Within-Field Rice Yield Estimation Based on Sentinel-2 Satellite Data

Rice is considered one of the most important crops in the world. According to the Food and Agriculture Organization of the United Nations (FAO), rice production has increased significantly (156%) during the last 50 years, with a limited increase in cultivated area (24%). With the recent advances in remote sensing technologies, it is now possible to monitor rice crop production for a better understanding of its management at field scale to ultimately improve rice yields. In this work, we monitor within-field rice production of the two main rice varieties grown in Valencia (Spain) JSendra and Bomba during the 2020 season. The sowing date of both varieties was May 22–25, while the harvesting date was September 15–17 for Bomba and October 5–8 for JSendra. Rice yield data was collected over 66.03 ha (52 fields) by harvesting machines equipped with onboard sensors that determine the dry grain yield within irregular polygons of 3–7 m width. This dataset was split in two, selecting 70% of fields for training and 30% for validation purposes. Sentinel-2 surface reflectance spectral data acquired from May until September 2020 was considered over the test area at the two different spatial resolutions of 10 and 20 m. These two datasets were combined assessing the best combination of spectral reflectance bands (SR) or vegetation indices (VIs) as well as the best timing to infer final within-field yields. The results show that SR improves the performance of models with VIs. Furthermore, the correlation of each spectral band and VIs with the final yield changes with the dates and varieties. Considering the training data, the best correlation with the yields is obtained on July 4, with R2 for JSendra of 0.72 at 10 m and 0.76 at 20 m resolution, while the R2 for Bomba is 0.87 at 10 m and 0.92 at 20 m resolution. Based on the validation dataset, the proposed models provide within-field yield modelling Mean Absolute Error (MAE) of 0.254 t×ha−1 (Mean Absolute Percentage Error, MAPE, of 3.73%) for JSendra at 10 m (0.240 t×ha−1; 3.48% at 20 m) and 0.218 t×ha−1 (MAPE 5.82%) for Bomba (0.223 t×ha−1; 5.78% at 20 m) on July 4, that is three months before harvest. At parcel level the model’s MAE is 0.176 t×ha−1 (MAPE 2.61%) for JSendra and 0.142 t×ha−1 (MAPE 4.51%) for Bomba. These results confirm the close correlation between the rice yield and the spectral information from satellite imagery. Additionally, these models provide a timeliness overview of underperforming areas within the field three months before the harvest where farmers can improve their management practices. Furthermore, it highlights the importance of optimum agronomic management of rice plants during the first weeks of rice cultivation (40–50 days after sowing) to achieve high yields.

Download Full-text

The floating traditional Rice Techno-Demonstration for Community Food Security

International Journal of Agricultural Extension ◽

10.33687/ijae.007.01.1778 ◽

2019 ◽

Vol 7 (1) ◽

pp. 09-11

Author(s):

Pasigan U. Buisan ◽

Nasrudin A. Buisan

Keyword(s):

Water Hyacinth ◽

Crop Production ◽

Management Practices ◽

Rice Variety ◽

The Philippines ◽

Rice Production ◽

Community Food Security ◽

Rice Varieties ◽

Significant Difference ◽

Traditional Rice

Rice ranks as the most important cereal in the Philippines. It is one of the country’s largest agricultural production systems, highest among crop production. The general objective of the study was to determine if traditional rice variety can give optimum yield when planted in the floating area. The materials used were traditional rice varieties. The water hyacinth was converted into a rice floating area. Data gathered was analyzed using the analysis of variance. The LSD Test was employed to compare the significant difference between treatment mean values. Floating rice production is a viable option among small farmers particularly in areas where water availability is not control such as marshy area. Local materials such as water hyacinth as floater are widely available. There is a need however, for a sound cultural management to make it sustainable. The good agronomic characteristics of the rice varieties can be attributed by the types of variety and the management practices. The floating area can be converted a potential area for rice production for a certain community particularly in the marshy area. However, there a need for a sound management practices in the floating area that can farmers may adopt for a sustainable rice floating production area

Download Full-text

Assessing Opportunities to Increase Yield and Profit in Rainfed Lowland Rice Systems in Indonesia

Agronomy ◽

10.3390/agronomy11040777 ◽

2021 ◽

Vol 11 (4) ◽

pp. 777

Author(s):

Erythrina Erythrina ◽

Arif Anshori ◽

Charles Y. Bora ◽

Dina O. Dewi ◽

Martina S. Lestari ◽

...

Keyword(s):

Best Management Practices ◽

Management Practices ◽

Nutrient Management ◽

Agricultural Research ◽

Rice Yield ◽

Information Access ◽

Water Reservoir ◽

Dry Season ◽

Rice Production ◽

Yield Gap

In this study, we aimed to improve rice farmers’ productivity and profitability in rainfed lowlands through appropriate crop and nutrient management by closing the rice yield gap during the dry season in the rainfed lowlands of Indonesia. The Integrated Crop Management package, involving recommended practices (RP) from the Indonesian Agency for Agricultural Research and Development (IAARD), were compared to the farmers’ current practices at ten farmer-participatory demonstration plots across ten provinces of Indonesia in 2019. The farmers’ practices (FP) usually involved using old varieties in their remaining land and following their existing fertilizer management methods. The results indicate that improved varieties and nutrient best management practices in rice production, along with water reservoir infrastructure and information access, contribute to increasing the productivity and profitability of rice farming. The mean rice yield increased significantly with RP compared with FP by 1.9 t ha–1 (ranges between 1.476 to 2.344 t ha–1), and net returns increased, after deducting the cost of fertilizers and machinery used for irrigation supplements, by USD 656 ha–1 (ranges between USD 266.1 to 867.9 ha–1) per crop cycle. This represents an exploitable yield gap of 37%. Disaggregated by the wet climate of western Indonesia and eastern Indonesia’s dry climate, the RP increased rice productivity by 1.8 and 2.0 t ha–1, with an additional net return gain per cycle of USD 600 and 712 ha–1, respectively. These results suggest that there is considerable potential to increase the rice production output from lowland rainfed rice systems by increasing cropping intensity and productivity. Here, we lay out the potential for site-specific variety and nutrient management with appropriate crop and supplemental irrigation as an ICM package, reducing the yield gap and increasing farmers’ yield and income during the dry season in Indonesia’s rainfed-prone areas.

Download Full-text

Time-Series of Vegetation Indices (VNIR/SWIR) Derived from Sentinel-2 (A/B) to Assess Turgor Pressure in Kiwifruit

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi9110641 ◽

2020 ◽

Vol 9 (11) ◽

pp. 641

Author(s):

Alberto Jopia ◽

Francisco Zambrano ◽

Waldo Pérez-Martínez ◽

Paulina Vidal-Páez ◽

Julio Molina ◽

...

Keyword(s):

Time Series ◽

Crop Production ◽

Management Practices ◽

Water Status ◽

Vegetation Indices ◽

Turgor Pressure ◽

Security Risk ◽

Vegetative Development ◽

Water Sensor ◽

Sentinel 2

For more than ten years, Central Chile has faced drought conditions, which impact crop production and quality, increasing food security risk. Under this scenario, implementing management practices that allow increasing water use efficiency is urgent. The study was carried out on kiwifruit trees, located in the O’Higgins region, Chile for season 2018–2019 and 2019–2020. We evaluate the time-series of nine vegetation indices in the VNIR and SWIR regions derived from Sentinel-2 (A/B) satellites to establish how much variability in the canopy water status there was. Over the study’s site, eleven sensors were installed in five trees, which continuously measured the leaf’s turgor pressure (Yara Water-Sensor). A strong Spearman’s (ρ) correlation between turgor pressure and vegetation indices was obtained, having −0.88 with EVI and −0.81 with GVMI for season 2018–2019, and lower correlation for season 2019–2020, reaching −0.65 with Rededge1 and −0.66 with EVI. However, the NIR range’s indices were influenced by the vegetative development of the crop rather than its water status. The red-edge showed better performance as the vegetative growth did not affect it. It is necessary to expand the study to consider higher variability in kiwifruit’s water conditions and incorporate the sensitivity of different wavelengths.

Download Full-text

Integrating Statistical and System Dynamics Modelling to Analyse the Impacts of Climate Change on Rice Production in West Nusa Tenggara, Indonesia

10.26686/wgtn.17060483.v1 ◽

2021 ◽

Author(s):

◽

Muhamad Bahri

Keyword(s):

Climate Change ◽

System Dynamics ◽

Statistical Models ◽

Rice Yield ◽

Rice Production ◽

Maximum Temperature ◽

Dynamics Model ◽

Rice Varieties ◽

Changing Climate ◽

Impacts Of Climate Change

<p>Climate change, manifested as temperature rise and rainfall change, will pose significant challenges to rice farmers, leading to a possible rice shortage under a changing climate. This research aims to understand the impacts of climate variability and change on rice production through the rest of this century using Representative Concentration Pathway (RCP) scenarios, and combination of statistical and system dynamic modelling. The area of study is West Nusa Tenggara, Indonesia. Wetland and dryland farming types are assessed separately because they have different rice varieties and different agricultural practices. Overall, the research seeks to answer the question: How will climate change and climate variability affect rice production? Additional questions investigated are (1) What are the most significant supply uncertainties associated with a changing climate? and (2) What are possible solutions for reducing the impacts of climate change on rice production?. To answer these research questions, this study deals with three main research areas. First, based on observed data (1976-2011), this study developed regression-based statistical models in understanding the impacts of climate change on rice yield in West Nusa Tenggara. Statistical models find that the negative impacts of increased minimum temperature on rice yield are statistically significant. By contrast, the effects of maximum temperature on rice yield are not statistically significant. A key reason for this is that the highest maximum temperature (32⁰C) in the observed period (1976-2011) was lower than 35⁰C, a rice threshold for maximum temperature. By 2090 (2077-2100), rice yield in wetland and dryland is projected to decrease by about 3% (RCP2.6 scenario), 4% (RCP4.5 scenario), 5% (RCP6.0 scenario) and 14% (RCP8.5 scenario). Second, a system dynamics model was developed to assess the impacts of climate change on three issues including rice yield, harvested areas and rice production by 2090 (2077-2100). After embedding statistical models and estimating the impacts of maximum temperature on rice yield based on existing studies, the impacts of climate change on rice yield are projected. The system dynamics model is also equipped by August SOI to estimate the impacts of climate change on the timing of monsoon onset i.e the beginning of planting seasons. For assessing harvested areas under a changing climate, the system dynamics model is equipped by a mathematical relationship between seasonal rainfall and harvested areas. Because the system dynamics model includes the impacts of high maximum temperature, the projected loss of rice yield in wetland and dryland is relatively higher compared to that in statistical models. It is projected that rice yield loss will be about 3% (RCP2.6 scenario), 6% (RCP4.5 scenario), 10% (RCP6.0 scenario) and 23% (RCP8.5 scenario) by 2090 (2077-2100). Likewise, rice production loss in wetland and dryland is projected to be about 1% (RCP2.6 scenario), 2% (RCP4.5 scenario), 7% (RCP6.0 scenario) and 19% (RCP8.5 scenario) by 2090 (2077-2100). The projected loss of rice production is relatively lower than rice yield loss as wetland harvested areas are projected to experience a slight increase about 3% by 2090 (2077-2100) under a changing climate. This also means that the ranking of the impacts of climate change from the most significant to the least significant is its impact on rice yield, rice production and harvested areas. Third, policy options in overcoming the impacts of climate change on rice production are assessed. This study suggests that research on finding rice varieties with three main traits: heat tolerance, short growth duration and high yield is key to balance rice demand and rice supply in West Nusa Tenggara by 2090 (2077-2100). A failure to improve rice yield in such ways is likely to lead to significant reductions in rice supply in the face of climate change. This study makes theoretical contributions, including the development of statistical models for understanding the impacts of climate change on rice yield and a causal system for investigating the impacts of climate change on rice yield, rice production and harvested areas. Again, the combination of statistical and system dynamics modelling simultaneously investigates the impacts of climate change on rice yield, rice production and harvested areas. This means that this study provides a more holistic view of the impacts of climate change compared to existing studies. This study also offers practical contributions, advising that declining rice research should be avoided under a changing climate, and suggesting that farming intensification (more climate-resilient rice varieties) is more effective than farming extension (area expansion) in sustaining rice production under a changing climate. Again, research on developing more resilient-climate rice varieties is possible as projected rice yield in sustaining rice production by 2090 (2077-2100) is similar to rice’s yield potential.</p>

Download Full-text

Smallholder Farmers’ Knowledge, Perception and Management of Rice Blast Disease in Upland Rice Production in Tanzania

Journal of Agricultural Science ◽

10.5539/jas.v10n7p137 ◽

2018 ◽

Vol 10 (7) ◽

pp. 137 ◽

Cited By ~ 1

Author(s):

Ibrahim Hashim ◽

Delphina P. Mamiro ◽

Robert B. Mabagala ◽

Tadele Tefera

Keyword(s):

Rice Blast ◽

Management Practices ◽

Crop Residues ◽

Upland Rice ◽

Smallholder Farmers ◽

Household Survey ◽

Rice Production ◽

Rice Blast Disease ◽

Blast Disease ◽

Rice Varieties

The objective of this paper was to investigate farmers’ knowledge and management of rice blast disease in Tanzania. Farmers’ household survey was conducted in five districts namely Mvomero, Morogoro rural, Ulanga, Korogwe and Muheza in April and May 2017. Data were collected through face-to-face interviews using semi-structured questionnaire and observations made through transect walks across selected villages. Farmers observed symptoms of rice blast disease for the first time in the past 3 to 10 years, with higher severity of blast disease in April to May each year. About 46.3% of the respondents were not aware of the cause and spread of rice blast disease. About 39.9% of the respondents associated rice blast disease with drought, high rainfall and temperature (8.7%) and soil fertility problems (5.1%). About 18.7% of the farmers reported burning of crop residues, 17.0% use of ash, 4.0% use of nitrogen fertilizer and 6.3% application of fungicide for management of rice blast disease. The majority (54.0%) of farmers did not apply any management method. Most farmers planted local upland rice varieties, with only 7.7% using improved varieties. About 69.6% of the respondents shared information on disease management among themselves. Lack of knowledge, ability to afford and unavailability of effective blast disease control methods were reported to affect the management of the disease. Strengthening the capacity of farmers to identify the disease and proper management practices will sustainably solve the problem of rice blast disease in upland rice production.

Download Full-text

Evaluation of Different Nutrient Management Practices in Yield of Different Rice Cultivars in Lamjung District of Nepal

International Journal of Applied Sciences and Biotechnology ◽

10.3126/ijasbt.v4i2.15127 ◽

2016 ◽

Vol 4 (2) ◽

pp. 223-227

Author(s):

S. Ranabhat ◽

L.P. Amgain

Keyword(s):

Grain Yield ◽

Management Practices ◽

Nutrient Management ◽

Rice Yield ◽

Block Design ◽

Rice Cultivars ◽

Rice Varieties ◽

Fertilizer Recommendations ◽

Rice Yields ◽

Biological Yield

A field experiment was conducted on farmer’s field at two sites of Lamjung district of Nepal viz. Bhotewodar and Sundarbazaar to evaluate the performance of two commonly grown rice varieties viz US-382(hybrid) and Ramdhan (improved) under two nutrient management practices [Nutrient Expert®(NE) rice model recommendation, and government recommendation(GR)].Four replicates of four treatments were arranged in randomized completely block design. Rice varieties responded differently under the different nutrient management practices in terms of plant height, grains per panicle, sterility%, panicle weight, grain yield at 15% moisture, straw yield and harvesting index. The highest grain yield was obtained from NE field of US-382 variety which was followed by GR for US-382, NE for Ramdhan and GR for Ramdhan variety. NE based practices for US-382 variety produced higher biological yield as compared to GR. NE estimated attainable rice yield provided by the software compared with actual rice yield from the trials in farmer’s field and NE-based fertilizer recommendations proved the validity in reaching the yield targets estimated by the software. The observed rice yields recorded in the trials were higher than the NE estimated attainable yields, so NE recommendation for US-382 variety was found better over GR.Int J Appl Sci Biotechnol, Vol 4(2): 223-227

Download Full-text

Deep learning based F-Mask alternative for Sentinel-2 images in polar regions

10.5194/egusphere-egu21-15914 ◽

2021 ◽

Author(s):

Thorsten Seehaus ◽

Kamal Nambiar Gopikrishnan ◽

Veniamin Morgenshtern ◽

Philipp Hochreuther ◽

Matthias Braun

Keyword(s):

Deep Learning ◽

Data Augmentation ◽

Satellite Image ◽

Training Data ◽

Supervised Machine Learning ◽

Machine Learning Techniques ◽

Validation Dataset ◽

Polar Regions ◽

Unseen Data ◽

Sentinel 2

<div>Screening clouds, cloud shadows, and snow is a critical pre-processing step that needs to be performed before any meaningful analysis can be done on satellite image data. The state of the art 'F-Mask' algorithm, which is based on multiple pixel-level threshold tests, segments the image into clear land, cloud, cloud shadow, snow, and water classes. However, we observe that the results of this algorithm are not very accurate in polar and tundra regions. The unavailability of labeled Sentinel-2 training datasets with these classes makes the traditional supervised machine learning techniques difficult to implement. Experiments with large, noisy training data on standard deep learning classification tasks like CIFAR-10 and ImageNet have shown neural networks learn clean labels faster than noisy labels.&#160;</div><div>We present a multi-level self-learning approach that trains a model to perform semantic segmentation on Sentinel-2 L1C images. We use a large dataset with labels annotated using the F-mask algorithm for the training, and a small human-labeled dataset for validation. The validation dataset contains numerous examples where the F-mask classification would have given incorrect labels. At the first step, a deep neural network with a modified U-Net architecture is trained using a dataset automatically labeled with the F-mask algorithm. The performance on the validation dataset is used to select the best model from the step, which would then be used to generate more training labels from previously unseen data. In each of the subsequent steps, a new model is trained using the labels generated using the model from the previous step. The amount of data used for training increases with each step and the application of techniques like data augmentation and dropout improves the generalization of the trained model. We show that the final model from our approach can outperform its teacher, i.e. F-Mask algorithm.&#160;</div>

Download Full-text

APPRAISAL OF MAJOR DETERMINANTS OF RICE PRODUCTION AND FARMERS’ CHOICE OF RICE IDEOTYPES IN SOUTH SUDAN: IMPLICATIONS FOR BREEDING AND POLICY INTERVENTIONS

Experimental Agriculture ◽

10.1017/s0014479718000017 ◽

2018 ◽

Vol 55 (1) ◽

pp. 143-156

Author(s):

M. MOGGA ◽

J. SIBIYA ◽

H. SHIMELIS ◽

J. LAMO ◽

N. OCHANDA

Keyword(s):

Management Practices ◽

South Sudan ◽

Rice Production ◽

Eating Quality ◽

Rice Varieties ◽

Early Maturity ◽

Access To Credit ◽

Policy Interventions ◽

Poor Access ◽

Cooking And Eating Quality

SUMMARYRice (Oryza sativa L.) is an important staple crop in South Sudan. Current rice consumption in the country is approximately 23,000 mg year−1 of which more than 75% is imported. We investigated farmers’ perceived rice production and productivity constraints and choice of rice ideotypes across rainfed and irrigated rice agro-ecologies in South Sudan to guide breeding and policy interventions. Data were gathered through participatory rural approaches and a formal structured survey involving 136 rice farmers from major rice growing areas of South Sudan. Farmers generally cultivated old rice varieties with low yields varying from 0.4 to 1.6 mg ha−1. Pair-wise ranking based on respondents score indicated yield, early maturity, cooking and eating quality, nutritional importance and drought tolerance as the most desirable traits of rice ideotypes. Further, imported rice varieties were best ranked for their sweet and appealing taste, rich aroma, grain shape and size, swelling capacity and non-stickiness during cooking. Major perceived constraints to rice production were unreliable rainfall, poor access to credit facilities, poor soil and water management practices, poor rice storage facilities, inadequate and poor processing machines as well as limited technical skills in rice production. The results show a need for integration of farmers’ and stakeholders’ preferences in variety development process and the desirability of releasing site-specific rice cultivars given the differences in bio-physical, socio-cultural and farmers preferences across major rice growing areas of South Sudan.

Download Full-text

Yield response of selected irrigated rice varieties to varying N, P and K fertilizer regimes

International Journal of Agricultural Research Innovation and Technology ◽

10.3329/ijarit.v8i2.40554 ◽

2018 ◽

Vol 8 (2) ◽

pp. 42-46

Author(s):

J Chirchir ◽

G Chemining’wa ◽

F Nzuve ◽

R Murori

Keyword(s):

Grain Yield ◽

Crop Production ◽

Management Practices ◽

Block Design ◽

Grain Weight ◽

Yield Response ◽

Fertilizer Treatment ◽

Rice Varieties ◽

K Fertilizer ◽

Fertilizer Regimes

Rice production has been inconsistent in Mwea irrigation scheme due to poor crop management practices and reduced soil fertility. Proper rice nutrition conserves the environment, increases sustained crop production, farmer’s crop yield and profits. A field experiment was conducted at MIAD Center, in Kirinyaga County, during 2016-17 and 2017- 18 to determine the response of rice varieties to different N, P and K fertilizer treatments. The trial was conducted in randomized complete block design, with three replications of 13 N, P and K fertilizer regimes as the main plots treatments and varieties Basmati 370 and BW 196 as the sub-plot treatment. The N kg ha-1: P2O5 kg ha-1: K2O kg ha-1 fertilizer treatment ratios used were: 00:00:00, 60:40:40, 80:60:60, 100:80:80, 60:40:00, 80:60:00, 100:80:00, 60:00:40, 80:00:60, 100:00:80, 00:40:40, 00:60:60 and 00:80:80. Plant height, number of tillers hill-1, panicle length, and grain yield responded positively to fertilizer application, but 1000-grain weight did not. Variety BW 196 recording shorter plants and panicles, more tillers hill-1, higher grain weight and yield than variety Basmati 370. 00:40:40, 00:60:60, 00:80:80, 60:40:00 and the no-fertilizer control treatments recorded the least number of tillers hill-1. 60:40:40, 80:60:60, 100:80:80, fertilizer treatments had longer panicles than the no fertilizer control and 00:40:40. Except for 100:80:00 and 00:40:40, in the first season and 60:40:00, 80:60:00, 100:80:00 in the second season, all the fertilizer regimes increased grain yield relative to the control. The highest grain yield was realized in fertilizer regime 80:60:60, 100:80:80, 80:00:60 and 100:00:80. 80:00:60 is the recommended fertilizer regime. Int. J. Agril. Res. Innov. & Tech. 8 (2): 42-46, December, 2018

Download Full-text

The use of superior varieties to increase rice yields in lowland irrigated of Southeast Sulawesi

E3S Web of Conferences ◽

10.1051/e3sconf/202130601045 ◽

2021 ◽

Vol 306 ◽

pp. 01045

Author(s):

Muhammad Alwi Mustaha ◽

Cipto Nugroho ◽

Samrin ◽

Dian Rahmawati

Keyword(s):

Rice Yield ◽

Growth Yield ◽

Rice Production ◽

Growth And Yield ◽

Disease Dynamics ◽

Rice Varieties ◽

Rice Yields ◽

High Yielding Varieties ◽

Planting Season ◽

Experimental Garden

In order to increase national rice production, the Ministry of Agriculture has released some high yielding varieties. The research was aimed to obtain superior rice varieties adapted to irrigated lowland agro-ecosystems of Southeast Sulawesi. In the first planting season, research was carried out in Olo-oloho Village, Uepai sub district and the second season at the Wawotobi Experimental Garden. Eight varieties were tested, i.e. Inpari 31, Inpari 34, Inpari 35, Inpari 38, Inpari 39, Inpari 41, Inpari 42, Inpari 43 and Mekongga. The variables observed were growth, yield and pest and disease dynamics. The results shows that there was variation growth and yield of the tested varieties. The growth and yield also varied depend on the season. In the first planting season, Inpari 34 and Inpari 43 showed better growth than others, but Inpari 31 resulted the highest rice yield (4.83 t/ha). In the second planting season, only Inpari 34 showed consistent growth, but the variety that had showed the highest yield was Inpari 43 (8.13 t/ha).

Download Full-text