroradiometer (MODIS) and common land model. Journal of Geophysical Research, 109, D01103. White, J., Hoogenboomb, G., Kimball, B., & Wall, G. (2011). Methodologies for simulating impacts of climate change on crop production. Field Crops Research, 124, 357–368. Williams, J. R., Jones, C. A., Kiniry, J. R., & Spaniel, D. A. (1989). The EPIC growth model. Transactions of the American Society of Agricultural Engineering, 32, 479–511. Yang, H., Wang, L., Abbaspour, K. C., & Zehnder, A. J. B. (2006). Virtual water trade: an assessment of water use efficiency in the international food trade. Hydrology and Earth System Sciences, 10(3), 443–454. Yoshida, S. (1977). Physiological consequences of altering plant type and maturity. International Rice Commission Newsletter, 36, 5–16. Zhang, Y., Tang, Q., Zou, Y., Li, D., Qin, J., Yang, S., Chen, L., Xia, B., & Peng, S. (2009). Yield potential and radiation use efficiency of Bsuper^ hybrid rice grown under subtropical conditions. Field Crops Research, 114, 91–98.
Paddy crop yield estimation in Kashmir Himalayan rice bowl using remote sensing and simulation model Mohammad Muslim & Shakil Ahmad Romshoo & A. Q. Rather
Received: 7 October 2014 / Accepted: 23 April 2015 # Springer International Publishing Switzerland 2015
Abstract The Kashmir Himalayan region of India is expected to be highly prone to the change in agricultural land use because of its geo-ecological fragility, strategic location vis-à-vis the Himalayan landscape, its transboundary river basins, and inherent socio-economic instabilities. Food security and sustainability of the region are thus greatly challenged by these impacts. The effect of future climate change, increased competition for land and water, labor from non-agricultural sectors, and increasing population adds to this complex problem. In current study, paddy rice yield at regional level was estimated using GIS-based environment policy integrated climate (GEPIC) model. The general approach of current study involved combining regional level crop database, regional soil data base, farm management data, and climatic data outputs with GEPIC model. The simulated yield showed that estimated production to be 4305.55 kg/ha (43.05 q h−1). The crop varieties like Jhelum, K-39, Chenab, China 1039, China-1007, and Shalimar rice-1 grown in plains recorded average
M. Muslim (*) Department of Ecology, Environment and Remote Sensing, SDA colony Bemina, Srinagar, Kashmir, India 190018 e-mail: [email protected] S. A. Romshoo Department of Earth Sciences, University of Kashmir, Hazratbal, Srinagar, Kashmir, India 190006 A. Q. Rather Division of Environmental Sciences, Sher-e-kashmir University of Agricultural Science and Technology, Shalimar, Kashmir, India 191121
yield of 4783.3 kg/ha (47.83 q ha−1). Meanwhile, high altitude areas with varieties like Kohsaar, K-78 (Barkat), and K -33 2 record ed yield of 41 02.2 kg /ha (41.02 q ha−1). The observed and simulated yield showed a good match with R 2 = 0.95, RMSE = 132.24 kg/ha, respectively. Keywords Food security . Sustainability . GEPIC . GIS . RMSE
Introduction Forecasting crop yields well before harvest is crucial especially in region characterized by climatic uncertainties (Mohammad et al. 2010). The prediction and real-time estimation of crop yield at provincial and national level are of great use in planning and policy making at regional and national level (Basso et al. 2007; Hayes and Decker 1996). Monitoring agricultural crop conditions during the growing season and estimating the potential crop yields are both important for the assessment of seasonal production (Doraiswamy et al. 2003). Crop models have been used to estimate crop yield potential at scales ranging from a specific field (Yang et al. 2006) to a region or country (Lobell et al. 2005; Aggarwal et al. 2006; Aggarwal and Kalra 1994). In addition to quantifying a potential yield, crop models have been used to understand the reasons for a yield gap, the difference between actual and potential yield Lobell (2013). Several studies have shown that assessment of potential yield and yield gaps can help in identifying the
316
Page 2 of 12
yield-limiting factors and in developing suitable strategies to improve the productivity of a crop (Aggarwal and Kalra 1994; Lansigan et al. 1996; Evenson et al. 1997; Naab et al. 2004). Integrating of satellite data and crop productivity models is one of the most important quantitative analysis methodologies for yield estimation at regional level (Yang et al. 2006). In case of crop yield modeling using satellite data, several studies have been undertaken to establish relationship between spectral parameters through vegetation indices and the crop yield (Singh et al. 2000). The first systematic attempt in India directed towards crop inventory through remote sensing technique was carried out under a joint ISRO-ICAR experimental project named Agricultural Resources Inventory and Survey Experiment (ARISE) during 1974–1975 and inventory and acreage under various crops were estimated (Dhanju and Shankaranarayana 1978; Sahai et al. 1977). The developments in use of remote sensing in India for crop acreage estimation have been summarized in the reviews by (Navalgund and Sahai 1985; Sahai 1985; Sahai and Dadhwal 1990; Navalgund et al. 1991; Singh et al. 1992, 2000; Singh and Goyal 1993; Sridhar et al. 1994; Singh and Ibrahim 1996; Dadhwal 1999). Crop simulation modeling (CSM) and remote sensing (RS) linkage has a number of applications in regional crop forecasting, agroecological zonation, crop suitability, and yield gap analysis (Dadhwal et al. 2003). Various researchers (Lal et al. 1993; Thornton et al. 1995; Carbone et al. 1996; Rosenthal et al. 1998) applied various crop models in regional estimation of crop yield and variability. A number of crop growth models have been developed and widely used, such as EPIC (Williams et al. 1989), DSSAT (IBSNAT 1989), WOFOST (Hijmans et al. 1994), CropSyst (Stockle et al. 1994), YIELD (Burt et al. 1981), CropWat (Clarke et al. 1998), and CENTRURY (Parton et al. 1992). Most existing crop growth models are mainly used for point or site specific applications (Priya and Shibasaki 2001; Liu et al. 2007b). Combining crop growth models with other techniques is a common way to extend the applicability of these models (Fischer et al. 2002). Applicability of these models can be extended to boarder spatial scales by combining them with a Geographic Information System (GIS). Several researchers have demonstrated the strength of linking GIS and crop simulation crop models on a farm and at regional level (Engel et al. 1999; Hartkamp et al. 1999; Thornton 1991. There have
Environ Monit Assess (2015) 187:316
been some preliminary attempts to integrate crop growth models with a GIS (Curry et al. 1990; Rao et al. 2000; Priya and Shibasaki 2001; Ines et al. 2002; Stockle et al. 2003), and these attempts mainly focus on scales no higher than national ones. Recent research has integrated the EPIC model with a GIS for global and regional scale studies (Tan and Shibasaki 2003; Liu et al. 2007a, b). Regional scale agricultural food security challenges requires accurate and detailed understanding of the management and biophysical conditions driving agricultural production systems vexed by spatial and temporal variability. Approaches to access this information varies from interpretation of real world census data to complex plant growth production models which often tend to be complex and/or computationally expensive (Jara et al. 2013). It is also not always straightforward which techniques one should use, or whether high-resolution information is even necessary for approaching certain types of impact problems (Mearns et al. 2003). However, when undertaken at a regional scale, their arise need for balanced quantification of biophysical and management drivers often available on varied scale (high and low resolution) otherwise resulting in tradeoffs and compromising on effectiveness of the results (Adam et al. 2011). Hence, an optimum scale of assessment needs to be established depending on the availability of various data sets at regional level. The present study employed mesoescale (25 km) model resolution for representing the spatial crop-yield variability in flat to complex topographic zones (CTZ) of Kashmir Himalayan region. Several works concluded on the efficient performance of mesoescale models in the complex topographic zones and flat zones (Cooley et al. 2005; Meza et al. 2008; Palusso et al. 2011). Database from mesoscale crop simulation models have been used to evaluate different management practices (irrigation protocols and dynamics of pest and diseases) (Cooley et al. 2005) and overall climate change impacts on agriculture (Jara et al. 2013; Tan and Shibasaki 2003; White et al. 2011). Till date, many of models have been applied to climate impact assessment at country or subcontinental scale to estimate potential productivity, not actual productivity, using climate data, and agricultural practices which cannot be evaluated effectively. However, there is no attempt to apply crop simulation models at regional scale under existing agricultural and management practices and on evaluation of how much accuracy can be achieved by extending the idea of GIS/ crop simulation model by integration of data, available
Environ Monit Assess (2015) 187:316
at regional scale. In Kashmir region, agriculture is the back bone of the economy. About 73 % of the population in the state derives its livelihood directly or indirectly from the agriculture sector. The proportion of area under agriculture in the state is 30.71 %, with rice as principal agricultural crops. The state faces a formidable challenge to feed the growing population, which is estimated to reach 12.4 and 14.28 million by year 2010 and 2020, respectively. The possible adverse bearing manifest directly from changes in land use, water regimes, and crop management practices on the output of various crop, including paddy which forms the chief staple crop in this Himalayan region has not yet been assessed. In this context, present study was undertaken to simulate rice yield under water non-limiting conditions in Kashmir Himalayan region using combination of satellite data and crop simulation model and demonstrate potential of the modeling application, at regional scale, by establishing the accuracy of simulated yield.
Study area The Kashmir valley is a longitudinal depression in the great northwestern complex of the Himalayan ranges. It constitutes an important relief feature of tremendous geographic significance. Carved out tectonically, the valley has a strong genetic relationship with the Himalayan complex, which exercises an all-pervading influence on its geographic entity. Territorially, it forms the interior part of the Jammu and Kashmir. The latitudinal extent of the state is 32.17° N to 37.6° N, whereas the longitudinal extent is 73.26° E to 80.30° E (Fig. 1). The valley occupies an area of 15,856 km2. The rice zone in valley corresponds to the irrigated temperate zone that varies in altitude from 1524–2100 m asl.
Materials and methods Materials The important input data for GEPIC model include crop level data, soil data, climate, and management data are discussed as follows: The regional crop level soil level database for the study was taken from the regional crop land mapping done by (Muslim et al. 2010) which showed an area of 139,363 ha. The model can accept up to 20 parameters
Page 3 of 12 316
for 10 soil layers. However, only a minimum of seven parameters is required: depth, percent sand, percent silt, bulk density, PH, percent organic carbon, and percent calcium carbonate. Other soil parameters can be estimated by EPIC itself. Therefore only these seven parameters in four layers are applied in this study. The soildepth intervals are 0–0.1, 0.1–10, 10–30, 30–50, and 50–80 cm. This study used the soil data base product generated by division of soil science, Shere Kashmir university of agricultural science and Technology, Kashmir (SKUAST-K) Shalimar for entire Kashmir division under soil health assessment project which was downloaded from the home page of SKUAST-K, Shalimar (available at: http://shak.skuastkashmir.ac.in/ SoilTest.aspx). Soil physical and chemical properties across various physiographic zones of study area are given in (Table 1). An Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) DEM having a spatial resolution of 30 m was used to incorporate the elevation information. GEPIC also requires detailed descriptions of management practices. These descriptions must specify the timing of individual operations like plantation dates, duration of growing season, LAI at various stages of the crop, fertilizer application in growing season, irrigation, pesticide, and tillage control options. In current study, a mean transplantation date of 15 June and growing season length of 145 days was given as input to model as per package to practices of paddy cultivation published by Division of Agricultural Extensions SKUAST-K, Shalimar. The current MODIS 1-km LAI-FPAR product is retrieved from the reflectance of two bands (648 and 858 nm) and on an 8-day compositing period. The product also includes extensive quality control (QC) information regarding cloud and data processing conditions. During each 8-day period, the highest-quality LAI and FPAR are selected. These data are further composited over 4 (or 3) consecutive 8day periods to produce monthly data (Tian et al. 2004). This study uses the MODIS LAI product from April to September 2010, which was downloaded from the home page of Myneni’s Climate and Vegetation research group in the Department of Geography at Boston University (available: ftp://crsa.bu.edu/pub/rmyneni/ myneniproducts/datasets/MODIS/MOD15_BU/C4/). Some other parameters of management practices including fertilizer limits are described in details in (Table 2). The daily maximum and minimum temperature data for year 2010 was procured from the Indian Meteorological Department (IMD) for six IMD stations
316
Environ Monit Assess (2015) 187:316
Page 4 of 12
Fig. 1 Satellite image of study area showing Kashmir valley Table 1 Soil physical and chemical properties across various physiographic zones of study area Physiographic regions
Soil parameters Percent sand Percent silt Bulk density pH (g/cm3)
% OC
CEC (C mol EC (dS/m) (p+)/kg)
High altitude (1850–2100 m asl, slope >5 %) 30.1– 6.5
32.5–52.5
1.32–1.46
7.4–7.9 0.63–0.97 9.8–29.3
0.20–0.31
Low altitude (1600–1850 m asl, slope, 30
20–29
8.44
20.78
12.83
26.77
T max maximum recorded temperature, T min minimum recorded temperature
316
Page 6 of 12
Environ Monit Assess (2015) 187:316
Fig. 2 The general frame work of overall methodology
statistical crop yield for the region. This makes grid-togrid comparison between the simulated and statistical yields impossible. Considering the scarcely available
Fig. 3 Distribution of validation points across study area
data, the GEPIC model was validated by extensive field observations of crop areas in each grid cell, and comparing it with the simulated yield. The field observation
Environ Monit Assess (2015) 187:316
Page 7 of 12 316
was carried through GPS and the points were plotted on the simulated grid to make grid-point comparison between simulated and observed yields. More than five sample points were collected in each grid and averaged to assign a single value to a point for comparisons (Fig. 3). The observed and simulated yield for various points across the study for 2010 is shown in (Table 4). The simulated yields and the observed yields were quite comparable as indicated by various statistical tests done on observed and simulated yields. The slope of the trend line was not significantly different from 1, while all the intercepts was not significantly different from 0. The coefficient of determination was higher than 0.6. With the R 2 value is almost 0.956, RMSE = 132.24 kg/ha (Fig. 4) the observed and simulated values showed the good agreement. The percent error for predicted yield does not exceed±5.53 % for all the six districts of the valley (Table 4). The RMSE obtained was found to be 3 % of the simulated mean yield over the study area. Adjusted R2 and root mean square deviation are usually adopted to assess the goodness-of-fit clearly indicated the close match between observed and simulated yield.
Results Estimating the crop production at regional level is always challenging due to the variety of cropping systems and levels of technology used. A number of simplifying assumptions were used to allow the complex interaction between the main environmental variables influencing crop yields to be understood. The GEPIC assumes by default raster datasets with a resolution of 0.5°×0.5°, an extent of −80 to 180° longitude and −90 to 90° latitude, and therefore 720×360 rows and columns. In current study, the model was adjusted for resolution of 0.25× 0.25°, and extent of −73.93 to 75.43° longitude and 33.47° to 34.72° latitude, and therefore 5×6 rows and columns. Crop model was simulated for physiological effects of CO2 at 390 ppm obtained from the carbon projections for 2010 from Mauna Loa Observatory in Hawaii, USA, and other management practices. The output obtained at mesoscale (25 km) model resolution representing the spatial crop-yield variability through descriptive maps is shown in Fig. 5. The average yield simulated by model for 2010 was 4305.55 kg/ha (43.05 q/ha). Significant difference was seen in yield of plains and high altitudes. The crop varieties like Jhelum, K-39, Chenab, China
Table 4 Simulated and observed yield from various points across study area 2010 S. No.
Village
Lat/long
Yield kg/ha
Percent error (±)
Simulated
Observed
1
Ganderbal
34.26/74.76
5260
5500
2
Telbal
34.18/74.18
3880
4000
5.53 2.76
3
Pathushshi
34.40/74.64
4660
4500
3.68
4
Warpora
34.37/74.47
4660
4500
3.68
5
Haigam
34.17/74.55
3980
4000
0.46
6
Mukaam
33.90/75.27
3690
3500
4.37
7
Wahipora
34.39/74.28
4170
4000
3.91
8
Kullangam
34.34/74.36
4170
4000
3.91
9
Gopalpora
33.96/74.80
5450
5500
1.15
10
HMT
34.09/74.75
4130
4000
2.99
11
Marhama
33.81/75.06
4580
4500
1.84
12
Awantipora
33.92/74.98
4580
4500
1.84
13
Magam
34.06/74.62
3980
4000
0.46
14
Kulgam
33.64/75.00
4580
4500
1.84
15
Shar
34.01/74.99
4130
4000
2.99
16
Tral
33.93/75.11
4580
4500
1.84
316
Environ Monit Assess (2015) 187:316
Page 8 of 12
Fig. 4 Scatter plot of observed and simulated yields
6000 5700 5400 5100
y = 1.093x - 474.6 R² = 0.956 RMSE= 132.24kg/ha
4800 4500
Observed
4200
Simulated
3900 3600 3300 3000 3000 3300 3600 3900 4200 4500 4800 5100 5400 5700 6000
1039, China-1007, and Shalimar rice-1 grown in plains recorded average yield of 4783.3 kg/ha (47.83 q ha−1). Meanwhile, high altitude areas with varieties like Kohsaar, K-78 (Barkat), and K-332 recorded yield of 4102.2 kg/ha (41.02 q ha−1).
Fig. 5 Simulated paddy rice yield in kg/ha for the study area
Discussions Agricultural recommendations for the sustainable management of land resources need accurate crop yield data on a global or regional scale. In present study, an attempt
Environ Monit Assess (2015) 187:316
was made to estimate the crop yield in Kashmir Himalayan region using GEPIC model. The average yield simulated by model for 2010 was 4305.55 kg/ha (43.05 qha−1). The average yield in Kashmir region varies from (35–45 qha−1) has been reported by (Bali and Uppal 1992). Significant variation in crop yield was reported from rice varieties grown in plains and high altitude regions of the valley. This can be attributed to the climatic variations, crop varieties grown, and crop management practices adopted in these physiographic regions. The crop management, and soil and climatic factors play an important role in biomass and grain yield has be verified by (Arora et al. 2006; Katsura et al. 2008; Rashid et al. 2009; Zhang et al. 2009) while working on paddy rice in other regional countries, i.e., India, Bangladesh, and China. Variation in grain yield between different varieties under different agro climatic conditions in Kashmir Himalayan region has also been reported by (Bali and Uppal 1992; Dhiman et al. 1999; Singh et al. 2000; Gosh 2001; Ganajaxi et al. 2001). Also the percent error for predicted yield was found to be greater for the high altitude regions (Gandharbal= ±5.53, Mukaam=±4.37) while those of plains the percent error was relatively less (Haigam=±0.46, Magam= ±0.46). Altitude profoundly affects the soil’s inherent fertility and behavior (Mani 1990; Bowman et al. 2002). Many soil fertility characteristics (including organic matter content, pH, cation exchange capacity, phosphate sorption, and phosphorus availability) show significant altitudinal variations (Jobbagy and Jackson 2000). The high altitude of Himalaya probably provides a unique climate (Mandal et al. 1999). In low temp nitrogen assimilation is effected that has profound effect on rice biomass production and yield of rice (Hasanuzzaman et al. 2009, 2010; Mandal et al. 1999). The effects of low temperatures on crop yield have also been well illustrated by findings Sasaki and Wada (1975) clearly indicating significant variation in yield altitudes to that of plains. The simulated crop yield showed consistency with the other works at regional and national level. Also, the validation of the simulated results with the observed data showed good performance of the model.
Conclusions In current study, GEPIC model provided a practical tool for simulating crop yield. The model facilitated the
Page 9 of 12 316
effective use of spatially distributed climatic, soil, and land use data to estimate yield for each grid with a regional coverage. Integrating crop models with spatial data at regional level improved the model performance and gave additional strength of finding the outputs at finer scale. The results showed variation in yield with crop varieties grown in plains and high altitudes with crop varieties at low altitude recording higher yield. The simulated yield corresponded reasonably with the observed data showing good performance of the model. The results showed that GEPIC model could simulate regional yield with acceptable accuracy. Thus crop growth simulation models can replace the need for years of costly multi location, on station and on farm trials to select rice varieties. As research tools, model development and application can contribute to identify gaps in our knowledge, thus enabling more efficient and targeted research planning. There is ample scope for application of the calibrated model to identify better cultivars and management practices for different rice growing environments. We note that significant limitations exist in the capacity to evaluate current agricultural productivity in Kashmir Himalayan region. This includes uncertainties in and limited access to crop production and farm management data available at the regional level. In current study crop estimates were done under non-limiting condition of water implying that with water as non-limiting factor that model yield simulation will be governed by weather, cultivar, and management practices. In current study thus solely focused on relative productivity under basic management practices to estimate crop yield. Despite some limitations, the modeling approach has evolved as the best means of not only yield estimation but to assess the effects of future global climate change, thus helping in the formulation of national policies for adaptation and mitigation purposes.
Acknowledgments I acknowledge assistance extended by the EPIC support team at Backland Research Center, Texas and Swiss Federal Institute for Aquatic Science and Technology (EAWAG), for providing the software support. I acknowledge Mr. Christian Fobes, Swiss Federal Institute for Aquatic Science and Technology (EAWAG) for technical support in down scaling the model. I also like to thank IMD Srinagar for the supply of meteorological data for the study. I acknowledge the anonymous reviewers for the time and effort devoted to review this manuscript.
316
Page 10 of 12
References Adam, M., Van Bussel, L. G. J., Leffelaar, P. A., Van Keulen, H., & Ewert, F. (2011). Effects of modelling detail on simulated potential crop yields under a wide range of climatic conditions. Ecological Modelling, 222, 131–143. Aggarwal, P. K., & Kalra, N. (1994). Simulating the effect of climatic factors, genotype and management on productivity of wheat in India (p. 156). New Delhi: Indian Agricultural Research Institute Publication. Aggarwal, P. K., Banerjee, B., Daryaei, M. G., Bhatia, A., & Bala, A. (2006). InfoCrop: a dynamic simulation model for the assessment of crop yields, losses due to pests, and environmental impact of agro-ecosystems in tropical environments. II. Performance of the model. Agricultural Systems, 89, 47– 67. Arora, V. K., Gajri, P. R., & Uppal, H. S. (2006). Puddling, irrigation and transplanting-time effects on productivity of rice-wheat system on a sandy loam soil of Punjab, India. Soil and Tillage Research, 85, 212–220. Bali, A. S., & Uppal, H. S. (1992). Effect of date of transplantation and water management practices on yield of Basmati rice (Oryza sativa). Indian Journal of Agronomy, 40, 186–192. Basso, B., McVicar, T. R., & Lee, B. (2007). Remote sensing and GIS applications in agrometeorology. In K. Stitger (Ed.), Chapter 12: Guidelines of agrometeorological practices. Geneva: World Meteorological Organization of the United Nation. Bowman, W. D., Cairns, D. M., Baron, J. S., & Seastedt, T. R. (2002). Islands in the sky: alpine and treeline ecosystems of the Rockies. In J. S. Baron (Ed.), Rocky mountain futures: an ecological perspective (pp. 183–202). Washington (DC): Island Press. Burt, J. E., Hayes, J. T., O’Rourke, P. A., Terjung, W. H., & Todhunter, P. E. (1981). A parametric crop water use model. Water Resources Research, 17, 1095–1108. Carbone, G. J., Narumalani, S., & King, M. (1996). Application of remote sensing and GIS technologies with physiological crop models. Photogrammetric Engineering and Remote Sensing, 62, 171–179. Clarke, D., Smith, M., & El-Askari, K. (1998). CropWat for windows: User guide, version 4.2. Rome: Food and Agriculture Organization of the United Nations. Cooley, H. S., Riley, W. J., Torn, M. S., & He, Y. (2005). Impact of agricultural practice on regional climate in a coupled land surface mesoscale model. Journal of Geophysical Research: Atmospheres, (1984–2012):110.D3. Curry, R. B., Peart, R. M., Jones, J. W., Boote, K. J., & Allen, L. H. (1990). Simulation as a tool for analyzing crop response to climate change. Transactions of the ASAE, 33(3), 981–990. Dadhwal, V. K. (1999). Remote sensing and GIS for agricultural crop acreage and yield estimation. International Archives of Photogrammetry and Remote Sensing, XXXII, 7-W9, (ISPRS Commission VII/ WG 2 Symposium on Application of Remote Sensing and GIS for Sustainable Development). Dehradun, India, 58–67. Dadhwal, V. K., Sehgal, V. K., Singh, R. P., & Rajak, D. R. (2003). Wheat yield modeling using satellite remote sensing with weather data: recent Indian experience. Mausum, 54, 253– 262.
Environ Monit Assess (2015) 187:316 Dhanju, M. S., & Shankaranarayana, H. S. (1978). Agricultural Resources Inventory and Survey Experiment (ARISE), ISROSAC-TR-11-78. Bangalore: ISRO. Dhiman, S. D., Nandal, D. P., & Hariom. (1999). Performance of dwarf rice (Oryza sativa) varieties under different time of plantation. Indian Journal of Agronomy, 42, 235–255. Doraiswamy, P. C., Moulin, S., Cook, P. W., & Stern, A. (2003). Crop yield assessment from remote sensing. PE&RS, 69, 665–674. Engel, T., Hoogenboom, G., Jones, J. W., & Wilkens, P. W. (1999). AEGIS/Win: a computer program for the application of crop simulation models across geographical areas. Agronomy Journal, 1997, 919–928. Evenson, R., Herdt, R., & Hossain, M. (1997). Rice research in Asia: progress and priorities. CAB International and IRRI, 418. Fischer, G., van Velthuizen, H. T., Shah, M., & Nachtergaele, F. O. (2002). Global Agro- Ecological Assessment for Agriculture in the 21st Century: Methodology and Results. IIASA Research Report RR-02–002. International Institute for Applied Systems Analysis, Laxenburg, Austria. Ganajaxi, J., Mohan kumar, H. D., Hedge, Y., & Angadi, V. V. (2001). Effect of plantation dates on N levels on the grain yield of aromatic rice genotypes under rain fed conditions. Karnataka Journal of Agricultural Sciences, 14, 758–759. Gosh, M. (2001). Performance of hybrid and high yielding varieties in Teri region of West Bengal. Journal of Interacedemica, 55, 78–581. Hartkamp, A. D., White, J. W., & Hoogenboom, G. (1999). Interfacing geographic information systems with agronomic modeling: a review. Agronomy Journal, 91, 761–772. Hasanuzzaman, M., Nahar, K., Roy, T. S., Rahman, M. L., Hossain, M. Z., & Ahmed, J. U. (2009). Tiller dynamics and dry matter production of transplanted rice as affected by plant spacing and number of seedling per hill. Academic Journal of Plant Sciences, 2, 162–168. Hasanuzzaman, M., Ahamed, K. U., Rahmatullah, N. M., Akhter, N., Nahar, K., & Rahman, M. L. (2010). Plant growth characters and productivity of wetland rice (Oryza sativa L.) as affected by application of different manures. Emirates Journal of Food and Agriculture, 22, 46–58. Hayes, M. J., & Decker, W. L. (1996). Using NOAA AVHRR data to estimate maize production in the United States Corn Belt. International Journal of Remote Sensing, 173, 189–3200. Hijmans, R. J., Guiking-Lens, I. M., & van Diepen, C. A. (1994). WOFOST 6.0. (user’s guide for the WOFOST 6.0 crop growth simulation model). Technical Document 12 DLO Winand Staring Centre, Wageningen. IBSNAT. (1989). Decision Support System for Agrotechnology Transfer V2.10 (DSSAT V2.10). Honolulu: Department of Agronomy and Soil Science. College of Tropical Agriculture and Human Resources: University of Hawaii, Hawaii. Ines, A. V. M., Gupta, A. D., & Loof, R. (2002). Application of GIS and crop growth models in estimating water productivity. Agricultural Water Management, 54(3), 205–225. Jara, V., Meza, F. J., Zaviezo, T., & Chorbadjian, R. (2013). Climate change impacts on invasive potential of pink hibiscus mealybug, Maconellicoccus hirsutus (Green), in Chile. Climatic Change, 117(1–2), 305–317.
Environ Monit Assess (2015) 187:316 Jobbagy, E. G., & Jackson, R. B. (2000). The vertical distribution of soil organic and its relation to climate and vegetation. Ecology Application, 10, 423–426. Katsura, K., Maeda, S., Lubis, I., Horie, T., Cao, W., & Shiraiwa, T. (2008). The high yield of irrigated rice in Yunnan, China ‘A cross-location analysis’. Field Crops Research, 107, 1–11. Lal, H., Hoogenboom, G., Calixte, J. P., Jones, J. W., & Beinroth, F. H. (1993). Using crop simulation models and GIS for regional productivity analyses. Transactions of ASAE, 36, 175–184. Lansigan, F. P., Bouman, B. A. M., & Aggarwal, P. K. (1996). Yield gaps in selected rice-producing areas in the Philippines. In P. K. Aggarwal, F. P. Lansigan, T. M. Thiyagarajan, & E. G. Rubia (Eds.), Towards integration of models in rice research (pp. 11–18). Wegeningen: SAARP Research Proceedings. Liu, J., Wiberg, D., Zehnder, A. J. B., & Yang, H. (2007a). Modelling the role of irrigation in winter wheat yield, crop water productivity, and production in China. Irrigation Science, 26, 21–33. Liu, J., Williams, J. R., Zehnder, A. J. B., & Yang, H. (2007b). GEPIC—modelling wheat yield and crop water productivity with high resolution on a global scale. Agricultural Systems, 94, 478–493. Lobell, D. (2013). The use of satellite data for crop yield gap analysis. Field Crops Research, 143, 56–64. Lobell, D. B., Ortiz-Monasterio, J. I., Asner, G. P., Matson, P. A., Naylor, R. L., & Falcon, W. P. (2005). Analysis of wheat yield and climatic trends in Mexico. Field Crops Research, 94, 250–256. Mandal, C., Mandal, D. K., Srinivas, C. V., Sehgal, J., & Velayutham., M. (1999). Soil climatic database for crop planning in India. Technical Bulletin No. 53. NBSS and LUP, 1014. Mani, M. S. (1990). Fundamentals of high altitude biology (2nd ed.). New Delhi: Oxford and IBM Publishing Co. Pvt. Ltd. Mearns, L. O., Carbone, G., Doherty, R. M., Tsvetsinskaya, E., Mccarl, B. A., Adams, R. M., & Mcdaniel, L. (2003). The uncertainty due to spatial scale of climate scenarios in integrated assessments: an example from U.S. agricultura. Integrated Assessment, 4, 225–235. Meza, F., Silva, D., & Vigil, H. (2008). Climate change impacts on irrigated maize in Mediterranean climates: evaluation of double cropping as an emerging adaptation. Agricultural Systems, 98, 21–30. Mohammad, H. I., Ahmad, A. R. M., & Abbas, O. M. (2010). Utilization of water budget model for early season forcasting of sorghum and optimum growing dates in Gadaref mechanized rainfed areas—Sudan. Agriculture and Biology Journal of North America, 1(4), 510–525. Monteith, J. L. (1977). Climate and the efficiency of crop production in Britain (pp. 277–294). London: Philosophical Transactions of the Royal Society. Muslim, M., Romshoo, S. A., & Rather, A. Q. (2010). Using RSProducts for Mapping and Studying Agricultural Land Use System. Proceedings of 6th JK Science Congress, University of Kashmir. Naab, J. B., Singh, P., Boote, K. J., Jones, J. W., & Marfo, K. O. (2004). Using CROPGRO-peanut model to quantify yield gaps in the Guinean Savanna zone of Ghana. Agronomy Journal, 96, 1231–1242.
Page 11 of 12 316 Navalgund, R. R., & Sahai, B. (1985). Remote sensing applications in agriculture—Indian experience and plans. In: Proceedings of 4th Asian Agricultural Symposium. Feb. 28-Mar. 1, Kyushu Tokai Univ., Kumamoto, Japan, 329–343. Navalgund, R. R., Parihar, J. S., Ajai, & NageshwarRao, P. P. (1991). Crop inventory using remotely sensed data. Current Science, 61, 162–171. Palusso, T., Kersebaum, K. C., Angulo, C., Hlavinka, P., Moriondo, M., Olesen, J. E., Patilf, R., Rugetg, F., Rumbaurc, C., Takách, J., Trnkad, M., Bindii, M., Çaldaðj, B., Ewertc, F., Ferrisee, R., Mirschelb, W., Şaylanj, L., Šiškak, B., Röttera, R., & Rötter, R. (2011). Simulation of winter wheat yield and its variability in different climates of Europe: a comparison of eight crop growth models. European Journal of Agronomy, 35, 103–114. Parton, W. J., McKeown, B., Kirchner, V., & Ojima, D. S. (1992). CENTURY users’ manual. Fort Collins: Colorado State University, NREL Publication. Priya, S., & Shibasaki, R. (2001). National spatial crop yield simulation using GIS-based crop production model. Ecological Modelling, 136(2–3), 113. Rao, M. N., Waits, D. A., & Neilsen, M. L. (2000). A GIS-based modeling approach for implementation of sustainable farm management practices. Environmental Modelling & Software, 15(8), 745–753. Rashid, M. H., Alam, M. M., Khan, M. A. H., & Ladha, J. K. (2009). Productivity and resource use of direct-(drum)-seeded and transplanted rice in puddled soils in rice-rice and ricewheat ecosystems. Field Crops Research, 113, 274–281. Rosenthal, W. D., Hammer, G. L., & Butler, D. (1998). Predicting regional grain sorghum production in Australia using spatial data and crop simulation modeling. Agricultural and Forest Meteorology, 91, 263–274. Sahai, B. (1985). Agricultural remote sensing in the Indian context. In Proceedings of US-INDIA Symposiumcum-workshop on Remote Sensing Fundamentals and Applications. Space Applications Centre, Ahmedabad, India. Sahai, B., & Dadhwal, V. K. (1990). Remote sensing in agriculture. In J. P. Verma & A. Verma (Eds.), Technology blending and agrarian prosperity (pp. 83–98). New Delhi: Malhotra Publishing House. Sahai, B., Chandrasekhar, S., Barde, N. K., & NagBhushna, S. R. (1977). Agricultural resources inventory and surveys experiment. In M. J. Rycroft & A. C. Stickland (Eds.), COSPAR space research (pp. 3–8). Oxford: Pergamon Press. Sasaki, K., & Wada, S. (1975). Effect of nitrogen, phosphoric acid and potassium apply on the incidence of sterility in rice plant. Proceedings of the Crop Science Society of Japan, 44, 250– 254. Singh, R., & Goyal, R. C. (1993). Use of remote sensing technology in crop yield estimation surveys. Project Report. New Delhi: IASRI. Singh, R., & Ibrahim, A. E. I. (1996). Use of spectral data in Markov Chain model for crop yield forecasting. Journal of the Indian Society of Remote Sensing, 24(3), 145–152. Singh, R., Goyal, R. C., Saha, S. K., & Chhikara, R. S. (1992). Use of satellite spectral data in crop yield estimation surveys. International Journal of Remote Sensing, 13(14), 2583– 2592.
316
Page 12 of 12
Singh, R., Semwal, D. P., Rai, A., & Chhikara, R. S. (2000). Small area estimation of crop yield using remote sensing satellite data. International Journal of Remote Sensing, 23(1), 49–56. Sridhar, V. N., Dadhwal, V. K., Sharma, R., Bairagi, G. D., Chaudhari, K. N., & Sharma, A. K. (1994). Wheat production forecasting for a predominantly unirrigated region in Madhya Pradesh (India). International Journal of Remote Sensing, 15, 1307–1316. Stockle, C. O., Martin, S. A., & Campbell, G. S. (1994). Cropsyst, a cropping systems simulation-model—water nitrogen budgets and crop yield. Agricultural Systems, 46(3), 335–359. Stockle, C. O., Donatelli, M., & Nelson, R. (2003). CropSyst, a cropping systems simulation model. European Journal of Agronomy, 18(3–4), 289–307. Tan, G., & Shibasaki, R. (2003). Global estimation of crop productivity and the impacts of global warming by GIS and EPIC integration. Ecological Modelling, 168(3), 357. Thornton, P. K. (1991). Application of crop simulation models in agricultural research and development in the tropics and subtropics. Paper series P-15. Muscle Shoals: International Fertilizer Development Center. Thornton, P. K., Saka, A. R., Singh, U., Kumwenda, J. D. T., Brink, J. E., & Dent, J. B. (1995). Application of a maize crop simulation model in the central region of Malawi. Experimental Agriculture, 31, 213–226.
Environ Monit Assess (2015) 187:316 Tian, Y., Dickson, R. E., Zhou, I., Zeng, X., Dai, Y., Myneni, R. B., Knyazikhin, Y., Zhang, X., Freidl, M., Yu, H., Wu, W., & Shaikh, M. (2004). Comparison of seasonal and spatial variations of leaf area index and fraction of absorbed photosynthetically active radiation from moderate resolution imaging spectroradiometer (MODIS) and common land model. Journal of Geophysical Research, 109, D01103. White, J., Hoogenboomb, G., Kimball, B., & Wall, G. (2011). Methodologies for simulating impacts of climate change on crop production. Field Crops Research, 124, 357–368. Williams, J. R., Jones, C. A., Kiniry, J. R., & Spaniel, D. A. (1989). The EPIC growth model. Transactions of the American Society of Agricultural Engineering, 32, 479–511. Yang, H., Wang, L., Abbaspour, K. C., & Zehnder, A. J. B. (2006). Virtual water trade: an assessment of water use efficiency in the international food trade. Hydrology and Earth System Sciences, 10(3), 443–454. Yoshida, S. (1977). Physiological consequences of altering plant type and maturity. International Rice Commission Newsletter, 36, 5–16. Zhang, Y., Tang, Q., Zou, Y., Li, D., Qin, J., Yang, S., Chen, L., Xia, B., & Peng, S. (2009). Yield potential and radiation use efficiency of Bsuper^ hybrid rice grown under subtropical conditions. Field Crops Research, 114, 91–98.
