Estimation of Net Rice Production through Improved CASA Model by Addition of Soil Suitability Constant ( ħα )

Net primary production (NPP) is an important indicator of the supply of food and wood. We used a hierarchy model and real time field observations to estimate NPP using satellite imagery. Net radiation received by rice crop canopies was estimated as 27,428 Wm−2 (215.4 Wm−2 as averaged) throughout the rice cultivation period (RCP), including 23,168 Wm−2 (118.3 Wm−2 as averaged) as shortwave and 4260 Wm−2 (34.63 Wm−2 as averaged) as longwave radiation. Soil, sensible and latent heat  fluxes were approximated as 3324 Wm−2, 16,549 Wm−2, and 7554 Wm−2,  respectively. Water stress on rice crops varied between 0.5838 and 0.1218 from the start until the end of the RCP. Biomass generation declined from 6.09–1.03 g/m2 in the tillering and ripening stages, respectively. We added a soil suitability constant (ħα) into the Carnegie‐Ames‐Stanford Approach (CASA) model to achieve a more  precise  estimate  of  yield.  Classification  results  suggest  that  the  total  area  under  rice cultivation was 8861 km2. The spatial distribution of rice cultivation as per suitability zone was: 1674 km2 was not suitable (NS), 592 km2 was less suitable (LS), 2210 km2 was moderately suitable (MS) and 4385 km2 was highly suitable (HS) soil type with ħα ranges of 0.05–0.25, 0.4–0.6, 0.7–0.75 and 0.85–0.95 of the CASA based yield, respectively. We estimated net production as 1.63 million tons, as  per  0.46  ton/ha,  1.2  ton/ha  1.9  ton/ha  and  2.4  ton/ha  from NS,  LS, MS  and HS  soil  types, respectively. The results obtained through this improved CASA model, by addition of the constant ħα, are likely to be useful for agronomists by providing more accurate estimates of NPP.


Introduction
About three billion of the world's population consume rice (Oryza Sativa) as an essential food/crop [1].Almost 88% of global rice production is obtained from Asia [2].The seven largest rice producing countries in the world are China (210.1 million tons), India (165.3 million tons), Indonesia (74.2 million tons), Bangladesh (53.1 million tons), Vietnam (44 million tons), Thailand (33.3 million tons) and Myanmar (28.3 million tons) in 2016-2017, which together accounts for 680.3 million tons [3], whereas in Pakistan, annual rice production is only 6 million tons.The global rice production is insufficient to cater for the needs of the community [4].Increasing population in rice consuming countries [5] and climate change in recent decades [6,7], has put massive pressure on rice producers to increase per-hectare yield.Therefore, it is important to ensure rice sustainability through the use of state-of-the-art agricultural practices.Precise estimation of NPP has become a great challenge for agronomists and economist to evaluate the exact quantity to import/export in case of shortfall/surplus.NPP is a significant indicator that gives accurate estimates for food and wood production [8].NPP is largely influenced by many controlling factors including climate, microbial

Introduction
About three billion of the world's population consume rice (Oryza Sativa) as an essential food/crop [1].Almost 88% of global rice production is obtained from Asia [2].The seven largest rice producing countries in the world are China (210.1 million tons), India (165.3 million tons), Indonesia (74.2 million tons), Bangladesh (53.1 million tons), Vietnam (44 million tons), Thailand (33.3 million tons) and Myanmar (28.3 million tons) in 2016-2017, which together accounts for 680.3 million tons [3], whereas in Pakistan, annual rice production is only 6 million tons.The global rice production is insufficient to cater for the needs of the community [4].Increasing population in rice consuming countries [5] and climate change in recent decades [6,7], has put massive pressure on rice producers to increase per-hectare yield.Therefore, it is important to ensure rice sustainability through the use of state-of-the-art agricultural practices.Precise estimation of NPP has become a great challenge for agronomists and economist to evaluate the exact quantity to import/export in case of shortfall/surplus.NPP is a significant indicator that gives accurate estimates for food and wood production [8].NPP is largely influenced by many controlling factors including climate, microbial

Introduction
About three billion of the world's population consume rice (Oryza Sativa) food/crop [1].Almost 88% of global rice production is obtained from Asia [2].The se producing countries in the world are China (210.1 million tons), India (165.3 million (74.2 million tons), Bangladesh (53.1 million tons), Vietnam (44 million tons), Thaila tons) and Myanmar (28.3 million tons) in 2016-2017, which together accounts for 68 [3], whereas in Pakistan, annual rice production is only 6 million tons.The global ri insufficient to cater for the needs of the community [4].Increasing population in countries [5] and climate change in recent decades [6,7], has put massive pressure on to increase per-hectare yield.Therefore, it is important to ensure rice sustainability of state-of-the-art agricultural practices.Precise estimation of NPP has become a gre agronomists and economist to evaluate the exact quantity to import/expo shortfall/surplus.NPP is a significant indicator that gives accurate estimates for production [8].NPP is largely influenced by many controlling factors including cli α ranges of 0.05-0.25,0.4-0.6,0.7-0.75 and 0.85-0.95 of the CASA based yield, respectively.We estimated net production as 1.63 million tons, as per 0.46 ton/ha, 1.2 ton/ha 1.9 ton/ha and 2.4 ton/ha from NS, LS, MS and HS soil types, respectively.The results obtained through this improved CASA model, by addition of the constant

ation of Net Rice Production through Improved A Model by Addition of Soil Suitability Constant hammad Hassan Raza * and Syed Amer Mahmood
Sensing group, Department of Space Science, University of the Punjab, Quaid-e-Azam Campus, 4590, Pakistan; amerpakistan@gmail.comspondence: smhn72@gmail.com;Tel.: +92-300-888-3131 : 16  , respectively.Water n rice crops varied between 0.5838 and 0.1218 from the start until the end of the RCP.Biomass ion declined from 6.09-1.03g/m 2 in the tillering and ripening stages, respectively.We added itability constant (ħα) into the Carnegie-Ames-Stanford Approach (CASA) model to achieve precise estimate of yield.Classification results suggest that the total area under rice ion was 8861 km 2 .The spatial distribution of rice cultivation as per suitability zone was: 1674 s not suitable (NS), 592 km 2 was less suitable (LS), 2210 km 2 was moderately suitable (MS) 5 km 2 was highly suitable (HS) soil type with ħα ranges of 0.05-0.25,0.4-0.6,0.7-0.75 and 5 of the CASA based yield, respectively.We estimated net production as 1.63 million tons, 0.46 ton/ha, 1.2 ton/ha 1.9 ton/ha and 2.4 ton/ha from NS, LS, MS and HS soil types, vely.The results obtained through this improved CASA model, by addition of the constant likely to be useful for agronomists by providing more accurate estimates of NPP.rds: net primary production; biomass estimation; photosynthesis; respiration; rice ion period; net radiation uction ut three billion of the world's population consume rice (Oryza Sativa) as an essential p [1].Almost 88% of global rice production is obtained from Asia [2].The seven largest rice g countries in the world are China (210.1 million tons), India (165.

Introduction
About three billion of the world's population consume rice (Oryza Sativa) as an essential food/crop [1].Almost 88% of global rice production is obtained from Asia [2].The seven largest rice producing countries in the world are China (210.1 million tons), India (165.3 million tons), Indonesia (74.2 million tons), Bangladesh (53.1 million tons), Vietnam (44 million tons), Thailand (33.3 million tons) and Myanmar (28.3 million tons) in 2016-2017, which together accounts for 680.3 million tons [3], whereas in Pakistan, annual rice production is only 6 million tons.The global rice production is insufficient to cater for the needs of the community [4].Increasing population in rice consuming countries [5] and climate change in recent decades [6,7], has put massive pressure on rice producers to increase per-hectare yield.Therefore, it is important to ensure rice sustainability through the use of state-of-the-art agricultural practices.Precise estimation of NPP has become a great challenge for agronomists and economist to evaluate the exact quantity to import/export in case of shortfall/surplus.NPP is a significant indicator that gives accurate estimates for food and wood production [8].NPP is largely influenced by many controlling factors including climate, microbial characteristics of soil, topographic disturbances and the anthropogenic activities [9].Photosynthesis [10] and respiration [11] are the key processes of plant production that are indebted to sunlight.The impacts of sunlight on NPP have been analyzed in various ecosystems [8,12,13].These analyses are important to achieve detailed description of productivity.
Terrestrial NPP estimates are used to distinguish various biomes.Sufficient knowledge of NPP is required to understand carbon related processes e.g., carbon emissions and sequestrations [14,15].NPP estimations indicate the removal of carbon dioxide (CO 2 ) from the air that is stored in the plant body to support the growth of both the short-lived (foliage roots) and long-lived tissues [16].NPP estimates are deemed to evaluate ecological disturbances on a regional through to global scale [9,17,18].Various research has been conducted in recent decades to determine significant factors affecting NPP [19,20], such as climatic factors, including relative humidity, pressure and temperature [21,22].NPP estimates are difficult to compute on a large scale with high accuracy, thus, approximated results are calibrated with real time field observations to precisely investigate spatiotemporal variations in agroindustry [23].The first method to estimate NPP was the Miami model which generated a regression correlation between various productivity levels with averaged temperature and precipitation without accounting for the influence of other climatic factors [24].The evolution of remote sensing techniques opened avenues to develop a variety of models based on satellite data to simulate NPP of terrestrial ecosystems [9,[25][26][27].One of these models is the Carnegie-Ames-Stanford Approach (CASA) model that is based on light use efficiency (LUE) which relates NPP to vegetation characteristics and incorporates climatic variables.Several researchers have successfully implemented the CASA model in South America, North America, Australia, Africa and Eurasia [9,17,28,29].CASA based studies conducted in China, presented a clear picture of an increase in agroindustry and its impact on global climate change [17,30,31].However, some limitations exist in the CASA model e.g., maximum LUE was fixed as 0.389 gc/MJ for all types of vegetation [9,26], which results in high variations between actual and estimated NPP values [32].Another limitation in the CASA model was the computation of soil moisture content, which relates to soil parameters such as soil type, the water holding capacity of soil, field moisture capacity, etc.These parameters have high spatial diversity [26], therefore, it is a difficult job to obtain reliable values for soil parameters.
Generally, three types of NPP models are considered to be important: (1) the climatic model; (2) the process model; and (3) the LUE model [33].The climatic model establishes a statistical relationship between NPP and climatic data.This model does not consider the actual vegetation type, so large variations exist between actual and estimated NPP calculations.
In the process model, NPP estimates are based on ecological and physiological processes where various parameters are based on phenology such as respiration, photosynthesis and dry matter partition [34].These parameters are difficult to estimate accurately under variable environmental conditions, and there is high uncertainty between ground validated and estimated NPP [33].
The LUE model has been extensively used to derive maximal LUE due to its handy mechanism along with its requirements for simple ecological and psychological parameters and easy combination with remotely sensed data [9].The LUE model describes how a fraction of solar energy is used by plants for respiration and the remaining energy is fixed as net production [35].The net production energy is divided among the plant organs and finally transported to the environment by various channels [26].This transfer function not only creates a balance between incoming and outgoing energy but also contributes to the dynamic behavior of the ecosystem [36].This model incorporates the climatic and physiological aspects of vegetation, therefore, NPP estimates are more accurate.
This research demonstrates a hierarchy model for biomass estimation by incorporating local ecology.The objective is to determine the impact of solar radiation, water stress, LUE and various heat fluxes, including soil heat, sensible heat and latent heat, on net productivity., respectively.Water on rice crops varied between 0.5838 and 0.1218 from the start until the end of the RCP.Biomass tion declined from 6.09-1.03g/m 2 in the tillering and ripening stages, respectively.We added uitability constant (ħα) into the Carnegie-Ames-Stanford Approach (CASA) model to achieve e precise estimate of yield.Classification results suggest that the total area under rice tion was 8861 km 2 .The spatial distribution of rice cultivation as per suitability zone was: 1674 as not suitable (NS), 592 km 2 was less suitable (LS), 2210 km 2 was moderately suitable (MS) 85 km 2 was highly suitable (HS) soil type with ħα ranges of 0.05-0.25,0.4-0.6,0.7-0.75 and .95 of the CASA based yield, respectively.We estimated net production as 1.63 million tons, 0.46 ton/ha, 1.2 ton/ha 1.9 ton/ha and 2.4 ton/ha from NS, LS, MS and HS soil types, tively.The results obtained through this improved CASA model, by addition of the constant e likely to be useful for agronomists by providing more accurate estimates of NPP.
The research design determines a sequential study of various parameters/processes affecting rice productivity throughout the cultivation period.Real time climatic data, as recorded by local metrological offices (LMOs) on a daily basis, assisted in the computation of various fluxes e.g., net radiation flux, soil heat flux, sensible heat flux and latent heat flux as these directly affect the net rice production.Water stress was estimated to evaluate soil wetness conditions throughout the RCP.Light used by the rice crop canopy was appraised to evaluate growth related process e.g., photosynthesis and respiration.Variations in light reflectance by the rice crop canopy were determined by analyzing temporal differences in normalized difference vegetation index (NDVI) values.These NDVI values assisted in the estimation of the biomass generated each day and hence the NPP.The estimated NPP was compared with actual NPP (computed using yield values as reported by local farmers), which helped to introduce a soil suitability constant (

Introduction
About three billion of the world's population consume rice (Oryza Sativa) as an essenti food/crop [1].Almost 88% of global rice production is obtained from Asia [2].The seven largest ric producing countries in the world are China (210.1 million tons), India (165.3 million tons), Indones (74.2 million tons), Bangladesh (53.1 million tons), Vietnam (44 million tons), Thailand (33.3 millio tons) and Myanmar (28.3 million tons) in 2016-2017, which together accounts for 680.3 million ton [3], whereas in Pakistan, annual rice production is only 6 million tons.The global rice production insufficient to cater for the needs of the community [4].Increasing population in rice consumin countries [5] and climate change in recent decades [6,7], has put massive pressure on rice produce to increase per-hectare yield.Therefore, it is important to ensure rice sustainability through the us of state-of-the-art agricultural practices.Precise estimation of NPP has become a great challenge fo agronomists and economist to evaluate the exact quantity to import/export in case o shortfall/surplus.NPP is a significant indicator that gives accurate estimates for food and woo production [8].NPP is largely influenced by many controlling factors including climate, microbi α).

Investigation Site
This research was carried out in several major rice producing districts, including Nankana Sahib, Sheikhupura, Lahore, Hafiz Abad and Gujranwala, located in Punjab province of Pakistan, as shown in Figure 1C.The spatial extent of the investigation site is (31 • -32.5 • N) and ( 73• -75 • E).This test site is located at 225 m above sea level and receives approximately 500 mm annual rainfall [37].It is a peneplain area where slope factor does not affect the water distribution process.Water distribution to rice crops is implemented through paved network channels and link canals by the Punjab Irrigation Department (PID).Crop information is maintained by a patwari (a person who preserves the complete record of the canal command area) [20].The temperature variations in the study area range between 1-4 metrological offices (LMOs) on a daily basis, assisted in the computation of various fluxes e.g., net radiation flux, soil heat flux, sensible heat flux and latent heat flux as these directly affect the net rice production.Water stress was estimated to evaluate soil wetness conditions throughout the RCP.Light used by the rice crop canopy was appraised to evaluate growth related process e.g., photosynthesis and respiration.Variations in light reflectance by the rice crop canopy were determined by analyzing temporal differences in normalized difference vegetation index (NDVI) values.These NDVI values assisted in the estimation of the biomass generated each day and hence the NPP.The estimated NPP was compared with actual NPP (computed using yield values as reported by local farmers), which helped to introduce a soil suitability constant (ħα).

Investigation Site
This research was carried out in several major rice producing districts, including Nankana Sahib, Sheikhupura, Lahore, Hafiz Abad and Gujranwala, located in Punjab province of Pakistan, as shown in Figure 1C.The spatial extent of the investigation site is (31°-32.5°N) and (73°-75° E).This test site is located at 225 m above sea level and receives approximately 500 mm annual rainfall [37].It is a peneplain area where slope factor does not affect the water distribution process.Water distribution to rice crops is implemented through paved network channels and link canals by the Punjab Irrigation Department (PID).Crop information is maintained by a patwari (a person who preserves the complete record of the canal command area) [20].The temperature variations in the study area range between 1-4° C in winter and 45-48° C in summer.The spatial location of the test site is covered by the Landsat satellite, path no.149 and row no.38, according to United States Geological Survey (USGS, Reston, VA, USA).

Flow of the Methodology
The methodology used in this research to estimate total rice production incorporating local ecological parameters, is described in detail with the flow chart shown in Figure 2

Flow of the Methodology
The methodology used in this research to estimate total rice production incorporating local ecological parameters, is described in detail with the flow chart shown in Figure 2 below.

CASA Model
We used the CASA model to examine the NPP in the investigation site.The CASA model is based on LUE, which is a commonly used index to model vegetation productivity at global scales [38][39][40][41] by fixing solar energy [42].Estimation of LUE is an important and key parameter for mapping NPP that helps to spatially diagnose the vegetation health conditions.LUE is considered as an indispensable characteristic of plants [43] that show spatial heterogeneity at different scales due to composition of the species and plant physiological characteristics.At plant scale, LUE is influenced by many factors, including the amount of chlorophyll content, leaf age, intensity of the incident light and the plant growth stage [39].However, at canopy scale, LUE is affected by canopy structure, solar zenith angle, leaf inclination angle and the LAI.Dissimilar vegetation types at the regional scale can be identified by computing spatial variability in LUE through remote-sensed data [44,45].Several methods are available to compute LUE; these include the Eddy covariance technology method [46], productivity model inversion method [47] and quantum efficiency reckoning method [48,49].During the last decade, site-based measurement was commonly used to compute LUE differences [50,51], however, the photochemical reflectance index [52][53][54][55] has been more widely used in recent years.LUE determines a positive relationship between NPP and absorbed photosynthetic active radiation (APAR).APAR can be computed using the (NDVI as input.The CASA model incorporates NDVI to compute the light absorption fraction of plants [26].NDVI, a spectral index of crop canopy, relates linearly to the fraction of photosynthetic active radiation (PAR) intercepted by the crop canopy [56].
PAR is assigned a value ranging between 45% and 50% to represent averaged 24-h conditions [62] as in Equation (2).
where K 24 is incoming solar radiation incident on crops during its growth period.The ratio of PAR and APAR represents the fraction 'f' of absorbed radiation in comparison to the available radiation for photosynthesis [44,[63][64][65] and is computed in Equation (3).
The fraction 'f' was estimated using NDVI values according to Field et al. (1995) [9] in Equation ( 4), We obtained the Landsat satellite images listed in Table 1, to compute spatiotemporal variations in NDVI throughout the RCP.LUE varies for various crops depending upon the physiological characteristics of the plants [42].Asrar et al. (1985) [66] analyzed the impact of water stress on LUE.LUE results are maximal in ideal conditions and largely influenced by local temperature and moisture conditions.The LUE for rice crops was calculated as described in [9,35,43] Equation ( 5), determines the maximum conversion factor to compute biomass above the surface of the earth in optimal environmental conditions, and a value of 1.8 gC•MJ −1 is fixed for rice crops [67].T 1 and T 2 are the heat factors and W is water stress on the crop throughout its growth.LUE will always be lower than 18 [68].T 1 determines the impact of regional cold conditions, whereas T 2 reveals that LUE decreases with deviations in actual crop temperature in comparison with the optimal temperature required for rice crop growth.T 1 and T 2 were computed according to [26,43,67].
T 1 = 0.8 + 0.02 T opt − 0.0005 (T opt ) 2  (6) where T opt is mean air temperature for the month with the maximum leaf area index (LAI) for a particular crop (e.g., we recorded maximum LAI for rice crops in the month of September), and T mon is the mean monthly air temperature during the complete crop growth period.Variations in temperature were obtained for the complete RCP from LMO s located at district level and were averaged.W determines the amount of moisture content in the soil where the crop is sown [69].A range of 0 to 1 is fixed for W, where 0 indicates an oven-dry soil while 1 indicates surplus of water [70].In its early growth stages, rice crop need excess of water, that id, W > 0.5 which reduces to 0.1 before the ripening stage [71].W is normally applied in fields to examine the soil wetness level and estimated according to [67], where R n is net radiation (Wm −2 ), G o is soil heat flux (Wm −2 ) and λE is the latent heat flux (Wm −2 ).

Net Radiation (R n )
Rn is the key component of the earth's energy balance that determines the ratio of incoming and outgoing radiations [72].This energy balance closure affects the accuracy of the estimation of evapotranspiration, therefore, it is important to estimate R n accurately to investigate the agroclimatic relationship [73].Solar radiation is responsible for all physical and chemical processes occurring in the surface-atmosphere interface [74].The physiological aspects of the rice crop canopy are computed using various parameters such as air temperature, leaf surface temperature, pressure, dew point, extraterrestrial radiation, actual sunshine hours, clear sky radiation, albedo and the actual vapor pressure [73][74][75][76][77][78][79][80][81].R n can be computed using these parameters at local and regional scales.R n is the sum of the both incoming and outgoing shortwave (R ns ) and longwave (R nl ) radiation [72,82,83].R n represents the balance between R ns and R nl , computed as follows, R nl is four times proportional to the earth's surface temperature, as stated by Stephen Boltzman law and computed according to [73].
where σ is the Stephen Boltzman constant (σ = 4.903 × 10 −9 M J K −4 m −2 day −1 ), T maxk and T mink are daily maximum and minimum temperature in Kelvin, e a is actual vapor pressure in KP a , R s and R so are incoming and actual/clear sky solar radiation (Wm −2 ).Daily variations in temperature (T maxk and T mink ) and e a for the complete RCP were obtained from LMO s and averaged.
The solar radiation striking at the top of the Earth's atmosphere (TOA) is called extraterrestrial radiation (R a ) [84].R a changes throughout the year with the position of sun due to considerable variations in the solar incidence angle [85].The fraction of R a that actually approaches the surface of earth by interacting with the atmospheric gases is called incoming solar radiation (R s ), which depends mainly upon the cloud activity [85].R s was computed according to references [72,[84][85][86][87][88][89].
where A s and B s are the calibration constants and a value range of 0.25 and 0.50 is recommended, respectively [85].N and n are maximum and actual sunshine hours, respectively.The variations in N and n for the complete RCP were obtained by LMO s and averaged.R a was computed by substituting the required parameters (latitude, month and day) in the website http://www.engr.scu.edu/~emaurer/tools/calc_solar_cgi.pl.Clear sky radiation (R so ) is nearly 75% of R a, which shows the fraction of radiation reaching the earth's surface on cloud free days.The remaining 25% of R a is absorbed, scattered or reflected back by atmospheric gases.R so was computed using the Angstrom's formula [84,86], R ns was computed by canopy surface albedo and incoming solar radiation according to [72,85].
where R s is incoming solar radiation (Wm −2 ) calculated above in Equation ( 13) and α is albedo or the crop canopy reflection coefficient which is fixed for various crops e.g., albedo ranges between 0.15 and 0.25 for rice crops [90][91][92].

Soil Heat Flux Calculation (G o )
G o represents the conduction of heat in soil due to temperature gradient depending upon the soil characteristics [93].Being a key factor, temperature controls all the biochemical processes in soil that are essential for plant growth [94].G o has the smallest contribution in the energy balance equation, thus, it is often ignored but its exclusion may lead to considerable errors.G o was computed according to [95] as follows, where R n is net radiation and LAI is the leaf area index that varies throughout the crop growth.Estimation of LAI is important because G o largely depends upon the density of leaves through which heat penetrates to increase the soil temperature.We collected LAI values for the rice crops in various growth stages through 21 field observations using the instrument LAI 2200, these are mentioned in Figure 3. First observations regarding LAI were collected on 11 July 2017 and every five days thereafter and it was assumed that LAI remained relatively similar in the four days before the next observation was made, e.g., LAI was observed to be 0.225 m 2 m −2 on 11 July 2017 and it was assumed to be the same for the next four days (12-15) July 2017.

Latent (λE) and Sensible Heat Flux (H).
Sensible heat is the exchange of energy into the plant body without changing its state.It is important to accurately determine the proportion of sensible and latent heat flux, which is a difficult process [96,97].Therefore, a residual of energy balance equation is used to compute evaporation over the vegetation surface and a Bowen ratio (β) is applied to compute sensible heat flux [98,99].β can be computed using daily variations in temperature and actual vapor pressure for canopy-environment interface as follows, β = γ ΔT Δe (16) where γ is a psychometric constant that is 0.000665 times the atmospheric pressure P [100].ΔT and Δe are temperature and actual vapor pressure gradients, respectively.The expression for γ is as follows, γ = 0.000665 × P (17) where P is daily atmospheric pressure in KPa.We collected ΔT, Δe and P from LMOs and averaged them.Sensible heat was computed using the Bowen ratio [99].
Latent heat flux is related to the estimation of crop canopy resistance which is difficult to measure because canopy resistance is dependent upon complex interactions (many times non-linear interactions) of plants with atmospheric factors [74].Therefore, to avoid the canopy resistance method, the residual of energy balance equation was used to compute the latent heat flux [96][97][98][99][101][102][103][104] as follows, λE = Rn -Go -H (19)

Landsat Image Acquisition
Remotely sensed datasets are useful to extract spatial information (e.g., LAI and land cover type) to simulate carbon dynamics from regional to global scales [9,66].Global carbon monitoring systems can be developed through ecosystem models driven using remotely sensed datasets [105].We obtained five free Landsat 8 cloud images covering the complete RCP from the USGS, as mentioned

Latent (λE) and Sensible Heat Flux (H)
Sensible heat is the exchange of energy into the plant body without changing its state.It is important to accurately determine the proportion of sensible and latent heat flux, which is a difficult process [96,97].Therefore, a residual of energy balance equation is used to compute evaporation over the vegetation surface and a Bowen ratio (β) is applied to compute sensible heat flux [98,99].β can be computed using daily variations in temperature and actual vapor pressure for canopy-environment interface as follows, where γ is a psychometric constant that is 0.000665 times the atmospheric pressure P [100].∆T and ∆e are temperature and actual vapor pressure gradients, respectively.The expression for γ is as follows, where P is daily atmospheric pressure in KP a .We collected ∆T, ∆e and P from LMOs and averaged them.Sensible heat was computed using the Bowen ratio [99].
Latent heat flux is related to the estimation of crop canopy resistance which is difficult to measure because canopy resistance is dependent upon complex interactions (many times non-linear interactions) of plants with atmospheric factors [74].Therefore, to avoid the canopy resistance method, the residual of energy balance equation was used to compute the latent heat flux [96][97][98][99][101][102][103][104] as follows,

Landsat Image Acquisition
Remotely sensed datasets are useful to extract spatial information (e.g., LAI and land cover type) to simulate carbon dynamics from regional to global scales [9,66].Global carbon monitoring systems can be developed through ecosystem models driven using remotely sensed datasets [105].We obtained five free Landsat 8 cloud images covering the complete RCP from the USGS, as mentioned in Table 1, to survey spatiotemporal variability for NDVI, APAR and biomass generation at the investigation site.

Supervised Classification
To estimate net production, it is important to accurately assess the area under rice cultivation.We performed Land Use Land Cover (LULC) supervised classification on Landsat 8 multispectral imagery (path 149, row 038) and applied various indices (NDVI, the normalized difference built-up index (NDBI) and the normalized difference water index (NDWI)) to get a classified map with high accuracy.

Estimation of Solar Radiation Received by Rice Crop Canopy throughout the RCP
Solar radiation fluxes including clear sky radiation (R so ), actual incoming radiation (R s ), net radiation (R n ), net shortwave radiation (R ns ) and net longwave radiation (R nl ) were computed by substituting the input parameters (T, e a , n, N, A s , B s and α collected through local LMOs) in Equations ( 9)-( 14) and mapped the results in Figure 4. in Table 1, to survey spatiotemporal variability for NDVI, APAR and biomass generation at the investigation site.

Supervised Classification
To estimate net production, it is important to accurately assess the area under rice cultivation.We performed Land Use Land Cover (LULC) supervised classification on Landsat 8 multispectral imagery (path 149, row 038) and applied various indices (NDVI, the normalized difference built-up index (NDBI) and the normalized difference water index (NDWI)) to get a classified map with high accuracy.

Estimation of Solar Radiation Received by Rice Crop Canopy throughout the RCP
Solar radiation fluxes including clear sky radiation (Rso), actual incoming radiation (Rs), net radiation (Rn), net shortwave radiation (Rns) and net longwave radiation (Rnl) were computed by substituting the input parameters (T, ea, n, N, As, Bs and α collected through local LMOs) in Equations ( 9)-( 14) and mapped the results in Figure 4. Figure 4 shows the variations in Ra, Rso, Rs, Rn, Rns and Rnl throughout the RCP.Ra shows the maximum solar flux available at the TOA before entering into the earth's atmosphere, whereas Rso is showing exactly the same pattern with a decline of 25% due to the interactions of Ra with atmospheric gases that results in 75% of Ra approaching the surface of earth.Ra was observed to be about 490.1 Wm −2 on 20 June, 2017, which declined to 306.17 Wm −2 on 20 October, 2017 due to a change in solar angle.The calculation of Ra, Rso and Rs determines that 52,409 Wm −2 was recorded at TOA as Ra, a fraction of which 39,307 Wm −2 , 75% × Ra entered the earth's atmosphere as Rso, however, 28,960 Wm −2 could be received as actual incoming solar radiation by the earth's surface due to cloud activity at various spatial locations throughout the RCP.Rn was observed as 27,428 Wm −2 , which included 4260 Wm −2 as Rnl and 23,168 Wm −2 as Rns.Rn is observed as 1532 Wm −2 less than Rs, the main reason of this difference is heat conduction in the soil as Go.
The variations in ratio (n/N) are mapped in Figure 5 which shows the ratio of actual sunshine hours in comparison to total sunshine hours.Careful estimates show that a total number of 1584 sunshine hours were available throughout the RCP out of which 966 sunshine hours could be received by the rice crop canopy.The main reason for this difference is the cloud activity at various locations in the study site throughout the RCP.

Ra
Rns Rs Rso Rn Rnl Figure 4 shows the variations in R a , R so , R s , R n , R ns and R nl throughout the RCP.R a shows the maximum solar flux available at the TOA before entering into the earth's atmosphere, whereas R so is showing exactly the same pattern with a decline of 25% due to the interactions of R a with atmospheric gases that results in 75% of R a approaching the surface of earth.R a was observed to be about 490.1 Wm −2 on 20 June 2017, which declined to 306.17 Wm −2 on 20 October 2017 due to a change in solar angle.The calculation of R a , R so and R s determines that 52,409 Wm −2 was recorded at TOA as R a , a fraction of which 39,307 Wm −2 , 75% × R a entered the earth's atmosphere as R so , however, 28,960 Wm −2 could be received as actual incoming solar radiation by the earth's surface due to cloud activity at various spatial locations throughout the RCP.R n was observed as 27,428 Wm −2 , which included 4260 Wm −2 as R nl and 23,168 Wm −2 as R ns .R n is observed as 1532 Wm −2 less than R s , the main reason of this difference is heat conduction in the soil as G o .
The variations in ratio (n/N) are mapped in Figure 5 which shows the ratio of actual sunshine hours in comparison to total sunshine hours.Careful estimates show that a total number of 1584 sunshine hours were available throughout the RCP out of which 966 sunshine hours could be received by the rice crop canopy.The main reason for this difference is the cloud activity at various locations in the study site throughout the RCP.

Estimation of Go throughout RCP
Variations in Go were estimated by substituting Rn and LAI (mentioned in Equation ( 15) and mapped the results in Figure 6.A gradually decreasing trend is observed in Go which fluctuated between (65.25-9.43)Wm −2 for the complete RCP.The gradual decline in Go is due to the increasing density of rice plant leaves over the growth period that prohibits the sunlight reaching the surface of earth for conduction processes.A 2nd order regression model is applied in Figure 7 to determine the relationship between LAI and Go as it has a R² = 0.9282 with a RMSE = 3.4891.Figure 7 shows that LAI and Go are in inverse relation with each other for LAI ≤ 4.5 and Go is not affected by LAI > 4.5.

Estimation of G o throughout RCP
Variations in G o were estimated by substituting R n and LAI (mentioned in Equation ( 15) and mapped the results in Figure 6.A gradually decreasing trend is observed in G o which fluctuated between (65.25-9.43)Wm −2 for the complete RCP.The gradual decline in G o is due to the increasing density of rice plant leaves over the growth period that prohibits the sunlight reaching the surface of earth for conduction processes.

Estimation of Go throughout RCP
Variations in Go were estimated by substituting Rn and LAI (mentioned in Equation ( 15) and mapped the results in Figure 6.A gradually decreasing trend is observed in Go which fluctuated between (65.25-9.43)Wm −2 for the complete RCP.The gradual decline in Go is due to the increasing density of rice plant leaves over the growth period that prohibits the sunlight reaching the surface of earth for conduction processes.A 2nd order regression model is applied in Figure 7 to determine the relationship between LAI and Go as it has a R² = 0.9282 with a RMSE = 3.4891.Figure 7 shows that LAI and Go are in inverse relation with each other for LAI ≤ 4.5 and Go is not affected by LAI > 4.5.A 2nd order regression model is applied in Figure 7 to determine the relationship between LAI and G o as it has a R 2 = 0.9282 with a RMSE = 3.4891.Figure 7

Estimation of λE and H
Figure 8 shows the variations in λE and H calculated through Equations ( 16)-( 19) by incorporating T, P, Δe, Rn and Go on a daily basis throughout the RCP.Careful estimates show that 16,549 Wm −2 was taken by the rice crop canopy as H and 7554 Wm −2 was used for the transpiration

Estimation of λE and H
Figure 8 shows the variations in λE and H calculated through Equations ( 16)-( 19) by incorporating T, P, ∆e, R n and G o on a daily basis throughout the RCP.Careful estimates show that 16,549 Wm −2 was taken by the rice crop canopy as H and 7554 Wm −2 was used for the transpiration process.

Estimation of λE and H
Figure 8 shows the variations in λE and H calculated through Equations ( 16)-( 19) by incorporating T, P, Δe, Rn and Go on a daily basis throughout the RCP.Careful estimates show that 16,549 Wm −2 was taken by the rice crop canopy as H and 7554 Wm −2 was used for the transpiration process.

Estimation of W
W indicates the transition between the dry and wet season.We estimated the variations in W and plotted the results in Figure 9 by substituting Rn, Go and H in Equation (8).It shows a decreasing trend from the start to the end of the RCP.W fluctuated between 0.5838 and 0.1218 in our study site, which reflects the water surplus and water deficiency scenarios, respectively.

Estimation of W
W indicates the transition between the dry and wet season.We estimated the variations in W and plotted the results in Figure 9 by substituting R n , G o and H in Equation ( 8).shows a decreasing trend from the start to the end of the RCP.W fluctuated between 0.5838 and 0.1218 in our study site, which reflects the water surplus and water deficiency scenarios, respectively.

Estimation of LUE
The light used by the rice crop canopy was estimated on a daily basis by substituting έ = 1.8 gC•MJ −1 in Equation ( 5) and the results are plotted in Figure 10.Maximum light intake by the rice crop canopy was observed in the early growth stages while a decreasing trend observed toward the ripening stage or on the days with cloud activity.

Estimation of LUE
The light used by the rice crop canopy was estimated on a daily basis by substituting USV Symbol Macro(s) Description  5) and the results are plotted in Figure 10.Maximum light intake by the rice crop canopy was observed in the early growth stages while a decreasing trend was observed toward the ripening stage or on the days with cloud activity.

Biomass/NPP Estimation
To estimate biomass, the input parameter 'f', 'PAR' and 'APAR' were computed.To calculate 'f', daily variations in NDVI throughout the RCP were substituted in Equation ( 4) and the results are plotted in Figure 11.'f' varies between 0 and 1 which determines the fraction of PAR used by rice crop canopy for photosynthesis as APAR.If 'f' = 0, it means that APAR is nil, while 1 depicts that 0

Biomass/NPP Estimation
To estimate biomass, the input parameter 'f', 'PAR' and 'APAR' were computed.To calculate 'f', daily variations in NDVI throughout the RCP were substituted in Equation ( 4) and the results are plotted in Figure 11.'f' varies between 0 and 1 which determines the fraction of PAR used by rice crop canopy for photosynthesis as APAR.If 'f' = 0, it means that APAR is nil, while 1 depicts that APAR = PAR.Figure 11 shows a sharp decline in 'f' on 1 October 2017 because the rice crop turned to yellow at the ripening stage on this date.APAR is directly affected by NDVI, a decrease in NDVI leads to a drop in APAR and vice versa.

Biomass/NPP Estimation
To estimate biomass, the input parameter 'f', 'PAR' and 'APAR' were computed.To calculate 'f', daily variations in NDVI throughout the RCP were substituted in Equation ( 4) and the results are plotted in Figure 11.'f' varies between 0 and 1 which determines the fraction of PAR used by rice crop canopy for photosynthesis as APAR.If 'f' = 0, it means that APAR is nil, while 1 depicts that APAR = PAR.Figure 11 shows a sharp decline in 'f' on 1 October, 2017 because the rice crop turned to yellow at the ripening stage on this date.APAR is directly affected by NDVI, a decrease in NDVI leads to a drop in APAR and vice versa.We estimated biomass generation on a daily basis by substituting APAR and ε values in Equation ( 1) and the results are plotted in Figure 12.Biomass generation declined as NDVI dropped at the ripening stage mentioned in Figure 11, during the month of October.The decline in NDVI directly affects both the APAR and ε, because rice plants need comparatively less energy at the ripening stage as compared to its overall growth stages, consequently, APAR and ε dropped even in cloud free October.Peaks and dips in Figure 12 describe sunlight (n/N ratio) available for APAR and ε that directly affects the biomass potency.
Landsat 8, remotely sensed datasets listed in Table 1, were used to survey spatiotemporal variations in PAR, APAR, NDVI and biomass generation and the results are mapped in Figure 13.Lush green areas in Figure 13 determine the spatial extent of vegetation while blue areas represent non-vegetative features.We estimated biomass generation on a daily basis by substituting APAR and ε values in Equation ( 1) and the results are plotted in Figure 12.Biomass generation declined as NDVI dropped at the ripening stage mentioned in Figure 11, during the month of October.The decline in NDVI directly affects both the APAR and ε, because rice plants need comparatively less energy at the ripening stage as compared to its overall growth stages, consequently, APAR and ε dropped even in cloud free October.Peaks and dips in Figure 12 describe sunlight (n/N ratio) available for APAR and ε that directly affects the biomass potency.Landsat 8, remotely sensed datasets listed in Table 1, were used to survey spatiotemporal variations in PAR, APAR, NDVI and biomass generation and the results are mapped in Figure 13.Lush green areas in Figure 13 determine the spatial extent of vegetation while blue areas represent non-vegetative features.Careful estimates revealed that 4897 kg/ha biomass is generated in the study site throughout the RCP.The harvest index for rice crops is taken as 0.50 [106], which resulted in a dry mass of 2450 kg/ha (2.7 ton/ha).To compute net production, we extracted the area under rice cultivation through supervised classification.The classification results show that the total area under investigation was 13,680 km 2, of which 8861 km 2 (886,100 ha) was under rice cultivation.A field survey was conducted along with Maali patwari (a local expert) to cross validate the classification results which were 92.35% accurate.According to international standards the reliability of supervised neighborhood classification may be up to 94% [107].The net rice production was approximated as 2.39 million tons.

Discussion
Ratio (n/N) determines the cloud activity, which ranges between 0 and 1 (0 indicates a dense cloudy day and 1 is a cloud free day).This ratio has a direct impact on all productivity parameters e.g., R ns , R nl , R n , R s , H, λE, LUE as well as the biomass generation.The peaks and dips in Figures 4, 6, 8, 10 and 12 show the same pattern as the (n/N) ratio in Figure 5.A decline in (n/N), for example, 0.5 determines that half of the day is without sunshine and this results in a drop in observed R n , R ns , R nl and R s .R s is equivalent to R so on the day (where n = N).LUE is directly affected by the (n/N) ratio because it limits the sunlight falling on the rice crop canopy, therefore, photosynthesis/respiration process is halted or delayed.Ratio (n/N) can be significant or insignificant during various stages of rice crop growth (e.g., leaf emergence and tillering processes need comparatively high temperatures for proper growth where (n/N) ratio is feasible and near to 1), besides, the flowering, milk, dough and ripening stages remain good at lower temperatures when the (n/N) ratio should be less than 0.5.Figure 8, shows a decline for both λE and H for the days with partial or full cloud activity because less energy is required to maintain the plant internal temperature as compared to its surroundings.G o is also affected by the ratio (n/N) as shown in Figure 6, e.g., the dip in G o in Figure 5 on 24 July 2017 describes the dense cloud activity, which resulted in the non-availability of heat for conduction through the soil.Figure 6 shows that G o was comparatively high in the early growth stages due to less LAI as compared to forthcoming growth stages.
The difference in R n and G o is equivalent to the collective impact of λE and H on the rice crop canopy.λE and H are directly affected by the ratio (n/N).H is the transfer of heat to a rice plant to maintain its internal temperature according to the ambient air.λE is the energy consumption for the transpiration process.Temperature drops on a cloudy day due to non-availability of direct sunlight and hence less energy is consumed for both the λE and H.
W indicates the soil wetness level which ranges between 0 and 1. Temporal variations in W were mapped in Figure 9 and determine the soil wetness condition throughout the RCP.Rice crops in the form of green biomass in various growth stages had high concentration of water i.e., (W ≥ 0.5).W declined in the same way as green biomass turned to yellow during the final ripening stage and reached a level of 0.1218.W = 0 indicates that the rice plant is oven-dry.The same results for water stress were obtained by [9].T 1 and T 2 expresses the regulation of rice plants by temperature.Light used by the rice crop canopy may not be well computed, if these key parameters are ignored.T 2 represents the variations in the temperature regulation on the rice crop throughout the RCP while T 1 incorporates the impact of temperature fluctuations during the month with LAI max .LAI max was obtained through field observations that varied between (6.63-6.65)m 2 m −2 during the month of September 2017 with an optimum temperature of 30.5 • C. T 2 becomes 1 if T opt and T mon are equal to each other.T 2 = 0.5 determines that T mon differs with T opt by 10-15 • C. T opt and T mon were observed as slightly different from each other in our study site, however, significant variations in T opt and T mon may be observed by investigating agricultural productivity at a continental scale.The fraction 'f' of absorbed radiation in comparison to available radiation is mapped in Figure 11.This shows that about 80% of incoming radiation was used by the rice crop canopy and this suddenly dropped on 1 October 2017.This sharp decline was due to a decrement in NDVI on the same day, therefore, the reflective index of the rice crop canopy (NDVI) is an important parameter of productivity.As NDVI drops, the LUE of the plant declines and hence the biomass generation decreases regardless of weather conditions, as shown in Figures 10-12.The collective impact of all the parameters of productivity are mapped in Figure 12.There was a resemblance in Figures 5 and 12 which determines that the (n/N) ratio directly affects rice productivity.Figure 13 shows the spatiotemporal variations in PAR, APAR, NDVI and biomass generation.The lowest values in NDVI, APAR and biomass datasets reveal various features (e.g., a built-up area, a water body and some non-cropped and salinized land) other than green biomass.A gradual decreasing trend is observed in PAR values from July to October due to a change in sun inclination angle.NDVI was recorded as maximum in the month of September due to increased LAI; this suddenly declined in October in the ripening stage.Maximum NDVI results in enhanced biomass generation therefore, there is a direct relationship between NDVI and biomass generation.Biomass generation was observed least in October because the rice crop was in the ripening stage during that time.APAR was observed as maximal in the early growth stages of the rice plant because an increased growth is observed at the same time due to the emergence of leaves and tillering.APAR degraded in October because the rice crop turned yellow because photosynthesis was partially halted, even during sunshine hours.

Development of Soil Suitability Constant (hα)
Actual yield levels were obtained through 55 field observations taken randomly by interviewing local farmers.The observations sites were chosen keeping in mind the diversity in yield reported by farmers at rice monitoring centers.High variations were observed between actual (obtained through the field survey) and estimated (using the CASA model) yields.We obtained four ranges of actual yield, which were 400-500 kg/ha, 1000-1200 kg/ha, 1700-1800 kg/ha and 2100-2200 kg/ha.These ranges were 0.16-0.2%,0.4-0.48%,0.69-0.73%and 0.85-0.90% of the CASA based yield, respectively.To investigate further, an interpolated map was developed based on actual yield levels, as shown in Figure 14, and it was compared with soil suitability zones delineated by Hassan et al., 2018 [108].We observed that the lowest yield of 414 kg/ha (0.16% of the CASA based yield) was obtained from areas which were not suitable for rice cultivation with pH > 8.4, electric conductivity (EC) of 1.75-2.25 and excessively drained soil type (ST).In addition, for the other three yield ranges the following results were obtained: (1) 1100-1200 kg/ha with soil suitability levels, pH = 4-5, EC = 1.75-2.25 and highly drained ST; (2) 1700-1800 kg/ha with soil suitability levels, pH = 7.7-8.3,EC = 0.25-0.75 and well drained ST; and (3) 2100-2200 kg/ha with soil suitability levels, pH = 5.5-7.2,EC = 0.75-1.50and poorly drained ST.The CASA model demonstrates wide differences between actual and estimated yield [33,34,44,64,67,68] because it ignores soil suitability levels that must be incorporated to obtain precise yield.So, we developed a soil suitability constant, denoted as , respect stress on rice crops varied between 0.5838 and 0.1218 from the start until the end of the R generation declined from 6.09-1.03g/m 2 in the tillering and ripening stages, respectivel a soil suitability constant (ħα) into the Carnegie-Ames-Stanford Approach (CASA) mod a more precise estimate of yield.Classification results suggest that the total area cultivation was 8861 km 2 . The spatial distribution of rice cultivation as per suitability zo km 2 was not suitable (NS), 592 km 2 was less suitable (LS), 2210 km 2 was moderately s and 4385 km 2 was highly suitable (HS) soil type with ħα ranges of 0.05-0.25,0.4-0.6,0.85-0.95 of the CASA based yield, respectively.We estimated net production as 1.63 as per 0.46 ton/ha, 1.2 ton/ha 1.9 ton/ha and 2.4 ton/ha from NS, LS, MS and HS respectively.The results obtained through this improved CASA model, by addition of ħα, are likely to be useful for agronomists by providing more accurate estimates of NP Keywords: net primary production; biomass estimation; photosynthesis; resp cultivation period; net radiation α * HI (Kg/ha) There were other ground validations that caused an over estimation in yield, which were: (1) non-availability of open-source satellite images with high spatiotemporal resolution to determine growth variability on a daily basis; (2) poverty and water scarcity faced by local farmers; (3) lack of state-of-the-art technology, poor agricultural and educational practices; and (4) blind faith of farmers on the crop calendar, which has now become obsolete due to existing phenological changes.
We further superimposed the spatial distribution of rice cultivation to yield a variability map (already compared with soil suitability zones), [108] and found that the farmers had planted rice in

Introduction
About three billion of the world's population consume rice (Oryza Sativa) as an e food/crop [1].Almost 88% of global rice production is obtained from Asia [2].The seven larg producing countries in the world are China (210.1 million tons), India (165.3 million tons), In (74.2 million tons), Bangladesh (53.1 million tons), Vietnam (44 million tons), Thailand (33.3 tons) and Myanmar (28.3 million tons) in 2016-2017, which together accounts for 680.3 milli [3], whereas in Pakistan, annual rice production is only 6 million tons.The global rice produ insufficient to cater for the needs of the community [4].Increasing population in rice con countries [5] and climate change in recent decades [6,7], has put massive pressure on rice pr to increase per-hectare yield.Therefore, it is important to ensure rice sustainability through of state-of-the-art agricultural practices.Precise estimation of NPP has become a great challe agronomists and economist to evaluate the exact quantity to import/export in c shortfall/surplus.NPP is a significant indicator that gives accurate estimates for food and production [8].NPP is largely influenced by many controlling factors including climate, m α ranges according to soil suitability zones and computed 1.63 million tons as the net production in the study site, including 0.46 ton/ha for the not suitable zone, 1.2 ton/ha for the less suitable zone, 1.9 ton/ha for the moderately suitable zone and 2.4 ton/ha for the highly suitable zone.
poorly drained ST.The CASA model demonstrates wide differences between actual and estimated yield [33,34,44,64,67,68] because it ignores soil suitability levels that must be incorporated to obtain precise yield.So, we developed a soil suitability constant, denoted as ħα, on the basis of our results.The values of ħα ranges between 0.05-0.95.We further subdivided it into four ranges; 0.05-0.25 for not suitable, (0.4-0.6) for less suitable, (0.7-0.75) for moderately suitable and (0.85-0.95) for highly suitable soil types.These ħα based levels can be applied only if one has delineated cultivation sites based on physical and ecological grounds.The refining of ħα can improve the CASA based NPP.Finally, the improved method for computing NPP through the CASA model with the addition of ħα is given as, Net Productions = APAR * ε * ħα * HI (Kg/ha) We further superimposed the spatial distribution of rice cultivation to yield a variability map (already compared with soil suitability zones), [108] and found that the farmers had planted rice in 592 km 2 of less suitable soil, in 4385 km 2 of highly suitable soil, in 2210 km 2 of moderately suitable soil, and in 1674 km 2 of not suitable soil.We applied ħα ranges according to soil suitability zones and computed 1.63 million tons as the net production in the study site, including 0.46 ton/ha for the not suitable zone, 1.2 ton/ha for the less suitable zone, 1.9 ton/ha for the moderately suitable zone and 2.4 ton/ha for the highly suitable zone.

Conclusions
Remote sensing has become a useful tool for monitoring NPP and biomass generation.Remotely sensed datasets are of great importance to repeatedly monitor agricultural productivity.This research was conduction through the CASA model in collaboration with freely available Landsat imagery that approximated the rice productivity, resulted in significant differences as compared to actual estimates.The differences in productivity were due to the soil suitability zones being ignored, thus, we added a new soil suitability constant (ħα).
This research can be applied to surveys of spatiotemporal variations in yield that require actual sunshine hours, temperature, actual vapor pressure, rate of change in leaf area and the air pressure on a daily basis.This research opens avenues to investigate ħα ranges in detail.

Introduction
About three billion of the world's population consume rice (Oryza Sativa) as a food/crop [1].Almost 88% of global rice production is obtained from Asia [2].The seven producing countries in the world are China (210.1 million tons), India (165.3 million tons (74.2 million tons), Bangladesh (53.1 million tons), Vietnam (44 million tons), Thailand ( tons) and Myanmar (28.3 million tons) in 2016-2017, which together accounts for 680.3 [3], whereas in Pakistan, annual rice production is only 6 million tons.The global rice p insufficient to cater for the needs of the community [4].Increasing population in rice countries [5] and climate change in recent decades [6,7], has put massive pressure on ric to increase per-hectare yield.Therefore, it is important to ensure rice sustainability thro α ranges in detail.
Author Contributions: S.M.H.R. and S.A.M. generated the complete flow of methodology used in this research and arranged data on a daily basis from LMOs, rice monitoring centers and by local farmers.S.A.M. arranged funds and conducted field surveys for ground truthing.S.M.H.R. and S.A.M. produced the results and discussed with.A.A.K. (MNSUM) who checked the whole flow of the research.He also helped in reviewing this research by his three associates living in Germany, Brazil and Indonesia, regarding English related issues.

Figure 1 .
Figure 1.(A) Map of Punjab province in Pakistan; (B) Investigation site in district map of Punjab (C) Spatial extent of the investigation site.

Figure 1 .
Figure 1.(A) Map of Punjab province in Pakistan; (B) Investigation site in district map of Punjab; (C) Spatial extent of the investigation site.

Figure 2 .
Figure 2. Flow chart of the methodology.Figure 2. Flow chart of the methodology.

Figure 2 .
Figure 2. Flow chart of the methodology.Figure 2. Flow chart of the methodology.

Figure 3 .
Figure 3. Variations recorded in LAI on various dates with a temporal resolution of five days throughout the RCP.

Figure 4 .
Figure 4. Variations in extraterrestrial radiation (R a ); clear sky radiation (R so ); actual incoming radiation (R s ); net radiation (R n ); net longwave radiation (R nl ) and net shortwave radiation (R ns ) throughout the RCP.

Figure 6 .
Figure 6.Variations in Go for the complete RCP.

Figure 6 .
Figure 6.Variations in Go for the complete RCP.

Figure 6 .
Figure 6.Variations in G o for the complete RCP.

Figure 7 .
Figure 7. Correlation between Go and LAI.

Figure 8 .
Figure 8. Variation in λE and H throughout RCP.

Figure 8 .
Figure 8. Variation in λE and H throughout RCP.

Figure 9 .
Figure 9. Variation in W throughout the RCP.

Figure 10 .
Figure 10.Variations in LUE throughout the RCP.

Figure 10 .
Figure 10.Variations in LUE throughout the RCP.

Figure 12 .
Figure 12.Daily variation in biomass throughout the RCP.

Figure 12 .
Figure 12.Daily variation in biomass throughout the RCP.

Figure 14 .
Figure 14.(A) Spatial distribution of rice yield; (B) Rice cultivation mask with grey areas representing urban and built-up areas; (C) Integration of yield distribution with rice cultivation mask.
April 2018; Accepted: 28 May 2018; Published: 29 May 2018 Net primary production (NPP) is an important indicator of the supply of food and wood.d a hierarchy model and real time field observations to estimate NPP using satellite imagery.iation received by rice crop canopies was estimated as 27,428 Wm −2 t:

The incorporation of ation of Net Rice Production through Improved A Model by Addition of Soil Suitability Constant uhammad Hassan Raza * and Syed Amer Mahmood
• C in winter and 45-48 • C in summer.The spatial location of the test site is covered by the Landsat satellite, path no.149 and row no.38, according to United States Geological Survey (USGS, Reston, VA, USA).

Article Estimation of Net Rice Production through Impr CASA Model by Addition of Soil Suitability Con (ħα) Syed Muhammad Hassan Raza * and Syed Amer Mahmood
Net primary production (NPP) is an important indicator of the supply of fooWe used a hierarchy model and real time field observations to estimate NPP using satel Net radiation received by rice crop canopies was estimated as 27,428 Wm −2 Abstract: