Drought and Agricultural Prices in Mexico

Drought and Agricultural Prices in Mexico | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Drought and Agricultural Prices in Mexico Adrián Hernández-del-Valle, Herbert Kimura, Mario Duran Bustamante, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5827924/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Introduction : This study investigates the relationship between meteorological conditions and agricultural economics in Mexico from 1988 to 2024, with a focus on the impact of drought conditions measured by the Standard Precipitation Index (SPI) on agricultural prices, as indicated by the Agricultural Consumer Price Index (Agro CPI). Methods : Using SPI values over one-month and 24-month periods, we conducted a cluster analysis to examine how precipitation variability affects agricultural prices. The Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) was employed for this analysis due to its efficiency with large datasets and minimal parameter requirements. Results : The analysis revealed a significant negative correlation between prolonged drought periods and Agro CPI, especially in drought-prone clusters. The study found that the effects of drought on agricultural prices are cumulative, with the full impact appearing over successive growing seasons. The 24-month SPI was determined to be a better predictor of the long-term effects of drought on Agro CPI than the one-month SPI. Conclusion : These findings underscore the importance of considering long-term drought impacts in agricultural planning and market forecasting, highlighting critical insights for policymakers and stakeholders involved in agriculture and food security. Drought Standard Precipitation Index Agricultural Consumer Price Index Cluster analysis Climate variability Policy implications Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Plain Language Summary This study explores how droughts affect agricultural prices in Mexico, focusing on how variations in weather conditions influence the cost of agricultural products. By analyzing data from 1988 to 2024, the research highlights that droughts have a cumulative effect on agricultural prices over time. The findings emphasize the importance of considering long-term weather patterns when planning agricultural activities and setting market forecasts. This finding is crucial for farmers, policymakers, and other stakeholders to ensure food security and economic stability. 1. Introduction Among the seventeen Sustainable Development Goals (SDGs) set forth by the United Nations (UN) in 2015, Zero Hunger aims to ensure food security, improve nutrition, eradicate hunger, and advance sustainable agriculture by 2030. Thus, ensuring that people have "access to sufficient, safe, and nutritious food to meet their dietary needs and food preferences for an active and healthy life" is a major concern for all countries (Food and Agriculture Organization of the United Nations (FAO), 1996). However, the world is still far from achieving this goal. For instance, according to a recent report by various international organizations, including the FAO, International Fund for Agricultural Development (IFAD), United Nations International Children’s Emergency Fund (UNICEF), World Food Program (WFP), and World Health Organization (WHO), approximately 30% of the global population —2.4 billion people— still face food insecurity in 2022 and over 3.1 billion people worldwide, or 42% of the population, lack access to a healthy diet. More significantly, a recent estimate also indicates that by 2030, nearly 600 million people will suffer from chronic malnutrition (FAO, IFAD, UNICEF, WFP, and WHO, 2023). Therefore, eradicating hunger and food insecurity, particularly in less developed and developing countries where it is a top priority for public policy, are at the forefront of the global agenda. This was especially evident in the aftermath of the sharp increases in global food prices that occurred in 2007-2008 and 2021-2022. Figure 1 demonstrates this phenomenon. As can be seen, the FAO food price index (solid red line) displays a clear positive trend (dashed red line) between 1990 and 2024. The index doubled in 2011 compared to 1990, reaching almost 145 points in 2022, the highest point in the period 1990-2024 (Warr, 2014; Gregory and Coleman-Jensen, 2013). No single factor alone can explain this increasing trend, yet it is well-known that drought is just one of the significant interconnected factors that contribute to this increase in food prices worldwide (Carbon Brief, 2024; World Economic Forum, 2024). Droughts, which are usually characterized by a lack of precipitation over some time period, can affect economies (McKee et al., 1993). Droughts could be long-term phenomena that persist for months or years as well as short-term events. Thus, they are crucial for all economies because food prices, illustrated and discussed above, are directly affected by the frequency and severity of drought. This occurs through several interconnected mechanisms, such as reduced crop yields, supply-demand imbalances, or increased production costs. For instance, during severe drought conditions, a shortage of water stresses plants and reduces the amount of available agricultural products, which raises the price of agricultural products. Similarly, to offset the effects of drought, farmers may need to make larger investments in irrigation systems, which may result in higher production costs and prices overall. Therefore, as drought-induced costs are often passed on to consumers, droughts lead to an increase in agricultural consumer prices (Salazar et al. 2023; National Integrated Drought Information System (NIDIS), 2024; Ding et al., 2011). Drought is a phenomenon that can occur anywhere, but it is more common in some countries. Mexico is one such country, as it is located in a region that is historically prone to notable fluctuations in precipitation patterns. Therefore, Mexico's high rate of drought recurrence is an essential part of the country's climate, and this phenomenon makes Mexico highly vulnerable to the effects of drought, including rising agricultural consumer prices (Mishra and Singh, 2010; Stahle et al. 2016; Liverman, 1999; Neri and Magana, 2016; Dobler-Morales and Bocco, 2021; Boyd and Ibarraran, 2009). Not only historically but also currently and more likely in the future, drought will be one of the most significant issues discussed in Mexico. A report from the National Aeronautics and Space Administration (NASA), for instance, indicates that the drought in Mexico during 2020–2021 was among the most severe and extensive in recent times. Similarly, the drought in 2023, especially in early summer, became more intense and widespread, and climate change is expected to worsen it in the coming decades (Dobler-Morales and Bocco, 2021; NASA, 2021; NASA, 2024). The Mexican government addresses droughts through immediate and long-term measures. These plans include promoting modern irrigation techniques, enhancing infrastructure and water systems, encouraging the use of water-saving technologies, funding local projects, and developing new technologies. For example, at the institutional level, the Mexican government constructed a set of baseline measures through the National Water Commission (CONAGUA) to reduce the effects of drought, and the Mexican Federal authorities supported this initiative with the development of the National Drought Program (PRONACOSE) (Federman et al., 2014). However, as also emphasized by Dobler-Morales and Bocco (2021), the proposed programs, in general, place more emphasis on mitigating than on preventing disasters, and do not efficiently deal with certain obstacles. Sin Hambre is another example. The Mexican government addresses hunger on a national scale, implementing over 30 programs, with an annual expenditure equivalent to roughly 1% of the Gross Domestic Product (GDP) (Beltran-Silva, 2023; Hansen et al., 2022). In this study, we explore the relationship between drought conditions and agricultural prices in Mexico from 1988 to 2024, particularly considering the potential impact of drought on prices in different clusters. We mainly ask two major questions: (i) What is the relationship between drought and agricultural prices in Mexico and how does it change between different clusters? (ii) Is there any difference between the short- and long-term effects of drought on prices? In order to quantify drought and agricultural prices, we employ the Standard Precipitation Index (SPI) and the Agricultural Consumer Price Index (Agro CPI), respectively. We methodologically apply the Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) clustering algorithm developed by Campello et al. (2013). As also further discussed in the following sections, there are several reasons for choosing this approach. First, this clustering algorithm performs better when dealing with large datasets. As we also work with a large dataset, the HDBSCAN clustering algorithm fully suits our needs. Second, compared with other conventional clustering algorithms, such as K-means or DBSCAN, it requires fewer user-defined parameters, resulting in more significant and efficient clustering from the input data. Third, annual rain in Mexico varies greatly between regions. The south of the country receives almost half of the annual rain, while the north and center of the country, where most economic activity takes place, receive between 5 and 10 percent of the annual rain. This heterogeneity underscores the need for a clustering analysis, such as HDBSCAN. Fourth, clustering analysis is one of the widely used approaches in the literature currently to classify meteorological data and is highly effective in addressing agricultural challenges. We believe our timely study is important and contributes to the existing literature in three respects. First, as discussed previously, drought is a very important topic in Mexico—it has been for a long time and is very likely to continue to be so in the future. Therefore, many scholars paid considerable attention to drought, examined the topic from several perspectives, and produced a rich literature that contains important information on drought in Mexico and its effects on the country (Boyd and Ibarraran, 2009; Arceo-Gomez et al., 2022; Dobler-Morales and Bocco, 2021; Murray-Tortarolo and Jaramillo, 2019; Santacruz-De Leon et al. 2022; Gonzalez et al. 2023a; Gonzalez et al., 2023b). However, to the best of our knowledge, there is no study to investigate the drought-agricultural prices nexus in Mexico with a large dataset by using a data mining technique, especially clustering analysis. This study, for the first time, fills this gap in the existing literature. Second, we use the 24-month and one-month SPI values in this research and compare the results. As further discussed in the following sections, the one-month SPI values are useful for identifying brief variations in precipitation patterns and understanding the immediate effects of precipitation in drought conditions. However, the 24-month SPI provides insightful data regarding anomalous precipitation patterns over a longer period. Incorporating the 24-month SPI into our analysis not only improves our understanding of drought monitoring and management but also gives important information about the effect of extended droughts or periods of excessive precipitation on agricultural prices. 1 Third, our study provides several useful insights for policymakers by shedding light on a trending topic from the past to the present in Mexico. This is important because food consumption depends on economic constraints. Our study exploring the interplay between climate variables and food prices can provide critical insights to regulators, policy, and decision-makers to better promote and coordinate initiatives and highlight the importance of long-term drought considerations in agricultural planning and market forecasting. The understanding of weather-related risks as an underlying cause of food prices, especially in developing countries, such as Mexico, due to their higher share in CPI baskets, may also facilitate the design of policies seeking to reduce inflation (Gonzalez et al., 2023a). The remainder of this paper is organized as follows. In the second section, we review the literature with a specific focus on the socioeconomic impacts of drought in Mexico and the challenges of food security. Next, we introduce the method and dataset used in the study. We then discuss the main outcomes of our data analysis. Finally, we conclude the paper with policy implications and future research directions. 2. Literature Review Drought is a widespread phenomenon, and its detrimental impacts are evident in many segments of society, including the economy, environment, agriculture, or health (Gautier et al., 2016; Vicento-Serrano et al., 2020; Lester et al., 2022; Sena et al., 2014; Ding et al., 2011; Edwards et al. 2018; Barış-Tüzemen and Lyhagen, 2024). Therefore, it has received considerable attention, especially in the academic literature, from many different perspectives. This increased interest is further fueled by the large economic and social costs connected with them that have recently been observed on all continents, including Europe, Asia, Africa, and South America (Mishra and Singh, 2010; Edwards et al., 2018). The economic impacts of droughts on agriculture are multifaceted and there is a growing body of literature in this area (Diersen et al., 2002; Edwards et al. 2009; Horridge et al., 2005). For example, using the panel fixed effect model, Kuwayama et al. (2018) empirically investigated how the drought would affect crop yield and farm income in the US between 2001 and 2013. The empirical analysis outcomes show that there is a statistically significant negative relationship between crop yield and drought, but not when it comes to farm income. A recent study by Barış-Tüzemen and Lyhagen (2024) also finds that there is a statistically significant positive correlation between crop production and rainfall for eight South European countries on the Mediterranean coast for the period of 1996–2016, but this link turns out to be negative if rainfall is excessive. Drought has similar negative effects in Mexico as it does elsewhere in the globe. For example, as demonstrated by Boyd and Ibarraran (2009), due to Mexico's high susceptibility to the effects of drought, the country's economy is affected in several ways, including decreased economic output, a worsening trade balance, a rise in poverty, or a slowdown in economic development. Climate change is expected to make these droughts more frequent in Mexico in the years to come, which may further increase the negative impact of droughts, particularly on crops in the agriculture sector, which typically takes place in rural areas and is responsible for a significant share of economic activity (Dobler-Morales and Bocco, 2021; Boyd and Ibarraran, 2009). The drought-agriculture nexus for Mexico is also extensively studied in the existing literature. For example, Salas-Martinez et al. (2021) examine the relationship between the SPI and the yield of cattle and corn grain in the central zone of the state of Veracruz for the years 1980–2018. The findings unequivocally demonstrate the detrimental effects of drought on cattle production and corn yield in Mexico. The paper also shows that the degree of vulnerability to drought varies significantly amongst regions in Mexico, necessitating targeted interventions and vulnerability assessments. Regional variation in terms of vulnerability and risk to drought is also confirmed by Neri and Magaña (2016). Gonzalez et al. (2023a) investigate the effect of temperature shocks on vegetable prices in Mexico with monthly data at the city level. The results show that temperature has a detrimental effect on vegetable yields and this might be an important driver of higher prices. Boyd and Ibarraran (2009) use a dynamic Computable General Equilibrium (CGE) model to examine the impact of prolonged drought on the Mexican economy and evaluate the potential efficacy of agriculture sector adaptation measures. The findings show that the rural sectors with output in grains, livestock, and forestry suffer the largest production losses. Arceo-Gomez et al. (2020) examine how the 2011 drought affected the well-being of rural households in Mexico, focusing on per capita incomes, poverty, and children's attendance at school. Their findings indicate that droughts have a detrimental influence on the well-being of rural households. The impact of the 2011 drought is also studied by Murray-Tortarolo and Jaramillo (2019), but with an emphasis on livestock population. The findings indicate that the nation's cattle and goat stocks declined by roughly 3% due to drought. In conclusion, the existing literature demonstrates that Mexico's drought has had a significant economic impacts on the country, specifically through the agriculture sector, affecting rural households, livestock populations, economic output, and agricultural products. Data mining techniques, especially cluster analysis, have gained popularity in addressing agricultural challenges in the existing literature as they yield valuable insights across various domains. Chen et al. (2022), for instance, utilize K-means clustering to segment agricultural product clients, emphasizing its potential for targeted marketing. Similarly, Shestakov and Lovchikova (2023) employ a similar clustering approach in their regional agricultural analysis to facilitate the formulation of policies and the development of agribusinesses. Pascucci et al. (2018) highlight the superiority of multivariate functional principal components clustering in yield pattern recognition, promoting sustainable field management. Prakash et al. (2021) propose an efficient initial cluster center estimation method, enhancing clustering efficiency in agricultural databases. Simion et al. (2022) explore agglomerative clustering to improve farmers' competitiveness. In drought impact research, Liu et al. (2021) show the effectiveness of HDBSCAN clustering in meteorological data classification. In conclusion, the reviewed literature underscores the significance of clustering methodologies in addressing agricultural challenges and highlights the urgent need for comprehensive approaches to manage drought impacts and ensure food security. Integrated strategies informed by data insights are vital for sustainable agricultural development and mitigating the impacts of drought on food systems globally. 3. Methodology HDBSCAN is a clustering algorithm that stands out for its superior performance compared to other methods, such as OPTICS or AUTO-HDS clustering, especially when dealing with large datasets containing a substantial number of objects (Campello et al., 2013). The algorithm begins by computing a core distance, denoted as (dcore), for each data point (Xp), representing the distance to its nearest neighbor among all data points or vertices (X = {x1,…, xn}). An ε-core object is then defined as an element of (X) that satisfies (ε ≥ dcore (Xp)). These core objects play a crucial role in identifying the density-based structure of the clusters. One of the key strengths of HDBSCAN lies in its ability to establish reachability between distances and vertices through a more efficient hierarchical simplification process. This is achieved by computing the reachability distance, denoted as (dreachability), defined as the maximum core distance among two data points (Xp) and (Xq), as well as the distance between them (dcore(Xp, Xq)). Mathematically, this can be expressed as dreachability = max{dcore(Xp), dcore(Xq), dcore(Xp, Xq)}, where (Xp) and (Xq) are the elements of (X). By incorporating reachability into the clustering process, HDBSCAN generates more robust and significant clusters from the input data. Unlike many other clustering algorithms, HDBSCAN offers the advantage of requiring fewer parameters to be defined by the user. This reduction in parameterization simplifies the process of applying the algorithm to real-world datasets and makes it more accessible to users with varying levels of expertise. In traditional clustering algorithms like K-means or DBSCAN, users typically need to specify parameters such as the number of clusters (k) or the maximum distance between points (ε). In HDBSCAN, the only parameter that users can set is the minimum cluster size, and even this can be set by the algorithm. HDBSCAN can provide an initial estimate of clusters without any user-specified parameters. This parameter determines the minimum number of points required for a cluster to be considered valid. By minimizing the number of user-defined parameters, HDBSCAN streamlines the clustering process and reduces the need for trial-and-error experimentation to find optimal parameter values. This simplicity and ease of use make HDBSCAN particularly well-suited for applications where a quick and efficient clustering solution is desired, without the burden of extensive parameter tuning. Andrzejuk (2018), for instance, identifies clusters in the agricultural emissions of the OECD countries by comparing the K-means algorithm with HDBSCAN. The paper emphasizes the advantages of HDBSCAN over traditional K-means, noting that the latter splits the data and assigns all of the information to a cluster, potentially leading to forced findings, while the former eliminates data that is unrelated to any cluster. The results further show that agriculture was comparable to Mexico and Australia in the initial analysis with K-means. However, after applying HDBSCAN, Mexico was omitted from the cluster and was instead categorized as an outlier with Japan, the Netherlands, New Zealand, Poland, the Republic of Korea, Spain, Türkiye, and the UK. Notwithstanding its benefits, HDBSCAN possesses many significant limitations that warrant consideration. The method is proficient at identifying clusters of diverse densities; however, it may encounter difficulties with datasets featuring clusters of markedly varied densities, thereby overlooking crucial groups in sparse data regions. Moreover, while HDBSCAN minimizes the necessity for parameter adjustment relative to other clustering methods, its susceptibility to the minimum cluster size parameter can still influence outcomes, especially when addressing datasets with clusters of significantly diverse sizes. Additionally, the algorithm's computational complexity may pose a challenge with extensive datasets, as it necessitates the computing of pairwise distances between points, potentially resulting in longer processing times relative to more straightforward clustering techniques such as K-means. That said, HDBSCAN demonstrates a notable advantage in processing information or data with noise, making it particularly effective for studies like ours, where variations in droughts and atypical rains in recent years introduce significant noise. Its ability to handle such noisy data establishes it as an appropriate method for the type of study we developed. 4. Data Analysis We use the one-month SPI data, first developed by McKee et al. (1993), to quantify drought in Mexico. The one-month SPI offers data on precipitation anomalies in relation to the long-term average for a particular month (Guttman, 1999). A positive one-month SPI shows that the amount of precipitation for the given month is higher than the average amount of precipitation over a lengthy period. This indicates the presence of above-average levels of moisture. A negative one-month SPI shows that the amount of precipitation during the given month is below the long-term average. This indicates arid conditions. An SPI score that is close to zero indicates that the precipitation for the specified month is quite similar to the long-term average, suggesting typical precipitation conditions. Greater positive or negative SPI values imply more substantial deviations from the average precipitation. In general, one-month SPI values are valuable for detecting brief deviations in precipitation patterns and comprehending the immediate consequences of precipitation in drought or wet situations. These numbers are included in a larger set of SPI values that are generated for various periods to evaluate the seriousness of drought conditions. On the other hand, the 24-month SPI offers valuable information about unusual precipitation patterns over a more extended timeframe. This is the way they operate (World Meteorological Organization (WMO), 2012): Long-Term Assessment: The 24-month SPI provides a more comprehensive view by considering variations in precipitation over a two-year timeframe. This facilitates a more thorough understanding of precipitation patterns and their influence on drought or wet conditions. Cumulative Effects: In contrast to the one-month SPI that only considers short-term changes, the 24-month SPI measures the combined impact of precipitation anomalies over a longer period. This is especially beneficial for evaluating long-lasting droughts or prolonged periods of excessive rainfall. Robustness: The 24-month SPI offers a more reliable evaluation of precipitation patterns by examining a longer period. This reduces the impact of temporary changes and provides a deeper understanding of long-term climate trends. Severity Assessment: The 24-month SPI indicates the severity of precipitation anomalies, similar to the one-month SPI. Positive values imply moisture levels that are higher than usual throughout the two years, while negative values suggest below-average precipitation, indicating dry conditions. Supplementary study: While the one-month SPI is useful for identifying short-term variances, the 24-month SPI enhances this study by offering insights into longer-term climate variability. Collectively, they provide a thorough understanding of precipitation patterns and their ramifications for monitoring and administering drought. Overall, the 24-month SPI is a useful tool for evaluating extended rainfall patterns and providing vital information on prolonged periods of drought or excessive precipitation. It complements the analysis provided by shorter-term SPI. 2 Mexico's National Consumer Price Index, —or ­­­Índice Nacional de Precios al Consumidor (INPC)—, which measures changes in the price level of a market basket of consumer goods and services purchased by households, is estimated and calculated by the Instituto Nacional de Estadística y Geografía (INEGI), which is the organization responsible for measuring and reporting on inflation in the country. The INPC has two components: core and non-core, which include more stable and volatile price components, respectively. This breakdown allows policymakers and analysts to distinguish between more stable price trends and those subject to short-term fluctuations, aiding in the formulation of economic policies and inflation expectations. The non-core agricultural prices (No Subyacente Agropecuarios) are of particular interest because they tend to be more volatile than the overall index. 5. Results Before starting our HDBSCAN clustering analysis, we first converted our biweekly Mexican Agro CPI data to the monthly form and built a compatible dataset with the one- and 24-month SPIs for all 418 weather stations in the Mexican Conagua System. We then eliminated missing values (NaN) by using Multiple Imputation of Chained Equations (MICE). Next, we performed the SelectKBest feature selection method to retain only the 20 regions with the highest influence on Agro CPI between January 1988 and February 2024. We then conducted a cluster analysis with HDBSCAN. 5.1. Short-Term Drought and Agro CPI Figure 1 illustrates the results of the cluster analysis. As can be seen, we identify three clusters: 1 (normal), 0, and -1 (anomalous). Interestingly, the 0 and -1 clusters only represent the period from December 2021 to the present date, presenting evidence of a change in short-term drought behavior. In Cluster 1 (the teal region), the average SPI is 0.11113, indicating wet conditions or excess precipitation. The average geometric growth of Agro CPI during this period is 22.69. In Cluster 0 (the yellow region), the average SPI changes to -0.4765, indicating dry conditions or drought. The geometric average growth of Agro CPI in Cluster 0 is 139.2. Sixteen out of the top 20 weather regions identified in the SelectKBest feature selection show a negative correlation with Agro CPI. The negative correlations between SPI and Agro CPI could indicate that as SPI decreases (indicating drier conditions), CPI tends to increase. This could suggest that drought conditions lead to lower agricultural yields, higher food prices, and thus higher Agro CPI. Figure 2 maps these regions in Cluster 0, and Figure 4 presents a heat map of the three clusters. Finally, in Cluster -1, the average SPI is -0.3, indicating dry conditions or drought. The geometric average growth of Agro CPI is 138.6. Six regions present a negative correlation with Agro CPI, as can be seen in Figure 3. In HDBSCAN, clusters are formed by grouping points that lie close together in the high-density regions of the data space. It is also equally important to note that the determination of Cluster -1 occurred during the construction of the density-based clustering hierarchy. Points that are not sufficiently close to any core points or do not meet the minimum cluster size criterion are assigned to the noise cluster. This cluster represents the outliers or noise points in the dataset, which do not belong to any dense regions or meaningful clusters. Consequently, HDBSCAN's inability to elucidate the dynamics of droughts during -1 periods underscores their enigmatic nature, extending beyond conventional explanations. Judging by the average SPI, drought conditions have been slightly less severe in Cluster -1, reflecting a slightly smaller growth of Agro CPI. However, Figure 4 shows that negative correlations prevail in clusters 0 and -1. The average correlation between Regional SPI and Agro CPI is 17% in Cluster 1, -9% in Cluster 0, and -8% in Cluster -1. 5.2. Long-Term Drought and Agro CPI This subsection presents an in-depth analysis of the relationship between the Agricultural CPI and the 24-month SPI in Mexico. Through a comparison of these two indicators, this study highlights the significant impact of prolonged drought conditions on agricultural pricing and the broader economic implications for consumers and producers. Figure 5 illustrates the results of the cluster analysis with the 24-month SPI. The clusters, shown in Figures 1 and 5, exhibit striking similarities. For example, Figure 5 shows an onset of anomalous behavior from September 2021 —three months earlier than the 1-month SPI analysis— to February 2024 (the end of our sample period). The key distinction between the comparison of the one-month and 24-month SPIs with Agro CPI lies in the considerably larger correlations observed in the 24-month analysis. Agro CPI experiences substantial growth, particularly evident in Clusters -1 and 0, corresponding to the anomalous sample observed in the 1-month SPI comparison. Table 1. Comparison of Clusters: Average 24-month-SPI and Geometric Average Growth of Agro CPI Clusters Average SPI Geometric average growth of Agro CPI -1 -1.31 137.17 1 -0.32 23.42 0 -1.26 141.72 Source: Own estimation with data from INEGI and Conagua Table 1 presents a comparison of clusters, highlighting the average 24-month SPI and geometric average growth of Agro CPI. As reported, the growth rates in Table 1 also coincide with those observed in the 1-month-SPI analysis. For example, Agro CPI shows substantial growth, especially in Clusters -1 and 0, aligning with the unusual samples reported in the 1-month SPI comparison. However, compared to the one-month SPI results, the results shown in Figure 6 indicate that the correlation between the 24-month SPIs and Agro CPI is much larger than those observed with the one-month SPI, indicating that a persistent lack of rainfall over longer periods is more influential in determining agricultural prices in Mexico in the last part of the sample. Therefore, the effects of drought on production are observed to be delayed, as extended periods of water scarcity negatively impact agricultural yields throughout successive growing seasons. Prolonged droughts can result in the loss of crucial water resources, worsening the impact on crop yields and livestock productivity, and gradually translating to increased food costs. When we concentrate on Cluster -1 and compare the one-month and 24-month correlation between regions and the Agricultural CPI, we obtain the following inferences. In the 1-month analysis (Figure 1), the correlation for Cluster -1 ranges from -.50 to .32, whereas in the 24-month analysis (Figure 5), the correlation for Cluster -1 ranges from around -1.00 to -.40. Therefore, by focusing only on Cluster -1, we can clearly see that the correlations are substantially higher (more negative) in the 24-month analysis, compared to the one-month analysis. The maximum negative correlation in the one-month analysis for Cluster -1 is -.50; however, in the 24-month analysis, the maximum negative correlation for Cluster -1 is around -1.00, which is twice as extreme. This indicates that for the regions/areas in Cluster -1, which likely experienced more severe or prolonged drought conditions, the 24-month SPI has a much stronger negative correlation with Agricultural CPI increases than the one-month SPI. The longer 24-month period is better able to capture the full delayed and compounding effects of drought on agricultural pricing in these severely impacted areas. Thus, especially for Cluster -1, the finding clearly demonstrates the importance of considering longer drought timescales when analyzing pricing impacts. A prolonged drought period as identified in this analysis could have several effects: Delayed impact of drought on production: Agricultural productivity is particularly vulnerable to extended drought conditions that span multiple growing seasons. Such long-term water deficits do not immediately affect crop yields, which may explain the weaker correlations with the one-month SPIs. Depletion of water resources: Long-term droughts can lead to significant depletion of key water resources, including reservoirs, groundwater, and soil moisture. As these resources diminish, the consequences for crop yields and livestock productivity become more pronounced, directly influencing food prices. Cumulative drought effects: The adverse effects of drought on agricultural systems tend to build up over longer periods. This accumulation of damage to crops, pastures, livestock, supply chains, and infrastructure has far-reaching impacts on food prices. Through the comparative analysis of 24-month SPIs and Agro CPI, this section demonstrates the complex interplay between climate events and economic outcomes in Mexico's agricultural sector. Understanding this relationship is critical for policymakers and stakeholders to create resilient agricultural practices and safeguard the economy against the consequences of climate variability. 6. Conclusion In this paper, we examine the relationship between the one-month and 24-month SPI, as documented in Mexico's Conagua System, and the Agro CPI in Mexico. Our results show that the relationships between the Agro CPI and the 24-month SPI are significantly stronger than those between the Agro CPI and the one-month SPI. The overwhelming data indicates a possible long-term change in Mexico's precipitation patterns, which might have significant effects on agricultural practices, food security, and overall CPI trends. Some key takeaways from the paper are as follows: Multiyear drought conditions have a significant and long-lasting effect on agriculture that goes beyond the immediate growing seasons. We find significant increases in the Agro CPI in persistently dry areas, especially in Clusters -1 and 0. This suggests a clear link between longer droughts and higher food prices. Our findings underscore the reality that the effects of drought on agriculture are not immediate but have a delayed onset, escalating over successive periods of water scarcity. Our paper suggests the following implications for policy and practice: The evident correlation between extended drought and Agro CPI underscores the need for proactive strategies. Agricultural policies must incorporate climate forecasts, mitigate risks with resilient farming techniques, and effectively manage water resources. Implementing agricultural practices tailored to withstand climatic fluctuations can reduce the adverse economic outcomes associated with drought. This may include investing in drought-resistant crop varieties and advanced irrigation systems. Farmers, policymakers, and stakeholders are encouraged to utilize long-term climate predictions to inform planting decisions and resource allocation, thereby enhancing agricultural productivity and economic resilience. Further research into the potential of innovative agricultural technologies to buffer the impact of long-term climatic changes on crop yields and Agro CPI can enhance our findings. Furthermore, future research may assess the feasibility and effectiveness of policy interventions aimed at supporting farmers during extended periods of drought, including financial instruments like crop insurance and water rights trading. Finally, these conclusions serve as a clarion call for the integration of climate adaptability into the core of agricultural planning and policy-making. It is imperative that we harness our collective knowledge and resources to safeguard food security and economic vitality in the face of mounting climate variability. Declarations Acknowledgments : We thank our colleagues for their invaluable input and feedback during the research and writing process. Special thanks to the National Water Commission (CONAGUA) and the Instituto Nacional de Estadística y Geografía (INEGI) for providing essential data. Conflicts of Interest : The authors declare no conflicts of interest. Funding Declarations : This research was supported by the National Council of Humanities, Sciences, and Technologies (Conahcyt). Data Availability Statement The data that support the findings of this study are publicly available from the following sources: Standard Precipitation Index (SPI) data for Mexico were obtained from the National Meteorological Service (SMN) of the National Water Commission (CONAGUA) through their drought monitoring system, accessible at: https://smn.conagua.gob.mx/es/climatologia/monitor-de-sequia/spi Agricultural Consumer Price Index (Agro CPI) data were sourced from the National Institute of Statistics and Geography (INEGI) through their economic indicators database, accessible at: https://inegi.org.mx/app/indicadores/?tm=0&t=10000215#D1000021 The processed datasets and analysis code used in this study are available from the corresponding author upon reasonable request. References Andrzejuk, A. (2018). Classification of Agricultural Emissions Among OECD Countries with Unsupervised Techniques. Zeszyty Naukowe SGGW W Warszawie - Problemy Rolnictwa Światowego, 18(4), 80–91. https://doi.org/10.22630/PRS.2018.18.4.99 Arceo-Gómez, E. O., Hernández-Cortés, D., & López-Feldman, A. (2020). Droughts and rural households' wellbeing: evidence from Mexico. Climatic Change, 10. Barış-Tüzemen, Ö. & Lyhagen, J. (2024). Revisiting the role of climate change on crop production: Evidence from Mediterranean countries. Environment, Development and Sustainability. https://doi.org/10.1007/s10668-024-04991-x Beltran-Silva, F. (2023). Fighting against hunger: A country-wide intervention and its impact on birth outcomes. World Development 165, 106202. Birthal, P. S., Roy, D., Negi, D. S., Khan, T., & Aggarwal, P. K. (2015). Farm-level impacts of climate change on agriculture in India. Journal of Agricultural Economics, 66(1), 1-26. Boyd, R., & Ibarrarán, M. (2009). Extreme climate events and adaptation: an exploratory analysis of drought in Mexico. Environment and Development Economics, 74. Bouaziz, M., Medhioub, E. & Csaplovisc, E. (2021). A machine learning model for drought tracking and forecasting using remote precipitation data and a standardized precipitation index from arid regions. Journal of Arid Environments, 189, 104478. Chen, T. C., Alazzawi, F. J. I., Mavaluru, D., Mahmudiono, T., Enina, Y., Chupradit, S., Ahmed, A. A. A., Syed, M. H., Ismael, A. M., & Miethlich, B. (2022). Application of data mining methods in grouping agricultural product customers. Mathematical Problems in Engineering, 2022, Article 9876543. Diersen, M.A., Taylor, G. & May, A. (2002). Direct and indirect effects of drought on South Dakota’s economy. Economics Commentator, 432, 26. Dijk, A. I., Renzullo, L. J., & McVicar, T. R. (2013). Global analysis of seasonal streamflow predictability using an ensemble prediction system and observations from 6192 small catchments worldwide. Water Resources Research, 49(3), 1276-1291. Ding, Y., Hayes, M.J. & Widhalm, M. (2011). Measuring economic impacts of drought: a review and discussion. Disaster Prevention and Management, 20(4), 434-446. https://doi.org/10.1108/09653561111161752 Dobler-Morales, C. & Bocco, G. (2021). Social and environmental dimensions of drought in Mexico: An integrative review. International Journal of Disaster Risk Reduction, 55, 10267. Edwards, B., Gray, M. & Hunter, B. (2009). A sunburnt country: the economic and financial impact of drought on rural and regional families in Australia in an era of climate change. Australian Journal of Labour Economics, 12, 109–31. Edwards, B., Gray, M. & Hunter, B. (2018). The social and economic impacts of drought. Australian Journal of Social Issues, 54(1), 22-31. FAO (1996). Declaration on world food security and World Food Summit plan of action. World Food Summit 13-17 November 1996, Rome, Italy. FAO, IFAD, UNICEF, WFP and WHO (2023). The State of Food Security and Nutrition in the World 2023. Urbanization, agrifood systems transformation and healthy diets across the rural–urban continuum. Rome. https://doi.org/10.4060/cc3017en FAO (2024). World Food Situation. https://www.fao.org/worldfoodsituation/foodpricesindex/en/ Federman, D. K., Cortes, F. I. A. & Perez, M. L. (2014). Constructing a framework or National Drought Policy: The way forward in Mexico. Weather and Climate Extremes, 3, 90-94. Guttman, N. B. (1999). Accepting the standardized precipitation index: A calculation algorithm. Journal of the American Water Resources Association, 35(2), 311-322. Hansen, J., List, G, Downs, S., Carr, E., Diro, R., Baethgen, W., Kruczkiewicz, W., Braun, M., Furlow, J., Wals, K. & Magima, N. (2022). Impact pathways from climate services to SDG2 (“zero hunger”): A synthesis of evidence. Climate Risk Management, 35. Hijioka, Y., Ikeda, T., Oyoshi, K., & Hirota, T. (2018). Climate change impacts on crop yields and water use in Australia. Climate Dynamics, 51(5-6), 2289-2303. Horridge, M., Madden, J. & Wittwer, G. (2005). The impacts of the 2002–03 drought on Australia. Journal of Policy Modeling, 27(3), 285–308. Gautier, D., Denis, D. & Locatelli, B. (2016). Impacts of drought and responses of rural populations in West Africa: a systematic review. WIREs Climate Change, 7(5), 666-681. Gregory, C. A. & Coleman-Jensen, A. (2013). Do high food prices increase food insecurity in the United States?, Applied Economic Perspectives and Policy, 35(4), 679-707. Kuwayama, Y., Ikeda, T., & Hijioka, Y. (2018). Drought effects assessment on U.S. agriculture: Empirical evidence of panel data models with county-level data. PLOS ONE, 13(5), e0197650. Lester, L., Flatau, P. & Kyron, M. (2022). Understanding the social impacts of drught. Perth: The University of Western Australia. Liu, J., Sun, S., & Chen, C. (2021). Big data Analysis of Regional Meteorological Observation Based: on Hierarchical Density Clustering Algorithm HDBSCAN. In 2021 2nd International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) (pp. 111-116). IEEE. Liverman, D. M. (2016). Vulnerability and Adaptation to Drought in Mexico. National Resources Journal, 39, 99-115. Lubova, T. N., Salimova, G., Nigmatullina, G., Islamgulov, D., & Sharafutdinov, A. (2023). Modeling the development of agricultural production by cluster analysis. E3S Web of Conferences, 403, 01044. Madadgar, S., AghaKouchak, A., & Farahmand, A. (2017). Multivariate drought characteristics across climatic zones. Scientific Reports, 7(1), 1785. McKee, T. B., Doesken, N. J. & Kleist, J. (1993). The relationship of drought frequency and duration of time scales. Eighth Conference on Applied Climatology, American Meteorological Society, Jan17-23, 1993, Anaheim CA, 179-186. Mishra, A. K., Jha, M. K., & Hansen, J. W. (2015). Remote sensing and hydrological models for drought assessment in northeast India. Journal of Hydrology, 524, 354-365. Mishra, A. K. & Singh, V. P. (2010). A review of drought concepts. Journal of Hydrology, 391(1-2), 202-216. Murray-Tortarolo, G. N., & Jaramillo, V. J. (2019). The impact of extreme weather events on livestock populations: the case of the 2011 drought in Mexico. Climatic Change, 24. NASA (2021). Widespread Drought in Mexico. Retrieved from https://earthobservatory.nasa.gov/images/148270/widespread-drought-in-mexico. NASA (2024). Drought Parches Mexico. Retrieved from https://earthobservatory.nasa.gov/images/152908/drought-parches-mexico. Neri, C., & Magaña, V. (2016). Estimation of Vulnerability and Risk to Meteorological Drought in Mexico. Weather, Climate, and Society, 45. NIDIS (2024). Retrieved from https://www.drought.gov/sectors/agriculture#:~:text=The%20primary%20direct%20economic%20impact,through%20government%20disaster%20assistance%20programs. Pascucci, S., Carfora, M., Palombo, A., Pignatti, S., Casa, R., Pepe, M., & Castaldi, F. (2018). A comparison between standard and functional clustering methodologies: Application to agricultural fields for yield pattern assessment. Remote Sensing, 10(1), 79. Prakash, K. L. N. C., Narayana, G. S., Ansari, M. D., & Gunjan, V. K. (2021). Instantaneous approach for evaluating the initial centers in the agricultural databases using K-means clustering algorithm. Journal of Mobile Multimedia, 17(3), 297-320. Putsenteilo, P., & Kostetskyi, Y. (2020). Using cluster analysis methods to forecast the investment development of agricultural sector enterprises. Innovative Economy, 16(83), 8-18. Salas-Martínez, F., Valdés-Rodríguez, O. A., Palacios-Wassenaar, O., & Márquez-Grajales, A. (2021). Analysis of the Evolution of Drought through SPI and Its Relationship with the Agricultural Sector in the Central Zone of the State of Veracruz, Mexico. Agronomy, 5. Salazar, C., Acuna-Duarte, A. & Gil, J. M. (2023). Drought shocks and price adjustments in local food markets in Chile: Do product quality and marketing channel matter?. Agricultural Economics, 54(3), 349-363. Santacruz-De León, G., Morán-Ramírez, J., & Ramos-Leal, J. A. (2022). Impact of Drought and Groundwater Quality on Agriculture in a Semi-Arid Zone of Mexico. Agriculture, 3. Sena, A., Barcellos, C., Freitas, C. & Corvalan, C. (2014). Managing the Health Impacts of Drought in Brazil. International Journal of Environmental Research and Public Health, 11, 10737-10751. https://doi.org/10.3390/ijerph111010737 Shestakov, R., & Lovchikova, E. I. (2023). Clustering of regions using basic agricultural and economic criteria. Economy of Regions, 19(1), 320-334. Simion, G., Căleanu, C. D., Bucos M. and B. Drăgulescu (2022). Grouping farmers using agglomerative clustering on data generated from statistics, 2022 International Symposium on Electronics and Telecommunications (ISETC), Timisoara, Romania, 2022, pp. 1-4, doi: 10.1109/ISETC56213.2022.10010248. Stahle, D. W., Cook, E. R., Burnette, D. J., Villanueva, J., Cerano, J., Burns, J. N., Griffin, D., Cook, B. I., Acuña, R., Torbenson, M. C. A., Szejner, P. & Howard, I. M. (2016). The Mexican Drought Atlas: Tree-ring reconstructions of the soil moisture balance during the late pre-Hispanic, colonial, and modern eras. Quaternary Science Reviews, 149, 34-60. Tiwari, M., & Misra, B. (2011). Application of cluster analysis in agriculture–A review article. International Journal of Computer Applications, 36(5), 40-48. Wang, D., Wang, R., Wang, Y., & Zha, S. (2006). Research on clustering for agriculture information. In 2006 6th World Congress on Intelligent Control and Automation (pp. 7277-7281). IEEE. Warr, P. (2014). Food Insecurity and its determinants. The Australian Journal of Agricultural and Resource Economics, 58(4), 519-537. WMO (2012). Standardized Precipitation Index User Guide. (WMO-No. 1090), Geneva. Xing, J. (2011). Application of fuzzy dynamic cluster analysis in the division of agricultural economic system. Journal of Anhui Agricultural Sciences, 39(22), 13540-13542. Vicento-Serrano, S. M., Quiring, S. M., Pena-Gallardo, M., Yuan, S. & Dominguez-Castro, F. (2020). A review of environmental droughts: increased risk under global warming. Earth-Science Reviews, 201, 102953. Footnotes Some papers forecast SPI to understand drought conditions in Mexico (Magallanes-Quintanar et al., 2024; Esquivel-Saenz et al., 2024). For further information about the Mexico’s SPI, please visit the following website: https://smn.conagua.gob.mx/es/climatologia/monitor-de-sequia/spi Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5827924","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":420749620,"identity":"b29942c3-4513-463f-a129-14c3ea6f458e","order_by":0,"name":"Adrián Hernández-del-Valle","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5ElEQVRIiWNgGAWjYDACCQYDhgcMDDJ8DMwHQFwZ4rQkMDDwsDGwJYC4PKRo4TEA8Qlr4Z/dvPFBYpsdDxt7z+dXN2oseBjYDx/dgNeSO8eKDRLbknnYeM5us845BnQYT1raDbzW3Mgxk0g4w8zDJpG7zTiHDahFgscMrxb5GznmPxLO1POwyb95ZpzzjwgtBkBbGBIqDgNt4WF+nNtGhBbDG2nFEgkVx4F+STNjzu2TADHw+0XuRvLGDx8MquX42Q8//pzzrQ7EOIbf+0iATQJMEqscBJg/kKJ6FIyCUTAKRg4AAJgjQVTXR8UlAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0003-0963-2357","institution":"Instituto Politecnico Nacional","correspondingAuthor":true,"prefix":"","firstName":"Adrián","middleName":"","lastName":"Hernández-del-Valle","suffix":""},{"id":420749621,"identity":"9374f32e-7e1f-4533-a674-0d4496ca22b1","order_by":1,"name":"Herbert Kimura","email":"","orcid":"","institution":"University of Brasilia: Universidade de Brasilia","correspondingAuthor":false,"prefix":"","firstName":"Herbert","middleName":"","lastName":"Kimura","suffix":""},{"id":420749622,"identity":"e65aa362-a30d-4e15-b62c-095fde166256","order_by":2,"name":"Mario Duran Bustamante","email":"","orcid":"","institution":"Instituto Politécnico Nacional: Instituto Politecnico Nacional","correspondingAuthor":false,"prefix":"","firstName":"Mario","middleName":"Duran","lastName":"Bustamante","suffix":""},{"id":420749623,"identity":"056e432b-7766-4ac8-b2f6-2b713ebe0b4c","order_by":3,"name":"Sedat Alataş","email":"","orcid":"","institution":"Aydın Adnan Menderes Üniversitesi Kütüphane ve Dokümantasyon Daire Başkanlığı: Aydin Adnan Menderes Universitesi Kutuphane ve Dokumantasyon Daire Baskanligi","correspondingAuthor":false,"prefix":"","firstName":"Sedat","middleName":"","lastName":"Alataş","suffix":""}],"badges":[],"createdAt":"2025-01-14 14:31:05","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5827924/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5827924/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":77253483,"identity":"68ca0df8-339f-413c-8a04-bd0084fbf947","added_by":"auto","created_at":"2025-02-26 16:40:50","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":73380,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal Food Prices (1990-2024)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSource:\u003c/strong\u003e FAO (2024)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/9c64eede62c476aea5ba2f52.png"},{"id":77252708,"identity":"1d563977-2b21-4ba8-b0d2-9421b1e8fa83","added_by":"auto","created_at":"2025-02-26 16:32:50","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":53130,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 1. Cluster analysis of Agro CPI related to one-month drought conditions\u003c/p\u003e\n\u003cp\u003eSource: Own estimation with data from INEGI and Conagua\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/22e6b6e16de80aa96a7861e4.png"},{"id":77253484,"identity":"20f63557-1166-4c2a-80af-dabf9189a8ca","added_by":"auto","created_at":"2025-02-26 16:40:50","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":331454,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 2. Drought regions in Cluster 0\u003c/p\u003e\n\u003cp\u003eSource: Own estimation with data from INEGI and Conagua\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/3525de5b5b92cfd9950a699d.png"},{"id":77252717,"identity":"ce22084c-04f4-482d-8bec-4168051b9080","added_by":"auto","created_at":"2025-02-26 16:32:50","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":301626,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3. Drought regions in Cluster -1\u003c/p\u003e\n\u003cp\u003eSource: Own estimation with data from INEGI and Conagua\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/066277459f70926c2e2cef30.png"},{"id":77252707,"identity":"f9f754f5-5841-45f8-aa44-04134377408b","added_by":"auto","created_at":"2025-02-26 16:32:50","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":66990,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 4. Correlation between Regional 1-month-SPIs and Agro CPI across Clusters\u003c/p\u003e\n\u003cp\u003eSource: Own estimation with data from INEGI and Conagua\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/7f7f675dc55850bbf179c104.png"},{"id":77252711,"identity":"dc84b373-4c9c-4640-9d7c-0ffa5faae124","added_by":"auto","created_at":"2025-02-26 16:32:50","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":65256,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 5. Cluster analysis of Agro CPI related to 24-month drought conditions\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSource: Own estimation with data from INEGI and Conagua\u003c/em\u003e\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/462fdef89b73b5020ca172d0.png"},{"id":77252715,"identity":"8e8e9602-31b1-4fb8-a89a-f9ff21c13374","added_by":"auto","created_at":"2025-02-26 16:32:50","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":69157,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 6. Correlation between Regional 24-month-SPIs and Agro CPI across Clusters\u003c/p\u003e\n\u003cp\u003eSource: Own estimation with data from INEGI and Conagua\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/fe34e4861f71b247300e8a18.png"},{"id":77253753,"identity":"0f1b417c-0546-466e-849c-d2aea4997a42","added_by":"auto","created_at":"2025-02-26 16:48:50","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1311368,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5827924/v1/88c126ba-8c42-467d-ae12-c68e5a0e466e.pdf"}],"financialInterests":"","formattedTitle":"Drought and Agricultural Prices in Mexico","fulltext":[{"header":"Plain Language Summary","content":"\u003cp\u003eThis study explores how droughts affect agricultural prices in Mexico, focusing on how variations in weather conditions influence the cost of agricultural products. By analyzing data from 1988 to 2024, the research highlights that droughts have a cumulative effect on agricultural prices over time. The findings emphasize the importance of considering long-term weather patterns when planning agricultural activities and setting market forecasts. This finding is crucial for farmers, policymakers, and other stakeholders to ensure food security and economic stability.\u003c/p\u003e"},{"header":"1. Introduction","content":"\u003cp\u003eAmong the seventeen Sustainable Development Goals (SDGs) set forth by the United Nations (UN) in 2015, Zero Hunger aims to ensure food security, improve nutrition, eradicate hunger, and advance sustainable agriculture by 2030. Thus, ensuring that people have \u0026quot;access to sufficient, safe, and nutritious food to meet their dietary needs and food preferences for an active and healthy life\u0026quot; is a major concern for all countries (Food and Agriculture Organization of the United Nations (FAO), 1996). However, the world is still far from achieving this goal. For instance, according to a recent report by various international organizations, including the FAO, International Fund for Agricultural Development (IFAD), United Nations International Children\u0026rsquo;s Emergency Fund (UNICEF), World Food Program (WFP), and World Health Organization (WHO), approximately 30% of the global population \u0026mdash;2.4 billion people\u0026mdash; still face food insecurity in 2022 and over 3.1 billion people worldwide, or 42% of the population, lack access to a healthy diet. More significantly, a recent estimate also indicates that by 2030, nearly 600 million people will suffer from chronic malnutrition (FAO, IFAD, UNICEF, WFP, and WHO, 2023). Therefore, eradicating hunger and food insecurity, particularly in less developed and developing countries where it is a top priority for public policy, are at the forefront of the global agenda. This was especially evident in the aftermath of the sharp increases in global food prices that occurred in 2007-2008 and 2021-2022. Figure 1 demonstrates this phenomenon. As can be seen, the FAO food price index (solid red line) displays a clear positive trend (dashed red line) between 1990 and 2024. The index doubled in 2011 compared to 1990, reaching almost 145 points in 2022, the highest point in the period 1990-2024 (Warr, 2014; Gregory and Coleman-Jensen, 2013). No single factor alone can explain this increasing trend, yet it is well-known that drought is just one of the significant interconnected factors that contribute to this increase in food prices worldwide (Carbon Brief, 2024; World Economic Forum, 2024).\u003c/p\u003e\n\u003cp\u003eDroughts, which are usually characterized by a lack of precipitation over some time period, can affect economies (McKee et al., 1993). Droughts could be long-term phenomena that persist for months or years as well as short-term events. Thus, they are crucial for all economies because food prices, illustrated and discussed above, are directly affected by the frequency and severity of drought. This occurs through several interconnected mechanisms, such as reduced crop yields, supply-demand imbalances, or increased production costs. For instance, during severe drought conditions, a shortage of water stresses plants and reduces the amount of available agricultural products, which raises the price of agricultural products. Similarly, to offset the effects of drought, farmers may need to make larger investments in irrigation systems, which may result in higher production costs and prices overall. Therefore, as drought-induced costs are often passed on to consumers, droughts lead to an increase in agricultural consumer prices (Salazar et al. 2023; National Integrated Drought Information System (NIDIS), 2024; Ding et al., 2011).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDrought is a phenomenon that can occur anywhere, but it is more common in some countries. Mexico is one such country, as it is located in a region that is historically prone to notable fluctuations in precipitation patterns. \u0026nbsp;Therefore, Mexico\u0026apos;s high rate of drought recurrence is an essential part of the country\u0026apos;s climate, and this phenomenon makes Mexico highly vulnerable to the effects of drought, including rising agricultural consumer prices (Mishra and Singh, 2010; Stahle et al. 2016; Liverman, 1999; Neri and Magana, 2016; Dobler-Morales and Bocco, 2021; Boyd and Ibarraran, 2009).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNot only historically but also currently and more likely in the future, drought will be one of the most significant issues discussed in Mexico. A report from the National Aeronautics and Space Administration (NASA), for instance, indicates that the drought in Mexico during 2020\u0026ndash;2021 was among the most severe and extensive in recent times. Similarly, the drought in 2023, especially in early summer, became more intense and widespread, and climate change is expected to worsen it in the coming decades (Dobler-Morales and Bocco, 2021; NASA, 2021; NASA, 2024).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Mexican government addresses droughts through immediate and long-term measures. These plans include promoting modern irrigation techniques, enhancing infrastructure and water systems, encouraging the use of water-saving technologies, funding local projects, and developing new technologies. For example, at the institutional level, the Mexican government constructed a set of baseline measures through the National Water Commission (CONAGUA) to reduce the effects of drought, and the Mexican Federal authorities supported this initiative with the development of the National Drought Program (PRONACOSE) (Federman et al., 2014). However, as also emphasized by Dobler-Morales and Bocco (2021), the proposed programs, in general, place more emphasis on mitigating than on preventing disasters, and do not efficiently deal with certain obstacles. Sin Hambre is another example. The Mexican government addresses hunger on a national scale, implementing over 30 programs, with an annual expenditure equivalent to roughly 1% of the Gross Domestic Product (GDP) (Beltran-Silva, 2023; Hansen et al., 2022).\u003c/p\u003e\n\u003cp\u003eIn this study, we explore the relationship between drought conditions and agricultural prices in Mexico from 1988 to 2024, particularly considering the potential impact of drought on prices in different clusters. We mainly ask two major questions: \u003cem\u003e(i)\u003c/em\u003e What is the relationship between drought and agricultural prices in Mexico and how does it change between different clusters? \u003cem\u003e(ii)\u003c/em\u003e Is there any difference between the short- and long-term effects of drought on prices?\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn order to quantify drought and agricultural prices, we employ the Standard Precipitation Index (SPI) and the Agricultural Consumer Price Index (Agro CPI), respectively. We methodologically apply the Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) clustering algorithm developed by Campello et al. (2013). As also further discussed in the following sections, there are several reasons for choosing this approach. First, this clustering algorithm performs better when dealing with large datasets. As we also work with a large dataset, the HDBSCAN clustering algorithm fully suits our needs. Second, compared with other conventional clustering algorithms, such as K-means or DBSCAN, it requires fewer user-defined parameters, resulting in more significant and efficient clustering from the input data. Third, annual rain in Mexico varies greatly between regions. The south of the country receives almost half of the annual rain, while the north and center of the country, where most economic activity takes place, receive between 5 and 10 percent of the annual rain. This heterogeneity underscores the need for a clustering analysis, such as HDBSCAN. Fourth, clustering analysis is one of the widely used approaches in the literature currently to classify meteorological data and is highly effective in addressing agricultural challenges.\u003c/p\u003e\n\u003cp\u003eWe believe our timely study is important and contributes to the existing literature in three respects. First, as discussed previously, drought is a very important topic in Mexico\u0026mdash;it has been for a long time and is very likely to continue to be so in the future. Therefore, many scholars paid considerable attention to drought, examined the topic from several perspectives, and produced a rich literature that contains important information on\u0026nbsp;drought in Mexico and its effects on the country (Boyd and Ibarraran, 2009; Arceo-Gomez et al., 2022; Dobler-Morales and Bocco, 2021; Murray-Tortarolo and Jaramillo, 2019; Santacruz-De Leon et al. 2022; Gonzalez et al. 2023a; Gonzalez et al., 2023b). However, to the best of our knowledge, there is no study to investigate the drought-agricultural prices nexus in Mexico with a large dataset by using a data mining technique, especially clustering analysis. This study, for the first time, fills this gap in the existing literature. Second, we use the 24-month and one-month SPI values in this research and compare the results. As further discussed in the following sections, the one-month SPI values are useful for identifying brief variations in precipitation patterns and understanding the immediate effects of precipitation in drought conditions. However, the 24-month SPI provides insightful data regarding anomalous precipitation patterns over a longer period. Incorporating the 24-month SPI into our analysis not only improves our understanding of drought monitoring and management but also gives important information about the effect of extended droughts or periods of excessive precipitation on agricultural prices.\u003ca href=\"#_ftn1\" name=\"_ftnref1\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e1\u003c/sup\u003e Third, our study provides several useful insights for policymakers by shedding light on a trending topic from the past to the present in Mexico. This is important because food consumption depends on economic constraints. Our study exploring the interplay between climate variables and food prices can provide critical insights to regulators, policy, and decision-makers to better promote and coordinate initiatives and highlight the importance of long-term drought considerations in agricultural planning and market forecasting. The understanding of weather-related risks as an underlying cause of food prices, especially in developing countries, such as Mexico, due to their higher share in CPI baskets, may also facilitate the design of policies seeking to reduce inflation (Gonzalez et al., 2023a).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe remainder of this paper is organized as follows. In the second section, we review the literature with a specific focus on the socioeconomic impacts of drought in Mexico and the challenges of food security. Next, we introduce the method and dataset used in the study. We then discuss the main outcomes of our data analysis. Finally, we conclude the paper with policy implications and future research directions.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cp\u003eDrought is a widespread phenomenon, and its detrimental impacts are evident in many segments of society, including the economy, environment, agriculture, or health (Gautier et al., 2016; Vicento-Serrano et al., 2020; Lester et al., 2022; Sena et al., 2014; Ding et al., 2011; Edwards et al. 2018; Barış-T\u0026uuml;zemen and Lyhagen, 2024). Therefore, it has received considerable attention, especially in the academic literature, from many different perspectives. This increased interest is further fueled by the large economic and social costs connected with them that have recently been observed on all continents, including Europe, Asia, Africa, and South America (Mishra and Singh, 2010; Edwards et al., 2018).\u003c/p\u003e\n\u003cp\u003eThe economic impacts of droughts on agriculture are multifaceted and there is a growing body of literature in this area (Diersen et al., 2002; Edwards et al. 2009; Horridge et al., 2005). \u0026nbsp;For example, using the panel fixed effect model, Kuwayama et al. (2018) empirically investigated how the drought would affect crop yield and farm income in the US between 2001 and 2013. The empirical analysis outcomes show that there is a statistically significant negative relationship between crop yield and drought, but not when it comes to farm income. A recent study by Barış-T\u0026uuml;zemen and Lyhagen (2024) also finds that there is a statistically significant positive correlation between crop production and rainfall for eight South European countries on the Mediterranean coast for the period of 1996\u0026ndash;2016, but this link turns out to be negative if rainfall is excessive. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDrought has similar negative effects in Mexico as it does elsewhere in the globe. For example, as demonstrated by Boyd and Ibarraran (2009), due to Mexico\u0026apos;s high susceptibility to the effects of drought, the country\u0026apos;s economy is affected in several ways, including decreased economic output, a worsening trade balance, a rise in poverty, or a slowdown in economic development. Climate change is expected to make these droughts more frequent in Mexico in the years to come, which may further increase the negative impact of droughts, particularly on crops in the agriculture sector, which typically takes place in rural areas and is responsible for a significant share of economic activity (Dobler-Morales and Bocco, 2021; Boyd and Ibarraran, 2009).\u003c/p\u003e\n\u003cp\u003eThe drought-agriculture nexus for Mexico is also extensively studied in the existing literature. For example, Salas-Martinez et al. (2021) examine the relationship between the SPI and the yield of cattle and corn grain in the central zone of the state of Veracruz for the years 1980\u0026ndash;2018. The findings unequivocally demonstrate the detrimental effects of drought on cattle production and corn yield in Mexico. The paper also shows that the degree of vulnerability to drought varies significantly amongst regions in Mexico, necessitating targeted interventions and vulnerability assessments. Regional variation in terms of vulnerability and risk to drought is also confirmed by Neri and Maga\u0026ntilde;a (2016). Gonzalez et al. (2023a) investigate the effect of temperature shocks on vegetable prices in Mexico with monthly data at the city level. The results show that temperature has a detrimental effect on vegetable yields and this might be an important driver of higher prices.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBoyd and Ibarraran (2009) use a dynamic Computable General Equilibrium (CGE) model to examine the impact of prolonged drought on the Mexican economy and evaluate the potential efficacy of agriculture sector adaptation measures. The findings show that the rural sectors with output in grains, livestock, and forestry suffer the largest production losses. Arceo-Gomez et al. (2020) examine how the 2011 drought affected the well-being of rural households in Mexico, focusing on per capita incomes, poverty, and children\u0026apos;s attendance at school. Their findings indicate that droughts have a detrimental influence on the well-being of rural households. The impact of the 2011 drought is also studied by Murray-Tortarolo and Jaramillo (2019), but with an emphasis on livestock population. The findings indicate that the nation\u0026apos;s cattle and goat stocks declined by roughly 3% due to drought. In conclusion, the existing literature demonstrates that Mexico\u0026apos;s drought has had a significant economic impacts on the country, specifically through the agriculture sector, affecting rural households, livestock populations, economic output, and agricultural products.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData mining techniques, especially cluster analysis, have gained popularity in addressing agricultural challenges in the existing literature as they yield valuable insights across various domains. Chen et al. (2022), for instance, utilize K-means clustering to segment agricultural product clients, emphasizing its potential for targeted marketing. Similarly, Shestakov and Lovchikova (2023) employ a similar clustering approach in their regional agricultural analysis to facilitate the formulation of policies and the development of agribusinesses. Pascucci et al. (2018) highlight the superiority of multivariate functional principal components clustering in yield pattern recognition, promoting sustainable field management. Prakash et al. (2021) propose an efficient initial cluster center estimation method, enhancing clustering efficiency in agricultural databases. Simion et al. (2022) explore agglomerative clustering to improve farmers\u0026apos; competitiveness. In drought impact research, Liu et al. (2021) show the effectiveness of HDBSCAN clustering in meteorological data classification. In conclusion, the reviewed literature underscores the significance of clustering methodologies in addressing agricultural challenges and highlights the urgent need for comprehensive approaches to manage drought impacts and ensure food security. Integrated strategies informed by data insights are vital for sustainable agricultural development and mitigating the impacts of drought on food systems globally.\u003c/p\u003e"},{"header":"3. Methodology","content":"\u003cp\u003eHDBSCAN is a clustering algorithm that stands out for its superior performance compared to other methods, such as OPTICS or AUTO-HDS clustering, especially when dealing with large datasets containing a substantial number of objects (Campello et al., 2013). The algorithm begins by computing a core distance, denoted as (dcore), for each data point (Xp), representing the distance to its nearest neighbor among all data points or vertices (X = {x1,\u0026hellip;, xn}). An \u0026epsilon;-core object is then defined as an element of (X) that satisfies (\u0026epsilon; \u0026ge; dcore (Xp)). These core objects play a crucial role in identifying the density-based structure of the clusters.\u003c/p\u003e\n\u003cp\u003eOne of the key strengths of HDBSCAN lies in its ability to establish reachability between distances and vertices through a more efficient hierarchical simplification process. This is achieved by computing the reachability distance, denoted as (dreachability), defined as the maximum core distance among two data points (Xp) and (Xq), as well as the distance between them (dcore(Xp, Xq)). Mathematically, this can be expressed as dreachability = max{dcore(Xp), dcore(Xq), dcore(Xp, Xq)}, where (Xp) and (Xq) are the elements of (X). By incorporating reachability into the clustering process, HDBSCAN generates more robust and significant clusters from the input data.\u003c/p\u003e\n\u003cp\u003eUnlike many other clustering algorithms, HDBSCAN offers the advantage of requiring fewer parameters to be defined by the user. This reduction in parameterization simplifies the process of applying the algorithm to real-world datasets and makes it more accessible to users with varying levels of expertise. In traditional clustering algorithms like K-means or DBSCAN, users typically need to specify parameters such as the number of clusters (k) or the maximum distance between points (\u0026epsilon;). In HDBSCAN, the only parameter that users can set is the minimum cluster size, and even this can be set by the algorithm. HDBSCAN can provide an initial estimate of clusters without any user-specified parameters. This parameter determines the minimum number of points required for a cluster to be considered valid. By minimizing the number of user-defined parameters, HDBSCAN streamlines the clustering process and reduces the need for trial-and-error experimentation to find optimal parameter values. This simplicity and ease of use make HDBSCAN particularly well-suited for applications where a quick and efficient clustering solution is desired, without the burden of extensive parameter tuning. Andrzejuk (2018), for instance, identifies clusters in the agricultural emissions of the OECD countries by comparing the K-means algorithm with HDBSCAN. The paper emphasizes the advantages of HDBSCAN over traditional K-means, noting that the latter splits the data and assigns all of the information to a cluster, potentially leading to forced findings, while the former eliminates data that is unrelated to any cluster. The results further show that agriculture was comparable to Mexico and Australia in the initial analysis with K-means. However, after applying HDBSCAN, Mexico was omitted from the cluster and was instead categorized as an outlier with Japan, the Netherlands, New Zealand, Poland, the Republic of Korea, Spain, T\u0026uuml;rkiye, and the UK.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNotwithstanding its benefits, HDBSCAN possesses many significant limitations that warrant consideration. The method is proficient at identifying clusters of diverse densities; however, it may encounter difficulties with datasets featuring clusters of markedly varied densities, thereby overlooking crucial groups in sparse data regions. Moreover, while HDBSCAN minimizes the necessity for parameter adjustment relative to other clustering methods, its susceptibility to the minimum cluster size parameter can still influence outcomes, especially when addressing datasets with clusters of significantly diverse sizes. Additionally, the algorithm\u0026apos;s computational complexity may pose a challenge with extensive datasets, as it necessitates the computing of pairwise distances between points, potentially resulting in longer processing times relative to more straightforward clustering techniques such as K-means. That said, HDBSCAN demonstrates a notable advantage in processing information or data with noise, making it particularly effective for studies like ours, where variations in droughts and atypical rains in recent years introduce significant noise. Its ability to handle such noisy data establishes it as an appropriate method for the type of study we developed.\u003c/p\u003e"},{"header":"4. Data Analysis","content":"\u003cp\u003eWe use the one-month SPI data, first developed by McKee et al. (1993), to quantify drought in Mexico. The one-month SPI offers data on precipitation anomalies in relation to the long-term average for a particular month (Guttman, 1999).\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eA positive one-month SPI shows that the amount of precipitation for the given month is higher than the average amount of precipitation over a lengthy period. This indicates the presence of above-average levels of moisture.\u003c/li\u003e\n \u003cli\u003eA negative one-month SPI shows that the amount of precipitation during the given month is below the long-term average. This indicates arid conditions.\u003c/li\u003e\n \u003cli\u003eAn SPI score that is close to zero indicates that the precipitation for the specified month is quite similar to the long-term average, suggesting typical precipitation conditions.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eGreater positive or negative SPI values imply more substantial deviations from the average precipitation. In general, one-month SPI values are valuable for detecting brief deviations in precipitation patterns and comprehending the immediate consequences of precipitation in drought or wet situations. These numbers are included in a larger set of SPI values that are generated for various periods to evaluate the seriousness of drought conditions. On the other hand, the 24-month SPI offers valuable information about unusual precipitation patterns over a more extended timeframe. This is the way they operate (World Meteorological Organization (WMO), 2012):\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eLong-Term Assessment: The 24-month SPI provides a more comprehensive view by considering variations in precipitation over a two-year timeframe. This facilitates a more thorough understanding of precipitation patterns and their influence on drought or wet conditions.\u003c/li\u003e\n \u003cli\u003eCumulative Effects: In contrast to the one-month SPI that only considers short-term changes, the 24-month SPI measures the combined impact of precipitation anomalies over a longer period. This is especially beneficial for evaluating long-lasting droughts or prolonged periods of excessive rainfall.\u003c/li\u003e\n \u003cli\u003eRobustness: The 24-month SPI offers a more reliable evaluation of precipitation patterns by examining a longer period. This reduces the impact of temporary changes and provides a deeper understanding of long-term climate trends.\u003c/li\u003e\n \u003cli\u003eSeverity Assessment: The 24-month SPI indicates the severity of precipitation anomalies, similar to the one-month SPI. Positive values imply moisture levels that are higher than usual throughout the two years, while negative values suggest below-average precipitation, indicating dry conditions.\u003c/li\u003e\n \u003cli\u003eSupplementary study: While the one-month SPI is useful for identifying short-term variances, the 24-month SPI enhances this study by offering insights into longer-term climate variability. Collectively, they provide a thorough understanding of precipitation patterns and their ramifications for monitoring and administering drought.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eOverall, the 24-month SPI is a useful tool for evaluating extended rainfall patterns and providing vital information on prolonged periods of drought or excessive precipitation. It complements the analysis provided by shorter-term SPI.\u003ca href=\"#_ftn1\" name=\"_ftnref1\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eMexico\u0026apos;s National Consumer Price Index, \u0026mdash;or \u0026shy;\u0026shy;\u0026shy;\u0026Iacute;ndice Nacional de Precios al Consumidor (INPC)\u0026mdash;, which measures changes in the price level of a market basket of consumer goods and services purchased by households, is estimated and calculated by the Instituto Nacional de Estad\u0026iacute;stica y Geograf\u0026iacute;a (INEGI), which is the organization responsible for measuring and reporting on inflation in the country. The INPC has two components: core and non-core, which include more stable and volatile price components, respectively. This breakdown allows policymakers and analysts to distinguish between more stable price trends and those subject to short-term fluctuations, aiding in the formulation of economic policies and inflation expectations. The non-core agricultural prices (No Subyacente Agropecuarios) are of particular interest because they tend to be more volatile than the overall index.\u003c/p\u003e\n\u003cdiv id=\"ftn1\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003c/div\u003e"},{"header":"5. Results","content":"\u003cp\u003eBefore starting our HDBSCAN clustering analysis, we first converted our biweekly Mexican Agro CPI data to the monthly form and built a compatible dataset with the one- and 24-month SPIs for all 418 weather stations in the Mexican Conagua System. We then eliminated missing values (NaN) by using Multiple Imputation of Chained Equations (MICE). Next, we performed the SelectKBest feature selection method to retain only the 20 regions with the highest influence on Agro CPI between January 1988 and February 2024. We then conducted a cluster analysis with HDBSCAN.\u003c/p\u003e\n\u003ch2\u003e5.1. Short-Term Drought and Agro CPI\u003c/h2\u003e\n\u003cp\u003eFigure 1 illustrates the results of the cluster analysis. As can be seen, we identify three clusters: 1 (normal), 0, and -1 (anomalous). Interestingly, the 0 and -1 clusters only represent the period from December 2021 to the present date, presenting evidence of a change in short-term drought behavior.\u003c/p\u003e\n\u003cp\u003eIn Cluster 1 (the teal region), the average SPI is 0.11113, indicating wet conditions or excess precipitation. The average geometric growth of Agro CPI during this period is 22.69. In Cluster 0 (the yellow region), the average SPI changes to -0.4765, indicating dry conditions or drought. The geometric average growth of Agro CPI in Cluster 0 is 139.2. Sixteen out of the top 20 weather regions identified in the SelectKBest feature selection show a negative correlation with Agro CPI. The negative correlations between SPI and Agro CPI could indicate that as SPI decreases (indicating drier conditions), CPI tends to increase. This could suggest that drought conditions lead to lower agricultural yields, higher food prices, and thus higher Agro CPI. Figure 2 maps these regions in Cluster 0, and Figure 4 presents a heat map of the three clusters.\u003c/p\u003e\n\u003cp\u003eFinally, in Cluster -1, the average SPI is -0.3, indicating dry conditions or drought. The geometric average growth of Agro CPI is 138.6. Six regions present a negative correlation with Agro CPI, as can be seen in Figure 3. In HDBSCAN, clusters are formed by grouping points that lie close together in the high-density regions of the data space. It is also equally important to note that the determination of Cluster -1 occurred during the construction of the density-based clustering hierarchy. Points that are not sufficiently close to any core points or do not meet the minimum cluster size criterion are assigned to the noise cluster. This cluster represents the outliers or noise points in the dataset, which do not belong to any dense regions or meaningful clusters. Consequently, HDBSCAN\u0026apos;s inability to elucidate the dynamics of droughts during -1 periods underscores their enigmatic nature, extending beyond conventional explanations.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eJudging by the average SPI, drought conditions have been slightly less severe in Cluster -1, reflecting a slightly smaller growth of Agro CPI. However, Figure 4 shows that negative correlations prevail in clusters 0 and -1. The average correlation between Regional SPI and Agro CPI is 17% in Cluster 1, -9% in Cluster 0, and -8% in Cluster -1.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e5.2. Long-Term Drought and Agro CPI\u003c/h2\u003e\n\u003cp\u003eThis subsection presents an in-depth analysis of the relationship between the Agricultural CPI and the 24-month SPI in Mexico. Through a comparison of these two indicators, this study highlights the significant impact of prolonged drought conditions on agricultural pricing and the broader economic implications for consumers and producers.\u003c/p\u003e\n\u003cp\u003eFigure 5 illustrates the results of the cluster analysis with the 24-month SPI. \u0026nbsp;The clusters, shown in Figures 1 and 5, exhibit striking similarities. For example, Figure 5 shows an onset of anomalous behavior from September 2021 \u0026mdash;three months earlier than the 1-month SPI analysis\u0026mdash; to February 2024 (the end of our sample period). The key distinction between the comparison of the one-month and 24-month SPIs with Agro CPI lies in the considerably larger correlations observed in the 24-month analysis.\u003c/p\u003e\n\u003cp\u003eAgro CPI experiences substantial growth, particularly evident in Clusters -1 and 0, corresponding to the anomalous sample observed in the 1-month SPI comparison.\u003c/p\u003e\n\u003cp\u003eTable\u0026nbsp;1. Comparison of Clusters: Average 24-month-SPI and Geometric Average Growth of Agro CPI\u003c/p\u003e\n\u003cdiv align=\"\"\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"295\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003eClusters\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003eAverage SPI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45.7627%;\"\u003e\n \u003cp\u003eGeometric average growth of Agro CPI\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003e-1.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45.7627%;\"\u003e\n \u003cp\u003e137.17\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003e-0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45.7627%;\"\u003e\n \u003cp\u003e23.42\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 27.1186%;\"\u003e\n \u003cp\u003e-1.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45.7627%;\"\u003e\n \u003cp\u003e141.72\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eSource: Own estimation with data from INEGI and Conagua\u003c/p\u003e\n\u003cp\u003eTable 1 presents a comparison of clusters, highlighting the average 24-month SPI and geometric average growth of Agro CPI. As reported, the growth rates in Table 1 also coincide with those observed in the 1-month-SPI analysis. For example, Agro CPI shows substantial growth, especially in Clusters -1 and 0, aligning with the unusual samples reported in the 1-month SPI comparison. However, compared to the one-month SPI results, the results shown in Figure 6 indicate that the correlation between the 24-month SPIs and Agro CPI is much larger than those observed with the one-month SPI, indicating that a persistent lack of rainfall over longer periods is more influential in determining agricultural prices in Mexico in the last part of the sample. Therefore, the effects of drought on production are observed to be delayed, as extended periods of water scarcity negatively impact agricultural yields throughout successive growing seasons. Prolonged droughts can result in the loss of crucial water resources, worsening the impact on crop yields and livestock productivity, and gradually translating to increased food costs.\u003c/p\u003e\n\u003cp\u003eWhen we concentrate on Cluster -1 and compare the one-month and 24-month correlation between regions and the Agricultural CPI, we obtain the following inferences. In the 1-month analysis (Figure 1), the correlation for Cluster -1 ranges from -.50 to .32, whereas in the 24-month analysis (Figure 5), the correlation for Cluster -1 ranges from around -1.00 to -.40. Therefore, by focusing only on Cluster -1, we can clearly see that the correlations are substantially higher (more negative) in the 24-month analysis, compared to the one-month analysis. The maximum negative correlation in the one-month analysis for Cluster -1 is -.50; however, in the 24-month analysis, the maximum negative correlation for Cluster -1 is around -1.00, which is twice as extreme. This indicates that for the regions/areas in Cluster -1, which likely experienced more severe or prolonged drought conditions, the 24-month SPI has a much stronger negative correlation with Agricultural CPI increases than the one-month SPI. The longer 24-month period is better able to capture the full delayed and compounding effects of drought on agricultural pricing in these severely impacted areas. Thus, especially for Cluster -1, the finding clearly demonstrates the importance of considering longer drought timescales when analyzing pricing impacts.\u003c/p\u003e\n\u003cp\u003eA prolonged drought period as identified in this analysis could have several effects:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eDelayed impact of drought on production: Agricultural productivity is particularly vulnerable to extended drought conditions that span multiple growing seasons. Such long-term water deficits do not immediately affect crop yields, which may explain the weaker correlations with the one-month SPIs.\u003c/li\u003e\n \u003cli\u003eDepletion of water resources: Long-term droughts can lead to significant depletion of key water resources, including reservoirs, groundwater, and soil moisture. As these resources diminish, the consequences for crop yields and livestock productivity become more pronounced, directly influencing food prices.\u003c/li\u003e\n \u003cli\u003eCumulative drought effects: The adverse effects of drought on agricultural systems tend to build up over longer periods. This accumulation of damage to crops, pastures, livestock, supply chains, and infrastructure has far-reaching impacts on food prices.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThrough the comparative analysis of 24-month SPIs and Agro CPI, this section demonstrates the complex interplay between climate events and economic outcomes in Mexico\u0026apos;s agricultural sector. Understanding this relationship is critical for policymakers and stakeholders to create resilient agricultural practices and safeguard the economy against the consequences of climate variability.\u003c/p\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eIn this paper, we examine the relationship between the one-month and 24-month SPI, as documented in Mexico\u0026apos;s Conagua System, and the Agro CPI in Mexico. Our results show that the relationships between the Agro CPI and the 24-month SPI are significantly stronger than those between the Agro CPI and the one-month SPI. The overwhelming data indicates a possible long-term change in Mexico\u0026apos;s precipitation patterns, which might have significant effects on agricultural practices, food security, and overall CPI trends.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSome key takeaways from the paper are as follows:\u0026nbsp;\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eMultiyear drought conditions have a significant and long-lasting effect on agriculture that goes beyond the immediate growing seasons.\u003c/li\u003e\n \u003cli\u003eWe find significant increases in the Agro CPI in persistently dry areas, especially in Clusters -1 and 0. This suggests a clear link between longer droughts and higher food prices.\u003c/li\u003e\n \u003cli\u003eOur findings underscore the reality that the effects of drought on agriculture are not immediate but have a delayed onset, escalating over successive periods of water scarcity.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eOur paper suggests the following implications for policy and practice:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eThe evident correlation between extended drought and Agro CPI underscores the need for proactive strategies. Agricultural policies must incorporate climate forecasts, mitigate risks with resilient farming techniques, and effectively manage water resources.\u003c/li\u003e\n \u003cli\u003eImplementing agricultural practices tailored to withstand climatic fluctuations can reduce the adverse economic outcomes associated with drought. This may include investing in drought-resistant crop varieties and advanced irrigation systems.\u003c/li\u003e\n \u003cli\u003eFarmers, policymakers, and stakeholders are encouraged to utilize long-term climate predictions to inform planting decisions and resource allocation, thereby enhancing agricultural productivity and economic resilience.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eFurther research into the potential of innovative agricultural technologies to buffer the impact of long-term climatic changes on crop yields and Agro CPI can enhance our findings. Furthermore, future research may assess the feasibility and effectiveness of policy interventions aimed at supporting farmers during extended periods of drought, including financial instruments like crop insurance and water rights trading. Finally, these conclusions serve as a clarion call for the integration of climate adaptability into the core of agricultural planning and policy-making. It is imperative that we harness our collective knowledge and resources to safeguard food security and economic vitality in the face of mounting climate variability.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e: \u003c/p\u003e\n\u003cp\u003eWe thank our colleagues for their invaluable input and feedback during the research and writing process. Special thanks to the National Water Commission (CONAGUA) and the Instituto Nacional de Estad\u0026iacute;stica y Geograf\u0026iacute;a (INEGI) for providing essential data. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest\u003c/strong\u003e: \u003c/p\u003e\n\u003cp\u003eThe authors declare no conflicts of interest. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Declarations\u003c/strong\u003e:\u003c/p\u003e\n\u003cp\u003eThis research was supported by the National Council of Humanities, Sciences, and Technologies (Conahcyt).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are publicly available from the following sources:\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n\u003cli\u003eStandard Precipitation Index (SPI) data for Mexico were obtained from the National Meteorological Service (SMN) of the National Water Commission (CONAGUA) through their drought monitoring system, accessible at: https://smn.conagua.gob.mx/es/climatologia/monitor-de-sequia/spi\u003c/li\u003e\n\u003cli\u003eAgricultural Consumer Price Index (Agro CPI) data were sourced from the National Institute of Statistics and Geography (INEGI) through their economic indicators database, accessible at: https://inegi.org.mx/app/indicadores/?tm=0\u0026amp;t=10000215#D1000021\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe processed datasets and analysis code used in this study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAndrzejuk, A. (2018). Classification of Agricultural Emissions Among OECD Countries with Unsupervised Techniques. Zeszyty Naukowe SGGW W Warszawie - Problemy Rolnictwa Światowego, 18(4), 80\u0026ndash;91. https://doi.org/10.22630/PRS.2018.18.4.99\u003c/li\u003e\n \u003cli\u003eArceo-G\u0026oacute;mez, E. O., Hern\u0026aacute;ndez-Cort\u0026eacute;s, D., \u0026amp; L\u0026oacute;pez-Feldman, A. (2020). Droughts and rural households\u0026apos; wellbeing: evidence from Mexico. Climatic Change, 10.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eBarış-T\u0026uuml;zemen, \u0026Ouml;. \u0026amp; Lyhagen, J. (2024). Revisiting the role of climate change on crop production: Evidence from Mediterranean countries. Environment, Development and Sustainability. https://doi.org/10.1007/s10668-024-04991-x\u003c/li\u003e\n \u003cli\u003eBeltran-Silva, F. (2023). Fighting against hunger: A country-wide intervention and its impact on birth outcomes. World Development 165, 106202.\u003c/li\u003e\n \u003cli\u003eBirthal, P. S., Roy, D., Negi, D. S., Khan, T., \u0026amp; Aggarwal, P. K. (2015). Farm-level impacts of climate change on agriculture in India. Journal of Agricultural Economics, 66(1), 1-26.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eBoyd, R., \u0026amp; Ibarrar\u0026aacute;n, M. (2009). Extreme climate events and adaptation: an exploratory analysis of drought in Mexico. Environment and Development Economics, 74.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eBouaziz, M., Medhioub, E. \u0026amp; Csaplovisc, E. (2021). A machine learning model for drought tracking and forecasting using remote precipitation data and a standardized precipitation index from arid regions. Journal of Arid Environments, 189, 104478.\u003c/li\u003e\n \u003cli\u003eChen, T. C., Alazzawi, F. J. I., Mavaluru, D., Mahmudiono, T., Enina, Y., Chupradit, S., Ahmed, A. A. A., Syed, M. H., Ismael, A. M., \u0026amp; Miethlich, B. (2022). Application of data mining methods in grouping agricultural product customers. Mathematical Problems in Engineering, 2022, Article 9876543.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eDiersen, M.A., Taylor, G. \u0026amp; May, A. (2002). Direct and indirect effects of drought on South Dakota\u0026rsquo;s economy. Economics Commentator, 432, 26.\u003c/li\u003e\n \u003cli\u003eDijk, A. I., Renzullo, L. J., \u0026amp; McVicar, T. R. (2013). Global analysis of seasonal streamflow predictability using an ensemble prediction system and observations from 6192 small catchments worldwide. Water Resources Research, 49(3), 1276-1291.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eDing, Y., Hayes, M.J. \u0026amp; Widhalm, M. (2011). Measuring economic impacts of drought: a review and discussion. Disaster Prevention and Management, 20(4), 434-446. https://doi.org/10.1108/09653561111161752\u003c/li\u003e\n \u003cli\u003eDobler-Morales, C. \u0026amp; Bocco, G. (2021). Social and environmental dimensions of drought in Mexico: An integrative review. International Journal of Disaster Risk Reduction, 55, 10267.\u003c/li\u003e\n \u003cli\u003eEdwards, B., Gray, M. \u0026amp; Hunter, B. (2009). A sunburnt country: the economic and financial impact of drought on rural and regional families in Australia in an era of climate change. Australian Journal of Labour Economics, 12, 109\u0026ndash;31.\u003c/li\u003e\n \u003cli\u003eEdwards, B., Gray, M. \u0026amp; Hunter, B. (2018). The social and economic impacts of drought. Australian Journal of Social Issues, 54(1), 22-31.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFAO (1996). Declaration on world food security and World Food Summit plan of action. World Food Summit 13-17 November 1996, Rome, Italy.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFAO, IFAD, UNICEF, WFP and WHO (2023). The State of Food Security and Nutrition in the World 2023. Urbanization, agrifood systems transformation and healthy diets across the rural\u0026ndash;urban continuum. Rome. https://doi.org/10.4060/cc3017en\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFAO (2024). World Food Situation. https://www.fao.org/worldfoodsituation/foodpricesindex/en/\u003c/li\u003e\n \u003cli\u003eFederman, D. K., Cortes, F. I. A. \u0026amp; Perez, M. L. (2014). Constructing a framework or National Drought Policy: The way forward in Mexico. Weather and Climate Extremes, 3, 90-94.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eGuttman, N. B. (1999). Accepting the standardized precipitation index: A calculation algorithm. Journal of the American Water Resources Association, 35(2), 311-322.\u003c/li\u003e\n \u003cli\u003eHansen, J., List, G, Downs, S., Carr, E., Diro, R., Baethgen, W., Kruczkiewicz, W., Braun, M., Furlow, J., Wals, K. \u0026amp; Magima, N. (2022). Impact pathways from climate services to SDG2 (\u0026ldquo;zero hunger\u0026rdquo;): A synthesis of evidence. Climate Risk Management, 35.\u003c/li\u003e\n \u003cli\u003eHijioka, Y., Ikeda, T., Oyoshi, K., \u0026amp; Hirota, T. (2018). Climate change impacts on crop yields and water use in Australia. Climate Dynamics, 51(5-6), 2289-2303.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eHorridge, M., Madden, J. \u0026amp; Wittwer, G. (2005). The impacts of the 2002\u0026ndash;03 drought on Australia. Journal of Policy Modeling, 27(3), 285\u0026ndash;308.\u003c/li\u003e\n \u003cli\u003eGautier, D., Denis, D. \u0026amp; Locatelli, B. (2016). Impacts of drought and responses of rural populations in West Africa: a systematic review. WIREs Climate Change, 7(5), 666-681.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eGregory, C. A. \u0026amp; Coleman-Jensen, A. (2013). Do high food prices increase food insecurity in the United States?, Applied Economic Perspectives and Policy, 35(4), 679-707.\u003c/li\u003e\n \u003cli\u003eKuwayama, Y., Ikeda, T., \u0026amp; Hijioka, Y. (2018). Drought effects assessment on U.S. agriculture: Empirical evidence of panel data models with county-level data. PLOS ONE, 13(5), e0197650.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLester, L., Flatau, P. \u0026amp; Kyron, M. (2022). Understanding the social impacts of drught. Perth: The University of Western Australia.\u003c/li\u003e\n \u003cli\u003eLiu, J., Sun, S., \u0026amp; Chen, C. (2021). Big data Analysis of Regional Meteorological Observation Based: on Hierarchical Density Clustering Algorithm HDBSCAN. In 2021 2nd International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) (pp. 111-116). IEEE.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLiverman, D. M. (2016). Vulnerability and Adaptation to Drought in Mexico. National Resources Journal, 39, 99-115.\u003c/li\u003e\n \u003cli\u003eLubova, T. N., Salimova, G., Nigmatullina, G., Islamgulov, D., \u0026amp; Sharafutdinov, A. (2023). Modeling the development of agricultural production by cluster analysis. E3S Web of Conferences, 403, 01044.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMadadgar, S., AghaKouchak, A., \u0026amp; Farahmand, A. (2017). Multivariate drought characteristics across climatic zones. Scientific Reports, 7(1), 1785.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMcKee, T. B., Doesken, N. J. \u0026amp; Kleist, J. (1993). The relationship of drought frequency and duration of time scales. Eighth Conference on Applied Climatology, American Meteorological Society, Jan17-23, 1993, Anaheim CA, 179-186.\u003c/li\u003e\n \u003cli\u003eMishra, A. K., Jha, M. K., \u0026amp; Hansen, J. W. (2015). Remote sensing and hydrological models for drought assessment in northeast India. Journal of Hydrology, 524, 354-365.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMishra, A. K. \u0026amp; Singh, V. P. (2010). A review of drought concepts. Journal of Hydrology, 391(1-2), 202-216.\u003c/li\u003e\n \u003cli\u003eMurray-Tortarolo, G. N., \u0026amp; Jaramillo, V. J. (2019). The impact of extreme weather events on livestock populations: the case of the 2011 drought in Mexico. Climatic Change, 24.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eNASA (2021). Widespread Drought in Mexico. Retrieved from https://earthobservatory.nasa.gov/images/148270/widespread-drought-in-mexico.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eNASA (2024). Drought Parches Mexico. Retrieved from https://earthobservatory.nasa.gov/images/152908/drought-parches-mexico.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eNeri, C., \u0026amp; Maga\u0026ntilde;a, V. (2016). Estimation of Vulnerability and Risk to Meteorological Drought in Mexico. Weather, Climate, and Society, 45.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eNIDIS (2024). Retrieved from https://www.drought.gov/sectors/agriculture#:~:text=The%20primary%20direct%20economic%20impact,through%20government%20disaster%20assistance%20programs.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePascucci, S., Carfora, M., Palombo, A., Pignatti, S., Casa, R., Pepe, M., \u0026amp; Castaldi, F. (2018). A comparison between standard and functional clustering methodologies: Application to agricultural fields for yield pattern assessment. Remote Sensing, 10(1), 79.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePrakash, K. L. N. C., Narayana, G. S., Ansari, M. D., \u0026amp; Gunjan, V. K. (2021). Instantaneous approach for evaluating the initial centers in the agricultural databases using K-means clustering algorithm. Journal of Mobile Multimedia, 17(3), 297-320.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePutsenteilo, P., \u0026amp; Kostetskyi, Y. (2020). Using cluster analysis methods to forecast the investment development of agricultural sector enterprises. Innovative Economy, 16(83), 8-18.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSalas-Mart\u0026iacute;nez, F., Vald\u0026eacute;s-Rodr\u0026iacute;guez, O. A., Palacios-Wassenaar, O., \u0026amp; M\u0026aacute;rquez-Grajales, A. (2021). Analysis of the Evolution of Drought through SPI and Its Relationship with the Agricultural Sector in the Central Zone of the State of Veracruz, Mexico. Agronomy, 5.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSalazar, C., Acuna-Duarte, A. \u0026amp; Gil, J. M. (2023). Drought shocks and price adjustments in local food markets in Chile: Do product quality and marketing channel matter?. Agricultural Economics, 54(3), 349-363.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSantacruz-De Le\u0026oacute;n, G., Mor\u0026aacute;n-Ram\u0026iacute;rez, J., \u0026amp; Ramos-Leal, J. A. (2022). Impact of Drought and Groundwater Quality on Agriculture in a Semi-Arid Zone of Mexico. Agriculture, 3.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSena, A., Barcellos, C., Freitas, C. \u0026amp; Corvalan, C. (2014). Managing the Health Impacts of Drought in Brazil.\u0026nbsp;International Journal of Environmental Research and Public Health, 11, 10737-10751. https://doi.org/10.3390/ijerph111010737\u003c/li\u003e\n \u003cli\u003eShestakov, R., \u0026amp; Lovchikova, E. I. (2023). Clustering of regions using basic agricultural and economic criteria. Economy of Regions, 19(1), 320-334.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSimion, G., Căleanu, C. D., Bucos M. and B. Drăgulescu (2022). Grouping farmers using agglomerative clustering on data generated from statistics, 2022 International Symposium on Electronics and Telecommunications (ISETC), Timisoara, Romania, 2022, pp. 1-4, doi: 10.1109/ISETC56213.2022.10010248. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eStahle, D. W., Cook, E. R., Burnette, D. J., Villanueva, J., Cerano, J., Burns, J. N., Griffin, D., Cook, B. I., Acu\u0026ntilde;a, R., Torbenson, M. C. A., Szejner, P. \u0026amp; Howard, I. M. (2016). The Mexican Drought Atlas: Tree-ring reconstructions of the soil moisture balance during the late pre-Hispanic, colonial, and modern eras. Quaternary Science Reviews, 149, 34-60.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eTiwari, M., \u0026amp; Misra, B. (2011). Application of cluster analysis in agriculture\u0026ndash;A review article. International Journal of Computer Applications, 36(5), 40-48.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eWang, D., Wang, R., Wang, Y., \u0026amp; Zha, S. (2006). Research on clustering for agriculture information. In 2006 6th World Congress on Intelligent Control and Automation (pp. 7277-7281). IEEE.\u003c/li\u003e\n \u003cli\u003eWarr, P. (2014). Food Insecurity and its determinants. The Australian Journal of Agricultural and Resource Economics, 58(4), 519-537.\u003c/li\u003e\n \u003cli\u003eWMO (2012). Standardized Precipitation Index User Guide. (WMO-No. 1090), Geneva.\u003c/li\u003e\n \u003cli\u003eXing, J. (2011). Application of fuzzy dynamic cluster analysis in the division of agricultural economic system. Journal of Anhui Agricultural Sciences, 39(22), 13540-13542.\u003c/li\u003e\n \u003cli\u003eVicento-Serrano, S. M., Quiring, S. M., Pena-Gallardo, M., Yuan, S. \u0026amp; Dominguez-Castro, F. (2020). A review of environmental droughts: increased risk under global warming. Earth-Science Reviews, 201, 102953.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\n \u003cli\u003eSome papers forecast SPI to understand drought conditions in Mexico (Magallanes-Quintanar et al., 2024; Esquivel-Saenz et al., 2024).\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFor further information about the Mexico\u0026rsquo;s SPI, please visit the following website: https://smn.conagua.gob.mx/es/climatologia/monitor-de-sequia/spi\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Drought, Standard Precipitation Index, Agricultural Consumer Price Index, Cluster analysis, Climate variability, Policy implications","lastPublishedDoi":"10.21203/rs.3.rs-5827924/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5827924/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eIntroduction\u003c/strong\u003e:\u003c/p\u003e\n\u003cp\u003eThis study investigates the relationship between meteorological conditions and agricultural economics in Mexico from 1988 to 2024, with a focus on the impact of drought conditions measured by the Standard Precipitation Index (SPI) on agricultural prices, as indicated by the Agricultural Consumer Price Index (Agro CPI).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e:\u003c/p\u003e\n\u003cp\u003eUsing SPI values over one-month and 24-month periods, we conducted a cluster analysis to examine how precipitation variability affects agricultural prices. The Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) was employed for this analysis due to its efficiency with large datasets and minimal parameter requirements.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e:\u003c/p\u003e\n\u003cp\u003eThe analysis revealed a significant negative correlation between prolonged drought periods and Agro CPI, especially in drought-prone clusters. The study found that the effects of drought on agricultural prices are cumulative, with the full impact appearing over successive growing seasons. The 24-month SPI was determined to be a better predictor of the long-term effects of drought on Agro CPI than the one-month SPI.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e:\u003c/p\u003e\n\u003cp\u003eThese findings underscore the importance of considering long-term drought impacts in agricultural planning and market forecasting, highlighting critical insights for policymakers and stakeholders involved in agriculture and food security.\u003c/p\u003e","manuscriptTitle":"Drought and Agricultural Prices in Mexico","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-02-26 16:32:45","doi":"10.21203/rs.3.rs-5827924/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"63724c78-2502-49f7-8096-0eda7cd5e218","owner":[],"postedDate":"February 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-02-26T16:32:45+00:00","versionOfRecord":[],"versionCreatedAt":"2025-02-26 16:32:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5827924","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5827924","identity":"rs-5827924","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}